hash
stringlengths 32
32
| doc_id
stringlengths 7
13
| section
stringlengths 3
121
| content
stringlengths 0
2.2M
|
|---|---|---|---|
c283b321481640a37e74db6d6840f337
|
119 411-8
|
6.8.1 Fees
|
These policy requirements are not meant to imply any restrictions on charging for TSP's services.
|
c283b321481640a37e74db6d6840f337
|
119 411-8
|
6.8.2 Financial Responsibility
|
OVR-6.8.2-01: The requirements identified in ETSI EN 319 411-1 [3], clause 6.8.2 shall apply.
|
c283b321481640a37e74db6d6840f337
|
119 411-8
|
6.8.3 Confidentiality of Business Information
|
No policy requirement. ETSI ETSI TS 119 411-8 V1.1.1 (2025-10) 18
|
c283b321481640a37e74db6d6840f337
|
119 411-8
|
6.8.4 Privacy of Personal Information
|
OVR-6.8.4-01: The requirements identified in ETSI EN 319 411-1 [3], clause 6.8.4 shall apply.
|
c283b321481640a37e74db6d6840f337
|
119 411-8
|
6.8.5 Intellectual Property Rights
|
No policy requirement.
|
c283b321481640a37e74db6d6840f337
|
119 411-8
|
6.8.6 Representations and Warranties
|
OVR-6.8.6-01: The requirements specified in ETSI EN 319 411-1 [3], clause 6.8.6 shall apply.
|
c283b321481640a37e74db6d6840f337
|
119 411-8
|
6.8.7 Disclaimers of Warranties
|
See clause 6.8.6.
|
c283b321481640a37e74db6d6840f337
|
119 411-8
|
6.8.8 Limitations of Liability
|
Limitations on liability are covered in the terms and conditions as per clause 6.9.4.
|
c283b321481640a37e74db6d6840f337
|
119 411-8
|
6.8.9 Indemnities
|
No policy requirement.
|
c283b321481640a37e74db6d6840f337
|
119 411-8
|
6.8.10 Term and Termination
|
No policy requirement.
|
c283b321481640a37e74db6d6840f337
|
119 411-8
|
6.8.11 Individual notices and communications with participants
|
No policy requirement.
|
c283b321481640a37e74db6d6840f337
|
119 411-8
|
6.8.12 Amendments
|
No policy requirement.
|
c283b321481640a37e74db6d6840f337
|
119 411-8
|
6.8.13 Dispute Resolution Procedures
|
OVR-6.8.13-01: The requirements identified in ETSI EN 319 411-1 [3], clause 6.8.13 shall apply.
|
c283b321481640a37e74db6d6840f337
|
119 411-8
|
6.8.14 Governing Law
|
Not in the scope of the present document.
|
c283b321481640a37e74db6d6840f337
|
119 411-8
|
6.8.15 Compliance with Applicable Law
|
OVR-6.8.15-01: The requirements identified in ETSI EN 319 411-1 [3], clause 6.8.15 shall apply.
|
c283b321481640a37e74db6d6840f337
|
119 411-8
|
6.8.16 Miscellaneous Provisions
|
No policy requirement. ETSI ETSI TS 119 411-8 V1.1.1 (2025-10) 19
|
c283b321481640a37e74db6d6840f337
|
119 411-8
|
6.9 Other Provisions
| |
c283b321481640a37e74db6d6840f337
|
119 411-8
|
6.9.1 Organizational
|
OVR-6.9.1-01: The requirements identified in ETSI EN 319 411-1 [3], clause 6.9.1 shall apply.
|
c283b321481640a37e74db6d6840f337
|
119 411-8
|
6.9.2 Additional testing
|
OVR-6.9.2-01: The requirements identified in ETSI EN 319 411-1 [3], clause 6.9.2 shall apply.
|
c283b321481640a37e74db6d6840f337
|
119 411-8
|
6.9.3 Disabilities
|
OVR-6.9.3-01: The requirements identified in ETSI EN 319 411-1 [3], clause 6.9.3 shall apply.
|
c283b321481640a37e74db6d6840f337
|
119 411-8
|
6.9.4 Terms and conditions
|
OVR-6.9.4-01: The requirements specified in ETSI EN 319 411-1 [3], clause 6.9.4 shall apply. In addition: OVR-6.9.4-02 [QCP-n-eudiwrp] [QCP-l-eudiwrp]: The requirements specified in ETSI EN 319 411-2 [4], clause 6.9.4 shall apply. ETSI ETSI TS 119 411-8 V1.1.1 (2025-10) 20 Annex A (informative): Bibliography • Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation). ETSI ETSI TS 119 411-8 V1.1.1 (2025-10) 21 Annex B (informative): Change history Date Version Information about changes October 2025 V1.1.1 First publication ETSI ETSI TS 119 411-8 V1.1.1 (2025-10) 22 History Version Date Status V1.1.1 October 2025 Publication
|
d61f269b3bd6a8a4276ae1a17f53d29b
|
104 134
|
1 Scope
|
The present document concerns a methodology for including uncertainty and sensitivity aspects for avoided environmental impact calculations. The objective of the present document is to provide a standardized method to assess in a simplified manner the uncertainty of calculations for avoided environmental impact resulting from the introduction of Information and Communication Technology (ICT) Solutions. Moreover, the sensitivity of individual elements and the contribution to the total uncertainty is outlined. A method is defined based on existing standards, e.g. Recommendation ITU-T L.1480 [i.8] and recognized methods which allow for communication of the results to the public and consumers. The uncertainty and sensitivity calculation procedures are standardized for the method to be developed to make visible the relation between the degree of simplification and the ability to draw conclusions.
|
d61f269b3bd6a8a4276ae1a17f53d29b
|
104 134
|
2 References
| |
d61f269b3bd6a8a4276ae1a17f53d29b
|
104 134
|
2.1 Normative references
|
References are either specific (identified by date of publication and/or edition number or version number) or non-specific. For specific references, only the cited version applies. For non-specific references, the latest version of the referenced document (including any amendments) applies. Referenced documents which are not found to be publicly available in the expected location might be found in the ETSI docbox. NOTE: While any hyperlinks included in this clause were valid at the time of publication, ETSI cannot guarantee their long-term validity. The following referenced documents are necessary for the application of the present document. Not applicable.
|
d61f269b3bd6a8a4276ae1a17f53d29b
|
104 134
|
2.2 Informative references
|
References are either specific (identified by date of publication and/or edition number or version number) or non-specific. For specific references, only the cited version applies. For non-specific references, the latest version of the referenced document (including any amendments) applies. NOTE: While any hyperlinks included in this clause were valid at the time of publication, ETSI cannot guarantee their long-term validity. The following referenced documents may be useful in implementing an ETSI deliverable or add to the reader's understanding, but are not required for conformance to the present document. [i.1] L. Lakanen: "Developing handprints to enhance the environmental performance of other actors", 2023. [i.2] A.S.G. Andrae: "Method for Calculating the Avoided Impact of Specific Information and Communication Technology Services", International Journal of Environmental Engineering and Development, vol. 2, pp. 73-87, 2024. DOI: https://doi.org/10.37394/232033.2024.2.7. [i.3] AIOTI: "IoT and Edge Computing Carbon Footprint Measurement Methodology", 2024. [i.4] WBCSD: "Guidance on Avoided Emissions - Helping Business Drive Innovations and Scale Solutions Toward Net Zero", 2023. [i.5] J.C. Bieser, R. Hintemann, L.M. Hilty, S. Beucker: "A review of assessments of the greenhouse gas footprint and abatement potential of information and communication technology", Environmental Impact Assessment Review, 2023, vol. 99, p. 107033. DOI: 10.1016/j.eiar.2022.107033. ETSI ETSI TS 104 134 V1.1.1 (2025-09) 7 [i.6] A.S.G. Andrae: "Method for Uncertainty and Probability Estimation of Avoided Impacts from Information and Communication Technology Solutions", International Journal of Recent Engineering Science, vol. 11, no. 5, pp. 103-108, 2024. DOI: 10.14445/23497157/IJRES- V11I5P110. [i.7] D. Font Vivanco, J. Freire‐González, R. Galvin, T. Santarius, H.J. Walnum, T. Makov and S. Sala: "Rebound effect and sustainability science: A review", Journal of Industrial Ecology, vol. 26, no. 4, pp. 1543-1563, 2022. DOI: 10.1111/jiec.13295. [i.8] Recommendation ITU-T L.1480 (2022): "Enabling the Net Zero transition: Assessing how the use of information and communication technology solutions impact greenhouse gas emissions of other sectors". [i.9] W. Lu: "Study On The Advanced Technique of Environmental Assessment Based on Life Cycle Assessment Using Matrix Method", Ph.D. Thesis, The University of Tokyo, Tokyo, Japan, 2006. [i.10] A. Seidel, N. May, E. Guenther, F. Ellinger: "Scenario-based analysis of the carbon mitigation potential of 6G-enabled 3D videoconferencing in 2030", Telematics and Informatics, vol. 64, p. 101686, 2021. DOI: 10.1016/j.tele.2021.101686. [i.11] C.L. Thiel, N. Mehta, C.S. Sejo, L. Qureshi, M. Moyer, V. Valentino, J. Saleh: "Telemedicine and the environment: life cycle environmental emissions from in-person and virtual clinic visits", NPJ Digital Medicine, vol. 6, no. 1, 87, 2023. DOI: 10.1038/s41746-023-00818-7. [i.12] F. Bélorgey, J. Fournier, N.L. Omnes: "Application of International Telecommunication Union Recommendation L. 1480 on measuring the greenhouse gas emission effects to a use case for photovoltaic power generation equipment", Environmental Research: Energy, vol. 2, no. 1, 015004, 2025. DOI: 10.1088/2753-3751/ad9f64. [i.13] C. Mutel: "Brightway: An open source framework for Life Cycle Assessment", Journal of Open Source Software, vol. 2, no. 12, p. 236, 2017. DOI: 10.21105/joss.00236. [i.14] B. Steubing, D. de Koning, A. Haas, C.L. Mutel: "The Activity Browser — An open source LCA software building on top of the brightway framework", Software Impacts, vol. 3, p. 100012, 2020. DOI: 10.1016/j.simpa.2019.100012. [i.15] A.S.G. Andrae: "Proxy-Based Economic Factors for ICT Emission Avoidance in Cut-Off Frameworks", ResearchGate, 2025. DOI: 10.13140/RG.2.2.22177.31841.
|
d61f269b3bd6a8a4276ae1a17f53d29b
|
104 134
|
3 Definition of terms, symbols and abbreviations
| |
d61f269b3bd6a8a4276ae1a17f53d29b
|
104 134
|
3.1 Terms
|
For the purposes of the present document, the following terms apply: accuracy: closeness to the value of the perfect reference system NOTE: If the perfect reference system would have a score of 100 EI units and the score of the calculated system at hand would be 90 EI units, the accuracy of the LCA would be 90 %. avoided emission: emission reductions resulting from the use of a solution but occurring outside that solution's lifecycle or value chain NOTE: As defined in Recommendation ITU-T L.1480 [i.8]. direct rebound effect: rebound effect where increased efficiency, associated cost reduction and/or convenience of a product or service results in its increased use because it is cheaper or otherwise more convenient NOTE: As defined in Recommendation ITU-T L.1480 [i.8]. ETSI ETSI TS 104 134 V1.1.1 (2025-09) 8 economy-wide rebound effect: rebound effect where more efficiency drives economic productivity overall resulting in more economic growth and consumption at a macroeconomic level NOTE: As defined in Recommendation ITU-T L.1480 [i.8]. element: flow inputs or outputs to unit processes within the studied product system at hand EXAMPLE: Example of elements are CO2e emissions from "Car embodied" (output) and amount of "Use of cars" (input) used by "Use of vehicles" in Table A.1. a and b are elements. first order effect: direct environmental effect associated with the physical existence of an ICT solution, i.e. the raw materials acquisition, production, use and end-of-life treatment stages, and generic processes supporting those including the use of energy and transportation NOTE: As defined in Recommendation ITU-T L.1480 [i.8]. higher order effect: indirect effect (including but not limited to rebound effects) other than first and second order effects occurring through changes in consumption patterns, lifestyles and value systems NOTE: As defined in Recommendation ITU-T L.1480 [i.8]. net second order effect: resulting second order effect after accounting for emissions due to the first order effects of an ICT solution NOTE: As defined in Recommendation ITU-T L.1480 [i.8]. parameter: unit process within the studied product system at hand EXAMPLE: Examples of parameters are "Car embodied" (output) and "Use of cars" (input) used by "Use of vehicles" in Table A.1. rebound effect: increases in consumption due to environmental efficiency interventions that can occur through a price reduction or other mechanism including behavioural responses NOTE: As defined in Recommendation ITU-T L.1480 [i.8]. EXAMPLE: An efficient product being cheaper or in other ways more convenient and hence being consumed to a greater extent. second order effect: indirect impact created by the use and application of ICTs which includes changes of environmental load due to the use of ICTs that could be positive or negative NOTE: As defined in Recommendation ITU-T L.1480 [i.8].
|
d61f269b3bd6a8a4276ae1a17f53d29b
|
104 134
|
3.2 Symbols
|
For the purposes of the present document, the following symbols apply: Av Avoided environmental impacts SOE Second Order Effect FOE First Order Effect, ICT Scenario environmental impacts Rb Absolute environmental impacts for total rebound effect A Technology matrix p Process vector α Final demand vector β Final environmental load vector k Environmental load in β i Column in A or B j Row in A or B a Element in A b Element in B B Environmental load matrix Total CO2e (LCA) result ETSI ETSI TS 104 134 V1.1.1 (2025-09) 9 Summated CO2e scores based on Process-sum CO2e (LCA) data which are specific and granular for the system at hand Summated CO2e score based on EEIO CO2e (LCA) proxy data which cover the remaining processes Cut-off threshold RRb Relative total rebound effect
|
d61f269b3bd6a8a4276ae1a17f53d29b
|
104 134
|
3.3 Abbreviations
|
For the purposes of the present document, the following abbreviations apply: 2D Two Dimensional 3D Three Dimensional 5G Fifth-generation for wireless technology 6G Sixth-generation for wireless technology AI Artificial Intelligence CO2e Carbon Dioxide equivalents CUVP Contribution of individual element to total uncertainty EEIO Environmentally Extended Input-Output EI Environmental Impact IVP Input value of individual element LCA Life Cycle Assessment PC Personal Computer PV PhotoVoltaic RE Rebound Effect SVP Sensitivity of individual element TU Total Uncertainty of whole calculation result UVP Uncertainty of individual element
|
d61f269b3bd6a8a4276ae1a17f53d29b
|
104 134
|
4 Methodology
| |
d61f269b3bd6a8a4276ae1a17f53d29b
|
104 134
|
4.1 Framework
|
Equation 1 based on Equation 1 in [i.6] shows the main factors for the proposed method which shall be applied to any ICT Solution. = − + (1) where: = All avoided Environmental Impacts (EI) or avoided emissions from the use of the ICT Solution at hand per functional unit. This is the net second order effect of the ICT solution. = EI changes in the studied product system per functional unit for the Baseline Scenario created by the ICT Solution. This is the second order effect. = All ICT related EI from the studied product system per functional unit for the use of the ICT Solution Scenario. This is the first order effect. = Absolute EI for direct and economy-wide rebound effects from studied product system per functional unit for the ICT Solution Scenario. Equation 1 is in principle applicable to any standard for avoided impact calculations such as Recommendation ITU-T L.1480 [i.8]. ETSI ETSI TS 104 134 V1.1.1 (2025-09) 10
|
d61f269b3bd6a8a4276ae1a17f53d29b
|
104 134
|
4.2 Sensitivity of individual element
|
Equations 2 to 5 based on page 90 in [i.9], and Equation 6 based on pages 63 and 64 in [i.9], show how the rate sensitivity for activity and environmental load inventory flows shall be calculated. × = (2) = × (3) = × (4) = × × (5) = ∆ ∆ , ∆ ∆ (6) where: = Technology matrix. Activity flows arranged in a square matrix. = Environmental load matrix. = Process vector. = Final demand vector. = Final environmental load vector. = kth environmental load in the final environmental load vector. ∆ = Variation of the kth environmental load in the final environmental load vector due to a very small (tiny, miniscule) variation in . = Value of the element in the ith column in the jth row of A. ∆ = Very small (tiny, miniscule) variation of the value of element in the ith column in the jth row of A. = Value of the element in the ith column in the jth row of B. ∆ = Very small (tiny, miniscule) variation of the value of element in the ith column in the jth row of B. SVPij = sensitivity of individual element. NOTE 1: SVPij can be calculated manually or by specialized software programs such as those mentioned in Annex C (informative). To explain the factors of Equations 2 to 5 a fictive example (Table 1) is used: the production of one piece of a generic Product G. α (the final demand vector) in Table 1 is the amount of Product G necessary to fulfil the functional unit. The production of one piece of Product G may require 5 kWh of "Electricity 1" emitting 0,02 kg CO2e/kWh, 2 kWh of "Electricity 2" emitting 0,3 kg CO2e/kWh and 3 kWh of "Electricity 3" emitting 0,5 kg CO2e/kWh. Additionally Product G may need 10 kg Aluminium emitting 12 kg CO2e/kg and 0,05 kg IC emitting 1 300 kg CO2e/kg. ETSI ETSI TS 104 134 V1.1.1 (2025-09) 11 Table 1: Production of one piece of a generic Product G - a fictive example Electricity production 1 Unit Amount Output Electricity 1 kWh 1 Output CO2e kg 0,02 Electricity production 2 Output Electricity 2 kWh 1 Output CO2e kg 0,3 Electricity production 3 Output Electricity 3 kWh 1 Output CO2e kg 0,5 Aluminium production Output Aluminium kg 1 Output CO2e kg 12 IC production Output IC kg 1 Output CO2e kg 1 300 Product G production Output Product G pieces 1 Input Electricity 1 kWh 5 Input Electricity 2 kWh 2 Input Electricity 3 kWh 3 Input Aluminium kg 10 Input IC kg 0,05 Boundary α Product G piece 1 For the Product G example, a square A (in blue) is shown in Table 2. Table 2: Example of a square technology matrix A A Electricity production 1 Electricity production 2 Electricity production 3 Aluminium production IC production Product G production Electricity 1 1 kWh (output) 0 0 0 0 -20 kWh (input) Electricity 2 0 1 kWh (output) 0 0 0 -5 kWh (input) Electricity 3 0 0 1 kWh (output) 0 -3 kWh (input) Aluminium 1 kg (output) -10 kg (input) IC 1 kg (output) -0,05 kg (input) Product G 0 0 0 0 0 1 piece (output) α = 1 NOTE 2: The inputs to processes have to be designated with a minus (-) sign in the present methodology as otherwise the final environmental loadings would be expressed in negative numbers. This can be conveniently shown with numerical computation programs as shown in Annex D. For the Product G example, B is shown in Table 3. Table 3: Example of an environmental load matrix B B Electricity production 1 Electricity production 2 Electricity production 3 Aluminium production IC production Product G production CO2e 0,02 kg (output) 0,3 kg (output) 0,5 kg (output) 12 kg (output) 1 300 kg (output) 0 (output) NOTE 3: Occasionally B can be simplified to consider e.g. CO2e for each process as a whole instead of CO2, CH4, N2O, etc. or weighted EI values. This principle is applied in Annex A (informative) in the present document. Annex B (informative) on cut-off procedures for SOE and FOE (larger product systems) also uses CO2e for each process. ETSI ETSI TS 104 134 V1.1.1 (2025-09) 12 For the Product G example, A-1 (in yellow) and p are shown in Table 4. Table 4: Example of an inverse technology matrix A-1 and a process vector p Electricity production 1 Electricity production 2 Electricity production 3 Aluminium production IC production Product G production p = A-1 × α 1 0 0 0 0 5 5 0 1 0 0 0 2 2 0 0 1 0 0 3 3 0 0 0 1 0 10 10 0 0 0 0 1 0,05 0,05 0 0 0 0 0 1 1 NOTE 4: Each item in the p vector is the scaling factor corresponding to one unit process. For the Product G example, α is shown in Table 5. Table 5: Example of a final demand vector (α) Electricity production 1 Electricity production 2 Electricity production 3 Aluminium production IC production Product G production α Electricity 1 0 Electricity 2 0 Electricity 3 0 Aluminium 0 IC 0 Product G 1 NOTE 5: This α vector expresses the boundary condition for the economic flows at the system boundary. For the Product G example, β is shown in Table 6. Table 6: Example of a final environmental load vector β Electricity production 1 Electricity production 2 Electricity production 3 Aluminium production IC production Product G production Total Sum β = B × A-1 × α CO2e 0,02 × 5 × 1 = 0,1 kg 0,3 × 2 × 1 = 0,6 kg 0,5 × 3 × 1 = 1,5 kg 12 × 10 × 1 = 120 kg 1 300 × 0,05 × 1 = 65 kg 0 187,2 kg In summary manufacturing of one piece of Product G emits 187,2 kg CO2e. NOTE 6: In the present method the final results in Annex A (informative), e.g. 48 502 g CO2e in clause A.1, are examples of final environmental load vectors. Table 7 shows the SVPij for the generic Product G example. ETSI ETSI TS 104 134 V1.1.1 (2025-09) 13 Table 7: Sensitivity of individual elements in the Production of one piece of a generic Product G Electricity production 1 SVP Output Electricity 1 Output CO2e -0,000534 Electricity production 2 Output Electricity 2 Output CO2e -0,00320 Electricity production 3 Output Electricity 3 Output CO2e -0,00801 Aluminium production Output Aluminium Output CO2e -0,641 IC production Output IC Output CO2e -0,347 Product G production Output Product G Input Electricity 1 0,000534 Input Electricity 2 0,00320 Input Electricity 3 0,008013 Input Aluminium 0,641 Input IC 0,347
|
d61f269b3bd6a8a4276ae1a17f53d29b
|
104 134
|
4.3 Estimation of contribution to total uncertainty
|
Equation 7 shows how the share of the total uncertainty shall be calculated. = × × (7) where: CUVPij = contribution of an individual element to total uncertainty. NOTE 1: As shown in Annex A (informative), the CUVP is valid both for uncertainty contributions from environmental flows and from amount flows. NOTE 2: A is unitless and ∑ = 1. IVPij = input value of an individual element. UVPij = uncertainty of an individual element. TU = Total uncertainty of whole calculation result. Equation 7 helps prioritize the data for which the variability should be minimized in order to achieve robust conclusions. Equation 7 is generally applicable to any standard - such as Recommendation ITU-T L.1480 [i.8] - for avoided impact calculations.
|
d61f269b3bd6a8a4276ae1a17f53d29b
|
104 134
|
4.4 Estimation of relative rebound effect
|
Equation 8 shows how the relative total rebound effect shall be calculated. = ! (8) where: RRb = relative total rebound effect. Four examples are shown in Annex A (informative) on how the methodology in the present document can be applied. ETSI ETSI TS 104 134 V1.1.1 (2025-09) 14 Annex A (informative): Examples using the uncertainty and sensitivity methodology A.0 Introduction Here are included four examples which show how the proposed methodology is applied. A.1 Business meeting This example is based on [i.10]. Here the method is applied to a business meeting comparison between a physical business trip with travel and a virtual business trip with video. The comparison is done for a future case in 2030. The function is to enable a business meeting and the functional unit is "the enabling of a 10-hour business meeting (at a conference, seminar training, trade fair, exhibition) attended by a German in Germany in 2030". SOE in [i.10] is calculated as follows: 2 × 355,9 km × (64 % × 19,3 g CO2e/personkm for cars + 32 % × 20,5 g CO2e/personkm for trains + 4 % × 6,9 g CO2e/personkm for buses {Embodied of transport vehicles} + 2 × 355,9 km × (64 % × 115 g CO2e/personkm for cars + 32 % × 13 g CO2e/personkm for trains + 4 % × 29 g CO2e/personkm for buses {Use of transport vehicles} = 69 833,7 g CO2e/10-hour meeting. FOE in [i.10] is calculated as follows: 10 hours × 40,8 g CO2e/h {embodied of average of PC with display and laptop} + 10 hours × 189 g/kWh × (10 % 3D holographic × 0,0425 kW + 90 % 2D high-quality × 0,0375 kW) g CO2e {local cloud&on board computation, Local cloud + 6G AI + holographic data computing} + {Internet + local network} 10 hours × 189 g/kWh × (10 % 3D holographic × 0,1642 kW + 90 % 2D high-quality × 0,0169 kW) g CO2e = 539,6 g CO2e/10 hour meeting. The relative rebound effect is assumed to be 30 % [i.6]. NOTE: The relative rebound effect is calculated with Equation 8 as [38,53 × 539,6] / [69 833,7 - 539,6] = 0,3. Using the data with Equation 1: Av = SOE - (FOE+Rb) = 69 833,7 - (539,6 + 38,53 × 539,6) = 48 503 g CO2e. Table A.1 shows the data to be used for Equation 7 and Figure A.1. The uncertainty range values, UVP, have all been assumed. SVP can be derived with different software programs. Table A.1: CO2e intensities, uncertainties and sensitivities for proposed methodology applied to business meetings Parameter and combinations Unit used Proxy value (IVP) (g CO2e/unit), (mean value, µ) Uncertainty range for EI flow value and activity flow value (UVP), (2Ꝺ) Sensitivity factor (SVP) Contribution to total Uncertainty (CUVP) calculated by Equation 7 Car embodied (output) personkm 19,3 5 -0,181 1,27 % = ((48 502 / 19,3 × -0,181)2 × 52) / 20 1842 {share of the uncertainty of the CO2e emissions from Car embodied of the total uncertainty} Train embodied (output) personkm 20,5 5 -0,096 0,32 % Bus embodied (output) personkm 6,90 2,00 -0,00405 ≈ 0 % ETSI ETSI TS 104 134 V1.1.1 (2025-09) 15 Parameter and combinations Unit used Proxy value (IVP) (g CO2e/unit), (mean value, µ) Uncertainty range for EI flow value and activity flow value (UVP), (2Ꝺ) Sensitivity factor (SVP) Contribution to total Uncertainty (CUVP) calculated by Equation 7 Use of cars (output) personkm 115 20 -1,08 20,4 % = ((48 502 / 115 × -1,08)2 × 202) / 20 1842 {share of the uncertainty of the CO2e emissions from Car use of the total uncertainty} Use of trains (output) personkm 13 5 -0,061 0,32 % Use of buses (output) personkm 29 2 -0,017 ≈ 0 % Embodied of vehicles (output) personkm 1 Car embodied (input) personkm 0,64 0,05 0,181 0,12 % Train embodied (input) personkm 0,32 0,05 0,096 0,13 % Bus embodied (input) personkm 0,04 0,05 0,00405 0,01 % Use of vehicles (output) personkm 1 Use of cars (input) personkm 0,64 0,05 1,08 4,411 % = ((48 502 / 0,64 × 1,08)2 × 0,052) / 20 1842 {share of the uncertainty of the amount of "Use of cars", used by "Use of vehicles", of the total uncertainty} Use of trains (input) personkm 0,32 0,05 0,061 0,05 % Use of buses (input) personkm 0,04 0,05 0,017 0,26 % 10 hour meeting physical (output) piece 1 0 Embodied of vehicles (input) personkm 711,8 200 0,28 3,61 % Use of vehicles (input) personkm 711,8 200 1,15 61,15 % = ((48 502 / 711,8 × -1,15)2 × 2002) / 20 1842 {share of the uncertainty of the amount of "use of vehicles" used by the "10-hour meeting physical" of the total uncertainty} PC (display+laptop) embodied (output) hour 40,8 10 0,33 3,83 % German power 2030 (output) W 0,19 0,02 0,107 0,07 % 3D holographic local power (output) piece 1 - German power 2030 (input) W 42,5 4 0,0006 ≈ 0 % ETSI ETSI TS 104 134 V1.1.1 (2025-09) 16 Parameter and combinations Unit used Proxy value (IVP) (g CO2e/unit), (mean value, µ) Uncertainty range for EI flow value and activity flow value (UVP), (2Ꝺ) Sensitivity factor (SVP) Contribution to total Uncertainty (CUVP) calculated by Equation 7 2D high-quality local power (output) piece 1 German power 2030 (input) W 37,5 3 0,005 ≈ 0 % 3D holographic network power (output) piece 1 German power 2030 (input) W 164,5 4 0,00025 ≈ 0 % 2D high-quality network power (output) piece 1 German power 2030 (input) W 16,9 3 0,00023 ≈ 0 % local cloud&on board computation, Local cloud + 6G AI + holographic data computing (output) hours 1 PC (display+laptop) embodied (input) hours 1 3D holographic local power (input) piece 0,1 0,02 -0,0006 ≈ 0 % 2D high-quality local power (input) piece 0,9 0,18 -0,005 0,06 % Internet + local network (output) hours 1 3D holographic network power (input) piece 0,1 0,02 -0,00025 ≈ 0 % 2D high-quality network power (input) piece 0,9 0,18 -0,00023 ≈ 0 % Total meeting 10 hour 6G (output) piece 1 computation, Local cloud + 6G AI + holographic data computing (input) hours 10 -0,39 ≈ 0 % Internet + local network (input) hours 10 -0,048 ≈ 0 % Rebound effect (output) piece 1 Total meeting 10 hour 6G (input) pieces 38,53 7,72 0,428 4,24 % Sum of uncertainty contributions 100 % Avoided CO2e (Av) (output) piece 1 10 hour meeting physical (SOE) (input) piece 1 Total meeting 10 hour 6G (FOE) (output) piece 1 ETSI ETSI TS 104 134 V1.1.1 (2025-09) 17 Parameter and combinations Unit used Proxy value (IVP) (g CO2e/unit), (mean value, µ) Uncertainty range for EI flow value and activity flow value (UVP), (2Ꝺ) Sensitivity factor (SVP) Contribution to total Uncertainty (CUVP) calculated by Equation 7 Rebound effect (Rb) (output) piece 1 The interpretation of Table A.1 is that is most worthwhile to focus effort on reducing the uncertainty of the amount of personkm used for the physical meeting and also reduce the uncertainty for the emissions from the cars. Av result is 48 503 ± 20 184 g CO2e as shown in Figure A.1. TU in Equation 7 is here 20 184 g. Figure A.1: Resulting probability analysis of avoided emissions for changing ways to have a business meeting The conclusion that can be drawn from Figure A.1. is that the virtual meeting will help avoid emissions as the uncertainty is not too large. A.2 Health consultation This example is based on [i.2]. Here the method is applied to a health consultation comparison between physical and remote consultation. The function is "Providing health consultation of Computerized Tomography (CT) scans" and the functional unit is "A health consultation subsystem for 24 consultations per day involving analysis of CT scans to be suited for the needs of the purchasing customer". SOE in [i.2] is calculated as follows: 320 km × (4 cars / 250 000 km × 10 000) {Petrol car embodied} + 320 km × (5,58 dm3 / 100 km × 0,73 kg/dm3 × 0,45) {Petrol used} + 320 km × (5,58 dm3 / 100 km × 2,31) {Use of petrol car} + 8 hr × (243 / (4 × 8 760 hr) + 0,01 kW × 0,6) {PC embodied and use} + 8 hr × (400 / (4 × 8 760 hr) + 0,01 kW × 0,6) {Monitors embodied and use}} = 99 kg CO2e/24 consultations. FOE in [i.2] is calculated as follows: 3 pcs × 13 hr × (243 / (4 yr × 8 760 hr) + 0,01 kW × 0,6) {PC embodied and use} + 3 pcs × 13 hr × (400 / (4 yr × 8 760 hr) + 0,01 kW × 0,6) {Monitors embodied and use} + 5 GB/hr × 13 hr × 2 kWh / 44 GB × 0,6 {5G wireless network use} = 3 kg CO2e/24 consultations. The relative rebound effect is assumed to be 30 % [i.6]. NOTE: The relative rebound effect is calculated with Equation 8 as [9,6 × 3] / [99 - 3] = 0,3. 48503 0 10000 20000 30000 40000 50000 60000 70000 80000 90000 Avoided grams CO2e per 10 hour business meeting Enabling of a 10-hour business meeting 68687 48317 ETSI ETSI TS 104 134 V1.1.1 (2025-09) 18 Using the data with Equation 1: Av = SOE - (FOE+Rb) = 99 - (3 + 9,6 × 3) = 67,2 kg CO2e. Table A.2 shows data to be used for Equation 7 and Figure A.2. The uncertainty range values, UVP, have all been assumed. SVP factors can be derived with different software programs or manually. Table A.2: CO2e intensities, uncertainties and sensitivities for proposed methodology applied to health consultation Parameters and combinations Unit used Proxy value (IVP) (kg CO2e/unit), (mean value, µ) Uncertainty range for EI flow value and activity flow value (UVP), (2Ꝺ) Sensitivity factor (SVP) Contribution to total Uncertainty (CUVP) calculated by Equation 7 Electricity production (output) kWh 0,6 0,06 0,35 0,75 % Vehicle embodied (output) piece 10 000 1 900 -0,76 12,64 % Petrol production (output) kg 0,45 0,05 -0,087 0,07 % PC embodied (output) piece 243 24,3 0,041 0,01 % Monitor embodied piece 400 49 0,069 0,03 % Vehicle use (output) km 1 Vehicle embodied (input) piece s 1,6E-5 2,72E-6 0,76 10,12 % Petrol production (input) kg 0,04 0,004 0,08 0,05 % CO2e (output) kg 0,13 0,0013 -1,46 0,13 % PC use (output) hour 1 PC embodied (input) piece s 2,9E-5 2,9E-6 -0,041 0,01 % Electricity production (input) kWh 0,01 0,001 -0,036 0,01 % Monitor use (output) hour 1 Monitor embodied (input) piece s 2,9E-5 2,9E-6 -0,069 0,03 % Electricity production (input) kWh 0,01 0,001 -0,036 0,01 % 5G network use (output) GB 1 Electricity production (input) kWh 0,046 0,0046 -0,024 0,47 % 24 physical consultations (output) piece 1 Vehicle use (input) km 320 75 1,46 70,91 % PC use (input) hours 8 0,8 0,0015 0 % Monitor use (input) hours 8 0,8 0,002 0 % 24 remote consultations (output) piece 5G network use (input) GB 65 6,5 -0,28 0,07 % PC use (input) hours 39 3,9 -0,08 0,47 % Monitor use (input) hours 39 3,9 -0,107 0,04 % Rebound effect (output) piece 1 24 remote consultations (input) piece 9,6 1,92 0,378 4,3 % Sum of uncertainty contributions 100 % ETSI ETSI TS 104 134 V1.1.1 (2025-09) 19 Parameters and combinations Unit used Proxy value (IVP) (kg CO2e/unit), (mean value, µ) Uncertainty range for EI flow value and activity flow value (UVP), (2Ꝺ) Sensitivity factor (SVP) Contribution to total Uncertainty (CUVP) calculated by Equation 7 Avoided CO2e (Av) piece 1 24 physical consultations (SOE) (input) piece 1 24 remote consultations (FOE) (output) piece 1 Rebound effect (Rb) (output) piece 1 Av result is 67,22 ± 27,36 kg CO2e as shown in Figure A.2. Figure A.2: Resulting probability analysis for avoided emissions by changing health consultation technologies for CT scans Figure A.2 suggests that the conclusion that remote consultation will lead to avoided emissions is well-founded. NOTE: Example of cut-off procedure for clause A.2 is shown in clause B.1.1. A.3 Telemedicine This example is based on [i.11]. Here the method is applied to a health care clinic visit comparison between physical and mixed physical and remote consultation. The function is to enable hospital visits and the functional unit is "the enabling of 1 961 768 visits to the health care clinic in 2021". SOE in [i.11] is 43 160 132 kg CO2e/1 961 768 visits. FOE in [i.11] is 25 863 762 kg CO2e/1 961 768 visits. The relative rebound effect is assumed to be 30 % [i.6]. NOTE: The relative rebound effect is calculated with Equation 8 as: [0,2 × 25 863 762] / [43 160 132 - 25 863 762] = 0,3. Using the data with Equation 1: Av = SOE - (FOE + Rb) = 43 160 132 - (25 863 762 + 0,2 × 25 863 762) = 12 123 617 kg CO2e. 67.2 0 20 40 60 80 100 Avoided kg CO2e per 24 health consultations Health consultation of computerized tomography (CT) scans 94.6 39.9 ETSI ETSI TS 104 134 V1.1.1 (2025-09) 20 Table A.3 shows data to be used for Equation 7 and Figure A.3. The uncertainty range values, UVP, have all been assumed. SVP factors can be derived with different software programs or manually. Table A.3: CO2e intensities, uncertainties and sensitivities for proposed methodology applied to telemedicine Parameters and combinations Unit used Proxy value (IVP) (kg CO2e/unit), (mean value, µ) Uncertainty range for EI flow value and activity flow value (UVP), (2Ꝺ) Sensitivity factor (SVP) Contribution to total Uncertainty (CUVP) calculated by Equation 7 Video (output) minutes 0,0014 0,00028 0,0027 ≈ 0 % Phone (output) minutes 0,00072 0,000144 0,000119 ≈ 0 % Car travel (output) km 0,335 0,067 -0,701 6,60 % Air travel (output) pkm 0,129 0,0258 -0,301 1,22 % CV with Virtual Video (output) visit 1 Video (input) minutes 31,8 6,36 -0,0027 ≈ 0 % CV with Virtual Phone (output) visit 1 Phone (input) minutes 27,5 5,5 -0,000119 ≈ 0 % CV by Car in person, (output) visit 1 Car travel (input) km 39,35 7,9 2,13 61,03 % CV by Car in person, V, (output) visit 1 Car travel (input) km 33,5 6,7 -1,43 27,49 % CV by Air in person, (output) visit 1 Air travel (input) pkm 68,35 13,7 0,301 1,22 % Annual Person Visits (output) piece 1 CV by Car in person visits 1 961 768 CV by Air in person visits 1 961 768 Annual Virtual Visits (output) piece 1 CV with Virtual Video (input) visits 612 700 CV with Virtual Phone (input) visits 59 635 CV by Car in person, V, (input) visits 1 289 433 CV by Air in person, (input) visits 1 289 433 Rebound effect (output) piece 1 Annual Virtual Visits (input) piece 0,2 0,04 -0,43 2,44 % Sum of uncertainty contributions 100 % Avoided CO2e (Av) piece 1 Annual Person Visits (SOE) (input) piece 1 ETSI ETSI TS 104 134 V1.1.1 (2025-09) 21 Parameters and combinations Unit used Proxy value (IVP) (kg CO2e/unit), (mean value, µ) Uncertainty range for EI flow value and activity flow value (UVP), (2Ꝺ) Sensitivity factor (SVP) Contribution to total Uncertainty (CUVP) calculated by Equation 7 Annual Virtual Visits (FOE) (output) piece 1 Rebound effect (Rb) (output) piece 1 Av result is 12 123 617 ± 6 621 361 kg CO2e as shown in Figure A.3. Figure A.3: Resulting probability analysis for avoided emissions by changing visit type in hospitals Figure A.3 suggests that the conclusion that telemedicine will lead to avoided emissions is well-founded. A.4 Solar electricity This example is based on [i.12]. Here the method is applied to an electricity generation comparison between a PV solar plant installation and average grid mix. The function is to provide some of the electricity needs for one single-family detached house and to the grid. The functional unit is "generation of 39 968 kWh of electricity needed by one specific single-family detached house in Warzaw area in Poland and generation of 101 948 kWh of electricity needed elsewhere in Warzaw area in Poland during 25 years between 2022 and 2047". SOE in [i.12] is 63 365 kg CO2e/25 years. FOE in [i.12] is 10 215 kg CO2e/25 years. The higher-order effect in [i.12], including the rebound effect, is estimated to be 3 694 kg CO2e. NOTE: The relative rebound effect is calculated with Equation 8 as [3 694] / [63 365 - 10 215] = 0,0695. Using the data with Equation 1: Av = SOE - (FOE + Rb) = 63 365 - (10 215 + 3 694) = 49 456 kg CO2e. Table A.4 shows data to be used for Equation 7 and Figure A.4. The uncertainty range values, UVP, have all been assumed. ETSI ETSI TS 104 134 V1.1.1 (2025-09) 22 SVP factors can be derived with different software programs or manually. Table A.4: CO2e intensities, uncertainties and sensitivities for proposed methodology applied to solarization Parameters and combinations Unit used Proxy value (IVP) (kg CO2e/unit), (mean value, µ) Uncertainty range for EI flow value and activity flow value (UVP), (2Ꝺ) Sensitivity factor (SVP) Contribution to total Uncertainty (CUVP) calculated by Equation 7 Electricity mix which Solar will replace during 25 years (output) kWh 0,4465 0,12 -1,28 49,02 % PV panel (output) piece 456 91,2 0,16 0,81 % Inverter (output) piece 359,19 71,8 0,066 0,0013 % Other Solar embodied (output) piece 1 894 379 0,035 0,04 % Solar electricity (output) kWh 1 PV panel (input) piece 1,26E-4 2,52E-6 -0,16 0,81 % Inverter (input) piece 6,41E-6 1,28E-6 -0,066 0,0013 % Other Solar embodied (input) piece 6,41E-6 1,28E-6 -0,035 0,04 % Rebound effect (output) piece 3 694 739 0,075 0,17 % Avoided CO2e (Av) piece 1 Electricity mix which Solar will replace during 25 years (SOE) (input) kWh 141 916 28 383 1,28 49,02 % Solar electricity (FOE) (output) kWh 141 916 7 096 -0,21 0,08 % Rebound effect (Rb) (output) piece 1 0,05 -0,075 0,01 % Sum of uncertainty contributions 100 % Av result is 49 456 ± 18 100 kg CO2e as shown in Figure A.4. ETSI ETSI TS 104 134 V1.1.1 (2025-09) 23 Figure A.4: Resulting probability analysis for avoided emissions µby introducing solar electricity in detached houses Figure A.4 suggests that the conclusion that introducing solar electricity will lead to avoided emissions is well-founded. 49456 0 10000 20000 30000 40000 50000 60000 70000 80000 Avoided kg CO2e per 6.3 Wpeak installation during 25 years Avoided emissions by use of solar panels in Poland 2022-2047 67557 31356 ETSI ETSI TS 104 134 V1.1.1 (2025-09) 24 Annex B (informative): Method for knowing if enough data have been collected to meet cut-off threshold B.0 Introduction Generally, a complete life cycle product system with a perfect accuracy includes all connected processes and primary data for all. This will be challenging to achieve in most situations and therefore a smaller product system can be identified (for which more accurate and specific data should be used) for the technology matrix A mentioned in clause 4.2. In this annex, a method for knowing if enough data have been collected to meet the preset cut-off threshold, is outlined. It is based on pages 92-103 in [i.9]. The method should be applied separately to SOE, FOE and Rb as in the present document there is no method developed for aggregating one c for Equation 1. B.1 Method description B.1.0 Detailed description of the method The Process-Sum (PS) method is combined with the Environmentally Extended Input−Output (EEIO) method (Equation B.1). = + (B.1) The cut-off criterion equation is (Equation B.2). ≥1 − (B.2) where: = Total CO2e (LCA) result. = Summated CO2e scores based on Process-sum CO2e (LCA) data which are specific and granular for the system at hand. = Summated CO2e score based on EEIO CO2e (LCA) proxy data which cover the remaining processes. = cut-off threshold, e.g. 0,05 for 5 %. The method has the following steps: 1) Define the goal of the CO2e (LCA) analysis. 2) Compose a preliminary product system indiscriminately by including important processes from all life cycle stages. 3) Analyse the preliminary product system with the PS LCA method for and note which remaining (surplus) processes were not be modelled with PS. 4) Analyse these processes with the EEIO LCA method for obtaining . 5) Determine if ≥1 − . 6) If not, determine if there are some essential processes modelled with the EEIO method which need to be included and modelled with the PS method. 7) Analyse again the new preliminary product system with the PS LCA method for obtaining and note again which processes could not be modelled with PS. ETSI ETSI TS 104 134 V1.1.1 (2025-09) 25 8) Repeat steps 5 and 6. 9) When ≥1 − , i.e. ≤ , the cut-off threshold has been met. Next follows an example of avoided emissions of health consultation. B.1.1 Example application of the method B.1.1.0 Avoided emissions of health consultation The goal is to perform a CO2e analysis of the avoided impact associated with a health consultation comparison between physical and remote consultation shown in clause A.2. The scope is from cradle-to-use. The required value of is 0,05 for SOE, 0,05 for FOE and 0,5 for Rb. Next follows the cut-off method applied to SOE and FOE of clause A.2. B.1.1.1 Cut-off method applied to clause A.2 Table B.1 shows the CO2e score for the preliminary product system for SOE. The first iteration represents the first preliminary product system. Table B.1: Score for preliminary product system with the PS LCA method () Process name(s) (kg CO2e generated by PS method) Petrol car embodied 51,2 SUM of CO2e emission (kg) 51,2 Table B.2 shows the CO2e score for the remaining surplus processes. Table B.2: Scores for remaining processes analysed with the EEIO LCA method Process name (kg CO2e generated by EEIO method) Use of petrol car 2,25 kg CO2e/USD [i.15] × 20 USD = 45 kg Petrol production 0,772 kg CO2e/USD [i.15] × 20 USD = 15,44 kg Others 5 SUM of CO2e emission (kg) 65,44 Table B.3 shows how (Equation B.2) is determined gradually for SOE with the proposed method. Table B.3: Calculation of by iterative process for SOE in clause A.2 Iteration number Added processes kg CO2e + 1 Petrol car embodied 51,2 65,44 0,4389 which is < 1 - 0,05 2 Use of petrol car 92,44 20,44 0,8189 which < 1 - 0,05 3 Petrol production 98,31 5 0,9516 which > 1 - 0,05 The cut-off threshold has been met. ETSI ETSI TS 104 134 V1.1.1 (2025-09) 26 All added processes in column 2 of Table B.3 represent the final product system for which the PS method should be used. B.1.1.2 Cut-off method applied to FOE of clause A.2 Table B.4 shows the CO2e score for the preliminary product system for FOE. The first iteration represents the first preliminary product system. Table B.4: Score for preliminary product system for FOE with the PS LCA method () Process name(s) (kg CO2e generated by PS method) Use of 5G 1,77 SUM of CO2e emission (kg) 1,77 Table B.5 shows the CO2e score for the remaining surplus processes. Table B.5: Scores for remaining processes of FOE analysed with the EEIO LCA method Process name (kg CO2e generated by EEIO method) Production of Monitor 0,488 kg CO2e/USD [i.15] computers and electronics × 15 USD = 1,464 Production of PC 0,488 kg CO2e/USD [i.15] computers and electronics × 50 USD = 4,88 Use of PC 6 kg CO2e/USD [i.15] computers and electronics × 0,6 USD = 3,6 Use of Monitor 6 kg CO2e/USD [i.15] computers and electronics × 0,2 USD = 1,2 Others 0,1 SUM of CO2e emission (kg) 36,62 Table B.6 shows how (Equation B.2) is determined gradually for FOE with the proposed method. Table B.6: Calculation of by iterative process for FOE Iteration number Added processes kg CO2e + 1 Use of 5G 1,77 36,62 0,046 which is < 1 - 0,05 2 Production of Monitor 2,21 29,3 0,071 which is < 1 - 0,05 3 Production of PC 2,48 4,9 0,336 which is < 1 - 0,05 4 Use of PC 2,72 3,7 0,676 which is < 1 - 0,05 5 Use of Monitor 2,95 0,1 0,9673 which > 1 - 0,05 The cut-off threshold has been met. All added processes in column 2 of Table B.6 represent the final product system for which the PS method should be used. NOTE: The practitioner is advised to take care when applying cut-off to FOE or Rb in order to make sure that the avoided environmental impact result is realistic and representative and not only resulting from the use of cut-off. ETSI ETSI TS 104 134 V1.1.1 (2025-09) 27 B.1.1.3 Cut-off method applied to Rb for clause A.2 The way Rb is used in the present document, it is not tied to any measurable activity like SOE and FOE. Rb is therefore 100 % proxy and a required value of c cannot be applied. Therefore, there is no pathway to increase without redefining Rb so process-sum data and EEIO data can be separated. This is beyond the scope of the present document and it is acknowledged that Rb is 100 % proxy. ETSI ETSI TS 104 134 V1.1.1 (2025-09) 28 Annex C (informative): Examples of software programs for implementation This annex lists examples of software programs which can be used to implement the present document. Examples are openLCA found at https://www.openlca.org/openlca/openlca-features/, Brightway [i.13] at https://docs.brightway.dev/en/latest/index.html and the related Activity Browser [i.14]. Another is Chain management by Life Cycle Assessment, found at https://www.universiteitleiden.nl/en/research/research-output/science/cml-cmlca. ETSI ETSI TS 104 134 V1.1.1 (2025-09) 29 Annex D (informative): Example of code for implementation of clause 4.2 in the present document This code can be used in the program GNU Octave [https://octave.org/] to calculate the 187,2 kg CO2e for the example in clause 4.2: A=[1,0,0,0,0,-5;0,1,0,0,0,-2;0,0,1,0,0,-3;0,0,0,1,0,-10;0,0,0,0,1,-0.05;0,0,0,0,0,1] B=[0.02;0.3;0.5;12;1300;0] alfa=[0;0;0;0;0;1] p=inv(A)*alfa beta=transpose(p)*B A=[1,0,0,0,0,-5;0,1,0,0,0,-2;0,0,1,0,0,-3;0,0,0,1,0,-10;0,0,0,0,1,-0.05;0,0,0,0,0,1] A = 1.0000 0 0 0 0 -5.0000 0 1.0000 0 0 0 -2.0000 0 0 1.0000 0 0 -3.0000 0 0 0 1.0000 0 -10.0000 0 0 0 0 1.0000 -0.0500 0 0 0 0 0 1.0000 >> B=[0.02;0.3;0.5;12;1300;0] B = 2.0000e-02 3.0000e-01 5.0000e-01 1.2000e+01 1.3000e+03 0 >> alfa=[0;0;0;0;0;1] alfa = 0 0 0 0 0 1 >> p=inv(A)*alfa p = 5.0000e+00 2.0000e+00 3.0000e+00 1.0000e+01 5.0000e-02 1.0000e+00 >> beta=transpose(p)*B beta = 187.20 >> ETSI ETSI TS 104 134 V1.1.1 (2025-09) 30 Annex E (informative): Change history Date Version Information about changes June 2024 V0.0.1 Added some initial text and heading 12 November 2024 V0.0.2 Stable draft 22 November 2024 V0.0.3 Final draft for approval 3 January 2025 V0.0.4 Final draft for approval 17 January 2025 V0.0.5 Stable draft 31 January 2025 V0.0.6 Stable draft 14 February 2025 V0.0.7 Stable draft 16 April 2025 V0.0.8 Final draft for approval 16 April 2025 V0.0.9 Final draft for approval ETSI ETSI TS 104 134 V1.1.1 (2025-09) 31 History Version Date Status V1.1.1 September 2025 Publication
|
dd3b0bf937cd109481cadaa34e26502f
|
104 103
|
1 Scope
|
The present document identifies and describes the problem arising from pervasive encrypted traffic in electronic/digital communications networks. In addition, the present document states the requirements for allowing Encrypted Traffic Integration (ETI) across an abstracted network architecture. The present document identifies the impact of encrypted traffic on a number of stakeholders and how the stakeholders' objectives work together. The characterization of traffic as either user generated content, user generated signalling, network signalling, and metadata, and the relative impact on stakeholders is considered. The present document also addresses the role of compliance obligations on the development and deployment of encrypted traffic and how it impacts different stakeholders. The present document addresses the impact of pervasive encryption on stakeholders in order to assist future standards development activity in mitigating the negative impact on stakeholders whilst not adversely impacting the positive impacts of such a paradigm on stakeholders, including the regulatory and lawful dimensions. The present document is structured as follows: • Clause 4 outlines the role of encryption as it is being applied to networks from a mainly business perspective. • Clause 5 outlines and provides a model of the ETI problem. • Clause 6 presents the ETI model from a technical perspective. • Clause 7 summarizes the risk of pervasive encryption. • Clause 8 outlines the requirements for countering risks of pervasive encryption. • Annex A (normative) provides a summary of the impact of pervasive encryption on various formal compliance obligations. • Annex B (informative) gives an overview of the various common approaches to provide encryption in networks. • Annex C (informative) offers a number of examples of the impact of pervasive encryption. • Annex D (informative) addresses the application of the ETI requirements using the security controls approach. The present document includes requirements that implement the role of Zero Trust (ZT) and Zero Trust Architecture (ZTA) [1] in ETI, in order to provide an explicitly trusted communications environment across all enabled layers of the Open Systems Interconnection (OSI) model. In addition, the present document describes a ZTA security model, that enforces transparency and explicability of the role of security functions, particularly encryption.
|
dd3b0bf937cd109481cadaa34e26502f
|
104 103
|
2 References
| |
dd3b0bf937cd109481cadaa34e26502f
|
104 103
|
2.1 Normative references
|
References are either specific (identified by date of publication and/or edition number or version number) or non-specific. For specific references, only the cited version applies. For non-specific references, the latest version of the referenced document (including any amendments) applies. Referenced documents which are not found to be publicly available in the expected location might be found in the ETSI docbox. NOTE: While any hyperlinks included in this clause were valid at the time of publication, ETSI cannot guarantee their long-term validity. The following referenced documents are necessary for the application of the present document. [1] NIST Special Publication 800-207: "Zero Trust Architecture". [2] ETSI TS 104 102: "Cyber Security (CYBER); Encrypted Traffic Integration (ETI); ZT-Kipling methodology". ETSI ETSI TS 104 103 V1.1.1 (2025-09) 7 [3] ETSI TS 103 532: "CYBER; Attribute Based Encryption for Attribute Based Access Control". [4] Recommendation ITU-T X.800: "Security architecture for Open Systems Interconnection for CCITT applications". [5] ETSI TS 101 331: "Lawful Interception (LI); Requirements of Law Enforcement Agencies". [6] ETSI TS 102 656: "Lawful Interception (LI); Retained Data; Requirements of Law Enforcement Agencies for handling Retained Data". [7] ETSI TS 102 165-1: "Cyber Security (CYBER); Methods and protocols; Part 1: Method and pro forma for Threat, Vulnerability, Risk Analysis (TVRA)". [8] ETSI TS 102 165-2: "Telecommunications and Internet converged Services and Protocols for Advanced Networking (TISPAN); Methods and protocols; Part 2: Protocol Framework Definition; Security Counter Measures". . [9] ETSI TS 104 224: "Securing Artificial Intelligence (SAI); Explicability and transparency of AI processing". [10] ETSI TS 104 223: "Securing Artificial Intelligence (SAI); Baseline Cyber Security Requirements for AI Models and Systems".
|
dd3b0bf937cd109481cadaa34e26502f
|
104 103
|
2.2 Informative references
|
References are either specific (identified by date of publication and/or edition number or version number) or non-specific. For specific references, only the cited version applies. For non-specific references, the latest version of the referenced document (including any amendments) applies. NOTE: While any hyperlinks included in this clause were valid at the time of publication, ETSI cannot guarantee their long-term validity. The following referenced documents may be useful in implementing an ETSI deliverable or add to the reader's understanding, but are not required for conformance to the present document. [i.1] ISO/IEC 7498-1: "Information technology - Open Systems Interconnection - Basic Reference Model: The Basic Model". NOTE: The same text is available as Recommendation ITU-T X.200. [i.2] Directive 2014/53/EU of the European Parliament and of the Council of 16 April 2014 on the harmonisation of the laws of the Member States relating to the making available on the market of radio equipment and repealing Directive 1999/5/EC (Text with EEA relevance) (Radio Equipment Directive). [i.3] COM/2017/010 final: "Proposal for a Regulation of the European Parliament and of the Council concerning the respect for private life and the protection of personal data in electronic communications and repealing Directive 2002/58/EC (Regulation on Privacy and Electronic Communications)". [i.4] Regulation (EU) 1025/2012 of the European Parliament and of the Council of 25 October 2012 on European standardisation, amending Council Directives 89/686/EEC and 93/15/EEC and Directives 94/9/EC, 94/25/EC, 95/16/EC, 97/23/EC, 98/34/EC, 2004/22/EC, 2007/23/EC, 2009/23/EC and 2009/105/EC of the European Parliament and of the Council and repealing Council Decision 87/95/EEC and Decision No 1673/2006/EC of the European Parliament and of the Council Text with EEA relevance. [i.5] ETSI TR 103 456: "CYBER; Implementation of the Network and Information Security (NIS) Directive". [i.6] ENISA: "Gaps in NIS standardisation Recommendations for improving NIS in EU standardisation policy" V.1.0, November 2016. [i.7] ETSI TR 103 618: "CYBER; Quantum-Safe Identity-Based Encryption". ETSI ETSI TS 104 103 V1.1.1 (2025-09) 8 [i.8] ETSI TR 103 719: "Guide to Identity-Based Cryptography". [i.9] ETSI TS 103 458: "CYBER; Application of Attribute Based Encryption (ABE) for PII and personal data protection on IoT devices, WLAN, cloud and mobile services - High level requirements". [i.10] ETSI TR 103 369: "CYBER; Design requirements ecosystem". [i.11] ETSI TR 103 421: "CYBER; Network Gateway Cyber Defence". [i.12] IETF RFC 8404: "Effects of Pervasive Encryption on Operators". [i.13] U.S. Office of the Director of National Intelligence (ODNI): "Going Dark: Impact to intelligence and law enforcement and threat mitigation" (2017). [i.14] Tor project. [i.15] A. Young, M. Yung: "Cryptovirology: Extortion-Based Security Threats and Countermeasures". IEEE Symposium on Security & Privacy, May 6-8, 1996. pp. 129-141. IEEE Explore: Cryptovirology: extortion-based security threats and countermeasures. [i.16] ETSI TR 102 661: "Lawful Interception (LI); Security framework in Lawful Interception and Retained Data environment". [i.17] ETSI TR 103 936: "Cyber Security (CYBER); Implementing Design practices to mitigate consumer IoT-enabled coercive control". [i.18] ISO 7498-2: "Information processing systems -- Open Systems Interconnection -- Basic Reference Model -- Part 2: Security Architecture". [i.19] ETSI TR 103 309: "CYBER; Secure by Default - platform security technology". [i.20] ETSI TS 103 486: "CYBER; Identity Management and Discovery for IoT". [i.21] GSMA: "FS.37 GTP-U Security". [i.22] GSMA: "FS.40 5G Security Guide Version 3.0". [i.23] NIST Special Publication 800-53, Revision 5: "Security and Privacy Controls for Information Systems and Organizations". [i.24] ETSI TR 104 065: "Securing Artificial Intelligence (SAI); AI Act mapping and gap analysis to ETSI workplan". [i.25] EN 18031-1: "Common security requirements for radio equipment - Part 1: Internet connected radio equipment" (produced by CEN). [i.26] EN 18031-2: "Common security requirements for radio equipment - Part 2: radio equipment processing data, namely Internet connected radio equipment, childcare radio equipment, toys radio equipment and wearable radio equipment" (produced by CEN). [i.27] EN 18031-3: "Common security requirements for radio equipment - Part 3: Internet connected radio equipment processing virtual money or monetary value" (produced by CEN). [i.28] Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation). [i.29] Directive (EU) 2016/1148 of the European Parliament and of the Council of 6 July 2016 concerning measures for a high common level of security of network and information systems across the Union (NIS Directive). [i.30] Regulation (EU) 2024/2847 of the European Parliament and of the Council of 23 October 2024 on horizontal cybersecurity requirements for products with digital elements and amending Regulations (EU) No 168/2013 and (EU) 2019/1020 and Directive (EU) 2020/1828 (Cyber Resilience Act). ETSI ETSI TS 104 103 V1.1.1 (2025-09) 9 [i.31] Regulation (EU) 2024/1689 of the European Parliament and of the Council of 13 June 2024 laying down harmonised rules on artificial intelligence and amending Regulations (EC) No 300/2008, (EU) No 167/2013, (EU) No 168/2013, (EU) 2018/858, (EU) 2018/1139 and (EU) 2019/2144 and Directives 2014/90/EU, (EU) 2016/797 and (EU) 2020/1828 (Artificial Intelligence Act). [i.32] Regulation (EU) 2019/881 of the European Parliament and of the Council of 17 April 2019 on ENISA (the European Union Agency for Cybersecurity) and on information and communications technology cybersecurity certification and repealing Regulation (EU) No 526/2013 (Cybersecurity Act)". [i.33] Council of the European Union: "Resolution on Encryption - Security through encryption and security despite encryption" No. 13084/1/20, Brussels, 24 Nov 2020. [i.34] Regulation (EU) 2022/2065 of the European Parliament and of the Council of 19 October 2022 on a Single Market For Digital Services and amending Directive 2000/31/EC (Digital Services Act). [i.35] ETSI TS 102 165-3: "Cyber Security (CYBER); Methods and Protocols for Security; Part 3: Vulnerability Assessment extension for TVRA".
|
dd3b0bf937cd109481cadaa34e26502f
|
104 103
|
3 Definition of terms, symbols and abbreviations
| |
dd3b0bf937cd109481cadaa34e26502f
|
104 103
|
3.1 Terms
|
For the purposes of the present document, the following terms apply: compliance obligations: requirements imposed on parties to network communication arising from: governmental statutory or regulatory provisions or directives; judicial decisions, rules and orders; contractual obligations among providers or users; and from legal exposure to tort claims NOTE: As defined in ETSI TR 103 369 [i.10]. encryption: transformation of data by a cryptographic algorithm to produce a ciphertext Going Dark: phenomenon by which an authorized user lacks the technical or practical ability to access data NOTE: One result of going dark is the inability of an indirect party to network communication, e.g. the network operator or service provider, to meet a legal requirement or need because of pervasive encryption of the information transmitted or retained. integrity: property that data has not been altered or destroyed in an unauthorized manner perfect forward secrecy: property of an encryption system in which inspection of the data exchange that occurs during the key agreement phase of a session does not reveal the key used to encrypt the remainder of the session NOTE: This definition is slightly at variance to that found in ETSI TR 102 661 [i.16] which, in referring to asymmetric cryptographic keys, states "property that past confidentiality protected data will not be affected, if all certificates, concerning a specific time period, are revealed to an attacker" although the general role of a session key to protect only for the associated session does not allow an attacker to infer any knowledge of any key used in any other session holds for both terms. pervasive encryption: extensive encryption of data communicated "on the wire" or "at-rest" using transient techniques and practices among only a subset of the affected parties trust: level of confidence in the reliability and integrity of an entity to fulfil specific responsibilities Zero Trust Architecture (ZTA): cybersecurity model that seeks to eliminate implicit trust NOTE: In NIST SP 800-207 [1], this term is extended to address the evolving set of cybersecurity paradigms that move defences from static, network-based perimeters to focus on users, assets, and resources. ETSI ETSI TS 104 103 V1.1.1 (2025-09) 10
|
dd3b0bf937cd109481cadaa34e26502f
|
104 103
|
3.2 Symbols
|
Void.
|
dd3b0bf937cd109481cadaa34e26502f
|
104 103
|
3.3 Abbreviations
|
For the purposes of the present document, the following abbreviations apply: A2ApoA Application to Application point of Attachment A2SpoA Application to Service point of Attachment ACL Access Control List AI Artificial Intelligence ApoA Application point of Attachment CapEx Capital Expenditure CEP Communication End Point CIA Confidentiality Integrity Availability CI/CD Continuous Integration/ Continuous Delivery CSP Communications Service Provider DSA Digital Services Act E2E End to End EDF Ephemeral Diffie Helman ENISA European Network Information Security Agency ETI Encrypted Traffic Integration EU European Union GDPR General Data Protection Regulation HSM Hardware Security Module ICT Information and Communications Technologies IoT Internet of Things IP Internet Protocol LI Lawful Interception MAC Message Authentication Code NIS Network and Information Security NIST National Institute of Standards and Technology NoO Notice of Obligations OpEx Operational Expenditure OS Operating System OSI Open Systems Interconnection PII Person Identifying Information QSC Quantum Safe Cryptography RED Radio Equipment Directive RTE Run Time Explicability S2SpoA Service to Service point of Attachment S2TpoA Service to Transport point of Attachment SA Security Association SAI Securing Artificial Intelligence SpoA Service point of Attachment T2TpoA Transport to Transport point of Attachment TE Terminal Equipment TFA Technology Facilitated Abuse TpoA Transport point of Attachment TVRA Threat Vulnerability Risk Analysis UE User Equipment VPN Virtual Private Network WPA-3 Wi-Fi® Protected Access 3 ZT Zero Trust ZTA Zero Trust Architecture ETSI ETSI TS 104 103 V1.1.1 (2025-09) 11
|
dd3b0bf937cd109481cadaa34e26502f
|
104 103
|
4 Roles of encryption in networks
|
The role of encryption of information being transported between two end-points is perceived to have three widely recognized positive purposes depending on the context: • confidentiality protection of the transferred information or content; • enhanced trust in the identity of the parties associated with the information or content; and • enhanced trust in the integrity of the information or content during transport. The use of encryption as the default approach to enhance the security of communications has become increasingly common. While there are often benefits, in many scenarios the use of encryption as a default security approach exposes users and networks to threats from bad actors and malicious traffic which, if not caught in time by not being recognized as a result of being hidden behind encryption, can no longer be filtered out by the network operator to protect the network and its end users. Use of end-to-end encryption can restrict the ability of network management, anti-fraud, cyber security, and regulatory monitoring systems to manage data and communications flowing into, through, and out of networks. While encryption protects traffic flowing through a network from unauthorized inspection, encryption in itself does not protect the communicating end points from attack and reduces the ability of operators and their network management systems to remove malicious traffic by appropriate use of cybersecurity tools. NOTE: Protecting privacy is often conflated with providing confidentiality where in practical terms the intersection of the 2 concepts is incomplete. Private information can be stored and transferred confidentiality but may leak from the endpoint However, encryption may have a negative impact on third parties who do not have access to the encryption keys used and therefore do not have access to the content, but may have operational or legal responsibilities that require, or which are dependent on some level of knowledge of the information transported. Critical factors include how the keys were generated, who has knowledge of them, and how they are protected or shared. In figurative terms the impact of encryption is to make the content of something obscured or impenetrable to anyone without the key. EXAMPLE: There are many ways to describe how encryption works and one is to consider the analogy of a locked cabinet. The contents of the cabinet cannot be accessed without using the right key to unlock it. The network problem is that there are many such locked cabinets and the network cannot afford to get them mixed up. The network is also not in possession of the keys to open the cabinets to help them work out what to do with them. So conventionally each locked cabinet is marked on the outside in such a way that the network can identify who they belong to and make sure that Alice receives only her cabinets and Bob receives only his. Networks are somewhat complicated though and the reality is that, like the Russian nesting doll, cabinets are enclosed inside cabinets. In order to move cabinets to the correct destination each cabinet needs to be labelled and distinguished, and if those labels and distinguishing marks are themselves encrypted such that they cannot be easily read then their value is decreased. In the worst case scenario a Russian doll like model is tossed across a network between end points which take the subsequent dolls out of the package, if it is not for them they toss it back into the network. Encryption is often confused with and used as a shorthand in referring to a number of cryptographic processes. For the purposes of the present document, the following relatively simple meanings are adopted: • Encryption - where data is transformed by a cryptographic algorithm to produce a ciphertext with the intent to hide the information content of the data. • Trust - a trusted network is able to fulfil its obligations, if obligations cannot be fulfilled because of the use of encrypted content, it will become less trusted. • Integrity - in the social or business context integrity is closely aligned to trust that the data and the methods of handling the data cannot be altered or destroyed in an unauthorized manner. Taking the terms above and the impact of pervasive encryption into account the result is a partial denial of service to the operator. In simple terms the expected trusted relationship of the operator is denied. ETSI ETSI TS 104 103 V1.1.1 (2025-09) 12 Encryption can be accomplished as a service by multiple parties as the information is moved between endpoints. In broad communications network terms, when some parties in the process choose to apply encryption, the other parties are no longer able to trust or view the information transported across the network to endpoints. Making the operator and other stakeholders explicitly aware of the use and role of encryption, and other security techniques, in the system may allow mitigation of the negative effects of encryption whilst promoting their positive effects. The consequence for the present document is to make all security functions in a network explicit, with the further requirement to ensure that every transaction is made within a bounded set of Security Associations (SAs), while validating legitimacy of the transmitted data, that build to provide an explicit per transaction security model. The security model by being explicit shall then also be considered as making an explicit trust model for each transaction. The initial point shall be that the entire transaction and all the elements and data involved in the transaction are untrusted and insecure. The end point at which the transaction shall take place is that all elements and data in the connection are secured and trusted.
|
dd3b0bf937cd109481cadaa34e26502f
|
104 103
|
5 Model of ETI problem
|
The Going Dark challenge in which an authorized user lacks the technical or practical ability to access data has been exaggerated through the increasing use of pervasive encryption of traffic and signalling across networks - usually on an end-to-end basis. The adverse effects on cyber defence as well as a broad array of essential network operator functions are well recognized and documented (see references [i.11], [i.12] and [i.33]). Additionally, pervasive encryption poses significant difficulties for government authorities and imposition of requirements on communication service providers as documented in references [i.13] and [5]. In addition greater liability is being passed to providers of network and telecommunications services to manage access to certain forms of content which is made increasingly difficult if too much is encrypted and hidden from examination. This is addressed in part in Figure 2 with element C. In the context of the telecommunication environment Going Dark includes the inability of a normally authorized party such as the CSP's network management entity to function because of the encryption by end-point users or third parties. For example, the intersection of the two elements, A, representing network capabilities that, when content and headers are encrypted, pose extreme challenges to network operation, and B, representing Network capabilities that are core to development of cyber/digital business, should be minimized, whilst always seeking to eliminate A (see Figure 1). In addition, the relative scale of B should always be significantly greater than A. See Figure 1. Figure 1: Representation of encrypted content fostering the Going Dark in networks One consequence of pervasive encryption is that some of the obligations placed on operators and suppliers with respect to regulation, law or convention, or operator security policy, may be difficult to meet. These additional constraints on networks, may be viewed as a third dimension of the simplified Venn Diagram and are shown in Figure 2. A Network capabilities that pose extreme challenges to network operation B Network capabilities that are core to development of cyber/digital business ETSI ETSI TS 104 103 V1.1.1 (2025-09) 13 Figure 2: Refinement of the Going Dark by addition of externally imposed obligations It is recognized that in many cases the content of encrypted data should never be exposed to an ICT operator. Thus, the definition of an authorized user is made with respect to the content and the necessary header information required to transfer that content. It should be assumed that a single authorized entity is unlikely to exist for the entire distribution of data from source to destination, rather it should be assumed that distinct authorized entities exist at each layer of the normal OSI stack. In the scope of the "C" entity from Figure 2 are such things as ensuring the obligations to support Lawful Interception ETSI TS 102 656 [6], ETSI TS 101 331 [5], the GDPR [i.28], obligations under the NIS2 [i.29] directive, the Cyber Resilience Act [i.30], the AI Act [i.31] and the Cyber Security Act [i.32], and any national or regional requirements to be able to offer services. NOTE: A short summary of the impact of regulation or compliance obligations, as in element "C" of Figure 2, can be found in Annex A of the present document.
|
dd3b0bf937cd109481cadaa34e26502f
|
104 103
|
6 Technical view of the problem of ETI
| |
dd3b0bf937cd109481cadaa34e26502f
|
104 103
|
6.1 Simplified network model
|
There are a very large number of ways of presenting a telecommunications network depending on the level of abstraction to be presented. The purpose of the network model in the present document is to identify the impact of encrypted traffic on network function. Assumption#1: The present document considers packet-based communications where packets are considered within a layered communications model. Assumption#2: In a layered model of communication (e.g. the OSI 7 layer model [i.1]) the payload of layer N is not intended to be available to layer N-1 (layer N-1 is agnostic of layer N). Assumption#3: In a layered model of communication (e.g. the OSI 7 layer model [i.1]) the header of layer N informs the content of the header of layer N-1, and layer N+1. A Network capabilities that pose extreme challenges to network operation B Network capabilities that are core to development of cyber/digital business C Obligations and regulations to be met to offer ICT services ETSI ETSI TS 104 103 V1.1.1 (2025-09) 14 Terminal Equipment (representing the end point of communication) Payload 7 Application Header Payload 6 Presentation Header Payload 5 Session Header Payload 4 Transport Header Payload 3 Network Header Payload 2 Link Header Payload 1 Physical Header Payload NOTE: Whilst every layer is indicated as having a header and payload it is recognized that in some implementations the header element is not explicit (e.g. a physical layer header is unusual). Figure 3: Layering model showing data hiding layer-by-layer In general, it is only the content of the header at layer N that is required to allow layer N to perform at its optimal level. For simplicity the network is presented as an entity that allows a Communication End Point (CEP) to connect to its peer CEP. From the point of view of the application in the CEP the network should not be visible. In broad terms if the payload is opaque, it can be encrypted. However, in some instances the nature of the payload may need to be examined in order to validate the content of the header. If the payload is encrypted, and the key to decrypt is known only to the peer (i.e. the key for an encrypted payload at layer N is known only to the layer N peer and is not available to any layer N-1 entity), then it is not possible to verify the header against the content of the payload. NOTE: In end-to-end communication the terms Terminal Equipment (TE) or User Equipment (UE) are often used to refer to the end-point and may be considered as a synonym for the more general Communication End Point (CEP). If the CEP encrypts content before submission to the network stack it may be possible for the user to send prohibited information across a network. An examination of this is given in Annex C.
|
dd3b0bf937cd109481cadaa34e26502f
|
104 103
|
6.2 Layer obfuscation
|
The term layer obfuscation is used in the ETI context to address one of the many problems wherein a packet is "re-networked" such as found in most tunnelling protocols. For the Internet Protocol (IP) stack this is a case of inserting one IP packet (the inner packet) into another IP packet (the outer packet), the outer packet is visible to the network and all the network functionality is performed on the outer packet with the inner packet being carried transparently across the network. Most tunnelling protocols will encrypt the inner packet. In a VPN environment the intention is to present a remote IP packet as if it were local. In the most extreme case this form of layering within layers becomes "onion" like, and often referred to as onion- routing, as is used in the Tor project [i.14]. The Tor project further complicates onion-routing as each connection between CEPs is encrypted across a randomized set of relay points, with each relay leg having a different encryption key. ETSI ETSI TS 104 103 V1.1.1 (2025-09) 15 Figure 4: Layer overlaid on layer model as seen in VPNs and onion-routing In addition to layer obfuscation the more general model is that of layer violations. Many layer violations are designed in as part of network optimization and whilst the theoretical basis of layering is very much against layer violations the reality is that they exist and are often essential. Any dependence on layering violations is broken by extending data hiding to data encryption.
|
dd3b0bf937cd109481cadaa34e26502f
|
104 103
|
6.3 Stakeholder model
| |
dd3b0bf937cd109481cadaa34e26502f
|
104 103
|
6.3.0 General view
|
Stakeholders can be defined with respect to their relationship to the data, and to the nature of that relationship as one of owner, trusted party (each of these is considered as non-adversarial stakeholder), or adversary.
|
dd3b0bf937cd109481cadaa34e26502f
|
104 103
|
6.3.1 Adversarial stakeholders
|
The viewpoint of an adversarial stakeholder is to use pervasive encryption with an explicit intent to avoid any form of oversight (i.e. to explicitly act against the interests of parties in group A and C of the Venn diagram from Figure 2).
|
dd3b0bf937cd109481cadaa34e26502f
|
104 103
|
6.3.2 Non-adversarial stakeholders
|
A non-adversarial stakeholder uses encryption as offered by legitimate parties in order to ensure that privileged information is only made visible to authorized parties.
|
dd3b0bf937cd109481cadaa34e26502f
|
104 103
|
6.3.3 Network management stakeholders
|
Network managers are a special case of non-adversarial stakeholders that are network-resident and who aim to optimize availability of networks to serve the other non-adversarial stakeholders. In particular, one of the roles of network management stakeholders is to be able to inhibit the actions of adversarial stakeholders. The ability to validate traffic and signalling is critical to the success of network management. ETSI ETSI TS 104 103 V1.1.1 (2025-09) 16
|
dd3b0bf937cd109481cadaa34e26502f
|
104 103
|
7 Quantitative risk assessment
| |
dd3b0bf937cd109481cadaa34e26502f
|
104 103
|
7.1 Overview
|
The TVRA approach defined in ETSI TS 102 165-1 [7], and expanded on in ETSI TS 102 165-3 [i.35] for continuous vulnerability assessment in the context of the CSA [i.32], identifies risk as the product of the impact and likelihood of an attack. For assessment of likelihood a number of metrics are considered that determine the level of expertise, time and equipment required by the attacker to conduct the attack. In addition the method addresses how the motivation of an attacker may determine the willingness to achieve an appropriate level of technical knowledge to mount an attack.
|
dd3b0bf937cd109481cadaa34e26502f
|
104 103
|
7.2 Impact and likelihood assessment
|
The core concern of pervasive encryption has been outlined in clauses 4, 5 and 6 of the present document. In order to express the problem more quantitatively the risk calculation approach defined in ETSI TS 102 165-1 [7] is applied below. Risk is assessed as the weighted product of the likelihood of a threat, and the impact of that threat. As outlined in previous clauses of the present document the impact of pervasive encryption is scenario, or use case, dependent. In contrast the ease of deploying strong encryption indicates that, in general technical terms, the likelihood of deploying encryption is very high. This is shown in the assignment of the likelihood factors from ETSI TS 102 165-1 [7] to the deployment of end-to-end or pervasive encryption in Table 1 and the resulting assessment of attack potential as basic shown in Table 1 and Table 2. Table 1: Quantization of attack potential for application of encryption Factor Range (assigned file) Value Scoring rationale Time (elapsed time) ≤ 1 day 0 The level of detail widely available for applying encryption suggests that an attack can be planned and deployed in less than 1 day. Expertise Layman 0 Tools exist to easily switch encryption on without having to have detailed knowledge of the cryptography that underpins it. In part this is because most computers and their operating systems include basic capabilities to apply encryption. Knowledge Public 0 There is widespread publicly available data to guide a non-expert to apply encryption at end points on a public connection. Opportunity Easy 1 Most of the tools for applying encryption exist by default in common ICT platforms. Equipment Specialized 4 There may be a requirement for slightly more advanced software than comes bundled with a computer and its operating system. As with the gathering of expertise and knowledge however most of this equipment is readily available with some effort to search it out. From ETSI TS 102 165-1 [7] the attack potential is translated into natural English terms as in Table 2. Table 2: Quantization of attack resistance to application of encryption Attack potential values Attack potential required to exploit attack Resistant to attacker with attack potential of 0 to 9 Basic No rating With the assumption that the impact from random application of encryption is non-negligible the likelihood of attack is considered as at least "possible" and often "very likely" as described in ETSI TS 102 165-1 [7]. Depending on the use case (see clauses 4 and 5, and Annex B for an overview of some use cases where pervasive or maliciously applied encryption is deployed) the level of risk measured using the approach described in ETSI TS 102 165-1 [7] is either Major or Critical in most cases. ETSI ETSI TS 104 103 V1.1.1 (2025-09) 17 EXAMPLE 1: A ransomware attack in which assets are encrypted by an attacker using symmetric keying with the key released on payment of the ransom, or simply destroyed if the ransom is not paid, may result in critical and irrecoverable loss of essential assets. Overall risk assessment of ransomware should be that it is a critical risk. EXAMPLE 2: Masking of essential routing data by use of layer masking (a Tor like network or a simple lower layer VPN) may result in continuous reinjection of traffic to the network without it terminating efficiently and thus lowering network efficiency. In the extreme cases this may lead to over investment in devices (CapEx) and raise operational costs (OpEx) to cope with maliciously formed traffic and could place the network organization at critical risk levels. As indicated in the examples pervasive encryption can give rise to increases in both Capital Expenditure (CapEx) and Operational Expenditure (OpEx) by reducing the achievable efficiency of networks and services.
|
dd3b0bf937cd109481cadaa34e26502f
|
104 103
|
7.3 Motivation assessment
|
It can be suggested that the application of encryption to user content is somewhat benign and is motivated by a reasonable desire on behalf of the user applying the encryption to satisfy the security tenet of least privilege. Thus it is reasonable to encrypt the details of a bank transaction as the network provider is not a privileged user of that content.
|
dd3b0bf937cd109481cadaa34e26502f
|
104 103
|
7.4 Countermeasure considerations
|
The guiding principle of countering ETI problems is visibility, transparency, and the provision of a trust architecture. The general model of countermeasure identified in ETSI TS 102 165-1 [7] is that they (the countermeasures) are assets that are added to the system to reduce the weighted risk to the system. The purpose of countermeasures is to reduce either the likelihood of an attack or the attack impact. The general security model can be summarized in the triplets shown below for each of the model (i.e. the general ETI concern) and its implementation (i.e. means to address the ETI concern): • Model: {threat, security-dimension, countermeasure} • Implementation (for ETI): {pervasive encryption, integrity, transparency and explicability} The specific threat considered in the present document is "pervasive encryption". In this case the security dimension most impacted is integrity, in that the threat may not allow managed operation of the network, and may allow for prohibited content to be distributed. The latter threat in particular impacts the level of trust afforded to the network by its users and therefore impacts the security dimension of integrity. NOTE 1: The meaning given to integrity in this instance is that a trusted network acts honestly and without masking its behaviour from its stakeholders. NOTE 2: A framework of security countermeasures is described in ETSI TS 102 165-2 [8]. Whilst the present document identifies that in order to counter the threat to integrity from pervasive encryption the network and is constituent elements shall be both transparent and explicable it is recognized that there may be several alternative countermeasures and these should first be identified, then evaluated and compared to identify the costs and benefits of each so that an informed decision can be made of which countermeasures to select. The model of static and run-time transparency and explicability given in ETSI TS 104 224 [9], and the application of the ZT-Kipling method (in ETSI TS 104 102 [2]) in the context of ETSI TS 102 165-1 [7] may be used to provide the required countermeasure to the threat of pervasive encryption (see also clause 8.2). ETSI ETSI TS 104 103 V1.1.1 (2025-09) 18
|
dd3b0bf937cd109481cadaa34e26502f
|
104 103
|
8 Requirements to counter the ETI problem
| |
dd3b0bf937cd109481cadaa34e26502f
|
104 103
|
8.1 Introduction and overview of means to counter ETI
|
An ETI conformant network shall be able to demonstrate that for each connection there is an established trust contract, and an associated security contract. The present document identifies the role of Zero Trust Architecture (ZTA) [1] and its close association to an active (rather than static) implementation of the secure by default paradigm [i.19] in providing the transparency and explicability of the use of encryption, while adhering to ZTA principles within related technologies to mitigate the ETI problem. Thus the core requirements to counter the ETI problem are transparency and explicability of the role and purpose of every element in the network. NOTE: Whilst the present document uses the ZT and ZTA paradigm to achieve the transparency and explicability requirements other approaches may be used to achieve the same objectives. In a broad interpretation of the ETI problem statement addressed in clauses 4, 5 and 6, and specifically in clause 7.4 above of the present document, it is surmised that making the operator and other stakeholders explicitly aware of the use and role of encryption, and other security techniques, in the system will allow mitigation of the negative effects of encryption whilst promoting their positive effects. The consequence for the recommendations in the form of requirements outlined in the present document is to make all security functions in a network explicit, with the further requirement to ensure that every transaction is made within a bounded set of Security Associations (SAs), while validating legitimacy of the transmitted data, that build to provide an explicit per transaction security model. The security model by being explicit should then also be considered as making an explicit trust model for each transaction. The initial point should be that the entire transaction and all the elements and data involved in the transaction are untrusted and insecure. The end point at which the transaction should take place is that all elements and data in the connection are secured and trusted. As stated above, to support ETI the operator and other stakeholders should be explicitly aware of the use and role of encryption, and other security techniques, in the system. Thus all security functions in the ICT system or network that is ETI compliant shall be explicit. In practical terms this requires that each security function is transparent and the rationale for each function is explicable.
|
dd3b0bf937cd109481cadaa34e26502f
|
104 103
|
8.2 Transparency
|
It shall be possible for an authorized entity to request the encryption state of any connection or data at any involved, identifiable, and addressable object (hereinafter referred to as an entity), in the ICT system by direct interrogation of the entities participating in the connection. The requesting entity should be within the same trusted environment of the target entity (the one whose state is being interrogated) and therefore has to be identified and authenticated before being allowed to operate on the entity. The encryption state of any connection shall be reported as one of: encryption applied; encryption not applied; or unknown. In order to enable the transparency requirement, all entities in the communication chain shall have a well-defined (i.e. standardized) point of inspection, and a standardized query interface should be used. As part of the goal of achieving transparency the sub-goals of accountability and explicability are considered. For this to be achieved the base requirement of transparency above shall apply to the connection as a whole and the context in which encryption is applied. NOTE: A number of means exist that attempt to verify and provide proof of the path any packet has taken across a network by addition of data to the packet header for 3rd party verification. The approach in the present document is to develop a trust and security contract for the connection that provides an alternative approach to achieve such proofs. The model for explicability and transparency is identified below and summarized in Figure 5. ETSI ETSI TS 104 103 V1.1.1 (2025-09) 19 Figure 5: Components required in element documentation for transparency Every element shall be identified and able to explain its purpose in the system. Every element shall identify the forms of security association it supports, and for each security association the root of trust (as the point of liability) shall be identifiable. The general approach to static and run-time explicability identified in ETSI TS 104 224 [9] should be augmented by application of the ZT-Kipling method (ETSI TS 104 102) [2]. Table 3, Table 4, Table 5 and Table 6 below act as examples of the application of each of ETSI TS 104 102 [2] and ETSI TS 104 224 [9] to the pervasive encryption problem. Table 3: Application of ETSI TS 104 102 [2] in the ETI context Question form Question text example What What is being encrypted? (e.g. data at rest, data in transit, consumer data, signalling data) Why Why is encryption being applied (to that asset/data/signalling)? When When is the encryption applied (and removed)? How How is the encryption applied? (e.g. symmetrically, asymmetrically, transparently to the end-user, with the collusion of one or more other parties) Where Where is the encryption applied? (logically (say the OSI layer) and geographically)? Who Who applies the encryption? (i.e. who has the keys for the operation?) Table 4: Static explicability statement for ETI and the role of encryption (from ETSI TS 104 224 [9]) Documentation Element Element 1 Statement of system purpose 2a Identification of data source(s) 2b Purpose of data source(s) (in support of system purpose) 2c Method(s) used to determine data quality 3 Identity of liable party The purpose of explicability is to allow a lay person (i.e. not a professional programmer or system analyst) to gain a reasonable understanding of the main data flows and processing steps in the program. Table 5: Run time explicability statement ETI and the role of encryption (from ETSI TS 104 224 [9]) Documentation Element Element RTE-1 Static explicability statement RTE-2 What process does data undergo between acquisition and curation? RTE-3 What are the metrics that determine change in the learning/weighting of data? RTE-4 Identification of events to be logged RTE-5 Identification of performance target and associated metrics RTE-6 Identification of liable party (if different from that identified in the static explicability documentation) Element purpose Element security associations supported Element point of liabiilty ETSI ETSI TS 104 103 V1.1.1 (2025-09) 20 Table 6: Transparency statement for ETI and the role of encryption (from ETSI TS 104 224 [9]) Documentation Element Element T-1 Static explicability statement T-2 Run time explicability statement For each data source T-3a Verified identification of source of data T-3b Verified proof of liability of data source T-3c Verified proof of consent to use
|
dd3b0bf937cd109481cadaa34e26502f
|
104 103
|
8.3 Management of cryptographic keys
|
The trend in cryptographic protection is towards perfect forward secrecy in which session keys cannot be compromised even if the root key from which the session keys are derived is itself made known. Ephemeral keys are a consequence or attribute closely associated with trying to achieve forward secrecy. A key is described as ephemeral when it is created uniquely for each key establishment process. The assurance of forward secrecy requires that the ephemeral session key is discarded after use. For the purposes of ETI the legitimate use of forward secrecy should be maintained for each Security Association (SA) in a transaction. However the form of key agreement should be visible to authorized parties.
|
dd3b0bf937cd109481cadaa34e26502f
|
104 103
|
8.4 Identification of authorized parties
|
The encryption state of any connection of an ETI conformant system shall only be disclosed to authorized parties. An authorized party shall be unambiguously identified and that identity shall be authenticated. The identity of the authorized party may take a number of forms including those defined in ETSI TS 103 486 [i.20] and using forms of attribute rich identity coupled to attribute based authentication modes as described in ETSI TS 103 486 [i.20].
|
dd3b0bf937cd109481cadaa34e26502f
|
104 103
|
8.5 Trust architecture for ETI
|
A layered communications architecture, as defined for OSI in ISO/IEC 7498-1 [i.1], has implicit trust relationships at each layer determined by the functional model of each layer. The present document extends the OSI model to a wider concept of ZTA as in NIST SP 800-207 [1] beyond the enterprise network to a full public telecommunications network addressing the particular model described in clause 3.1.3 of [1] (ZTA Using Network Infrastructure and Software Defined Perimeters) to the entirety of the OSI stack. ZT a security strategy (or approach) is designed to detect and prevent breaches, while consistently (or better continuously) verifying all users, all devices, all layers (e.g. OSI layers, signalling, data, management, etc.), all applications across all locations in the real time (run time), and applying Continuous Integration and Continuous Delivery (CI/CD) pipeline security, resulting in preventative security from all attack vectors at all stages of the attacks. The rationale of the ZTA is that there should be no assumptions as to what happens before or after each hop in and across the infrastructure, starting with the source and ending with the destination of a particular data flow. Every device, application, microservice, and user shall be validated and verified in real time. With each step a user (or the proxy for the user) makes through the infrastructure, the following two aspects provide adherence to ZTA: • ETI conformant systems shall provide means to validate, authenticate, and apply threat prevention capabilities across all locations consistently. • ETI conformant systems shall provide means to validate, authenticate and verify all users, all devices, and all applications, and apply threat prevention capabilities across all locations to protect against all attack vectors consistently. • ETI conformant systems shall be able to validate the "who", the "what", the "where", the "when", the "why", and the "how" across all traffic flows throughout the lifecycle of those flows. NOTE: Applying the ZT-Kipling method defined in [2] addresses the means to do the "who", "what", "where", "when", "why", and "how" validation. ETSI ETSI TS 104 103 V1.1.1 (2025-09) 21 Whilst the abstracted model from ETSI TS 102 165-2 [8] is recommended for addressing the ETI problem it is recognized that the managed security of networks is broadly addressed by the following services as defined by the OSI 7-layer security model (see table 2 of Recommendation ITU-T X.800 [4] and its mirror ISO 7498-2 [i.18]) and the aspects of ZTA [1]: • At layer 7: - Identity Assertion ("who"). - Application Validation ("what"). - To-be-accessed Targeted Resources Destination ("where"). - Data Flow Time-stamping ("when"). - Data Classification ("why"). - Identity Assertion of the Targeted Resources Access ("how"). - Peer Entity Authentication. - Data Origin Authentication. - etc. • At layer 6: - Facilities provided by the presentation layer offer support to the provision of security services by the application layer to the application process. - The facilities provided by the presentation layer rely on mechanisms which can only operate on a transfer syntax encoding of data. - Security mechanisms in the presentation layer operate as the final stage of transformation to the transfer syntax on transmission, and as the initial stage of the transformation process on receipt. • At layer 5: No security services are provided in the session layer. • At layer 4: - Peer Entity Authentication; - Data Origin Authentication; - Access Control service; - Connection Confidentiality; - Connectionless Confidentiality; - Connection Integrity with Recovery; - Connection Integrity without Recovery; and - Connectionless Integrity. • At layer 3: - Peer entity authentication. - Data origin authentication. - Access Control List (ACL). - Connection confidentiality. - Connectionless confidentiality. ETSI ETSI TS 104 103 V1.1.1 (2025-09) 22 - Packet flow confidentiality. - Connection integrity without recovery. - Connectionless integrity. • At layer 2: - Connection confidentiality. - Connectionless confidentiality. • At layer 1: - Connection confidentiality. - Traffic flow confidentiality. Each security service in the OSI model exists as a peer to peer service, i.e. network layer to network layer, application layer to application layer. Each layer has an implicit security association determined by the key used to protect the services at that layer. The model in the present document extends the OSI peer-to-peer model with the ZTA defined in [1] and addresses the identity management requirements as an instance of the model for ZTA Using Enhanced Identity Governance ([1], clause 3.1.1). To be considered as ETI layers shall be developed as independent trust zones with clear visibility (see clause 8.2 Transparency) of the services offered. By the term independent trust zone it is intended that each layer should have autonomy from any other layer. As Figure 2 illustrates, ZTA adds the following attributes to each of the OSI layers: • Layer 7: Data Validation and Integrity check, Threat Intelligence, Microsegmentation of Services; Identity Assertion. • Layer 6: This is folded into Layer 7. • Layer 5: This is folded into Layer 7. • Layer 4: Pure Access Control Lists do not guarantee ZTA, hence this is folded into Layer 7. • Layer 3: Microsegmentation, Data Source validation, Data Destination validation, Data Destination Authentication. • Layer 2: This is folded into Layer 3. • Layer 1: Authenticity and validation checks of all hardware and software components. In summary, ZTA follows the principle of "never trust, always verify". The model of trust, on the other hand, is that whilst content of user communication may view the network as untrusted and the user may choose to apply application plane services to ensure confidentiality of user content, the lower layers are themselves contained in layer specific trust relationships. In this way all data required to enable layer operations should be visible to that layer. ETSI ETSI TS 104 103 V1.1.1 (2025-09) 23 Figure 6: Representation of ZTA mapping to OSI layers In general, whereas trust can be defined in spoken and written English as "firm belief in the reliability, truth, or ability of someone or something" this has to be translated into something more tangible and exact for ICT systems and has often been simplified into the assertion of integrity of an object where that assertion is made, or attested to, by a known entity. A number of forms of integrity assertion exist, summarized in ETSI TS 102 165-2 [8], including the use of Message Authentication Codes (MACs), Hash functions, and digital signatures. The present document suggests that security association is used as a synonym for trust association. An end-to-end connection is composed of at least one and more likely an indeterminate, but finite, number of security associations. E2E security is thus not just an end-point issue but a composition of SAs issue. Each security association is also a representation of a trust association and for the purposes of the present document trust is a weighting that applies to a security association. ∑. Each security association should be protected by a unique key. For compositions of security associations, where one security association is dependent on another security association within the overall composition, they should be protected by different (and also unique) keys. The nature of an SA, one-to-one, one-to-many (including broadcast), many-to-one and many-to-many, has a significant influence on the selection of keying strategy to protect to the SA, details of keying strategies are addressed in ETSI TS 102 165-2 [8]. NOTE 1: An SA does not imply encryption but rather explicitly identifies the nature of the security association, e.g. an SA and key for each of integrity/confidentiality/availability. Therefore SA should not be read as shorthand for encrypted link. In the ETI model each link shall represent trust for each attribute of the CIA paradigm: • How is data encrypted and decrypted? Who establishes the keys? ETSI ETSI TS 104 103 V1.1.1 (2025-09) 24 • Source authentication as a prerequisite for the other attributes. • How is data integrity preserved? • How is the identity of the asset and link assured? • How is access control to data and services related to the link assured? When the ZTA model is completed the result is a trust contract between end points that details the form of security association on each link and at each layer. NOTE 2: Not all links will be explicit as some links will essentially be passive (layer 1 and 2 links often have no complex security associations). The role of trust contracts is addressed in part by obligation of trust protocols (see ETSI TS 103 486 [i.20] and also ETSI TR 103 719 [i.8]). 8.6 Reference model of an ICT network for ETI The security model for ETI is developed from that found in ETSI TS 102 165-2 [8] and copied below for convenience. In this latter model the OSI model [i.1] is simplified to 3 planes: Transport; Service and Application. Between planes both horizontally and vertically are "points of attachment" and it is at these points of attachment that the security services lie. The transport plane approximates to the lower layers of the OSI model, the service plane approximates to the higher layers of the OSI model, with the application plane addressing the user level application. NOTE 1: The specific terminology from ETSI TS 102 165-2 [8] is drawn from a traditional telecoms consideration but with a relaxed interpretation can be mapped to non-telecoms environments, including those of conventional programming, to business practices and similar. Figure 7: Abstract architecture for security countermeasure application from ETSI TS 102 165-2 [8] The user connects to each layer using a layer specific point of Attachment (poA): • TpoA Transport point of Attachment (TpoA reference point). ETSI ETSI TS 104 103 V1.1.1 (2025-09) 25 • SpoA Service point of Attachment (SpoA reference point). • ApoA Application point of Attachment (ApoA reference point). The countermeasures are described with respect to the user interaction with each layer: • Inbound authentication at TpoA/SpoA/ApoA. • Outbound authentication at TpoA/SpoA/ApoA. NOTE 2: If an authentication exchange nests inbound and outbound authentication, it is termed mutual authentication. However if the exchanges are discrete and with different lifetimes the term mutual authentication is inappropriate. • Integrity of communication at TpoA/SpoA/ApoA. • Confidentiality of communication at TpoA/SpoA/ApoA. Within the system the countermeasures are extended to cover interactions between layers both vertically and horizontally. The set of countermeasures thus include: • Service to Service authentication. • Integrity of communication from Service to Service. • Confidentiality of communication from Service to Service. NOTE 3: The term Service is used as a synonym for any of the three abstract layers of the ICT architecture. The services apply to the following points on Figure 4: • A2SpoA Application to Service reference point. • S2TpoA Service to Transport reference point. • A2ApoA Application to Application reference point. • S2SpoA Service to Service reference point. • T2TpoA Transport to Transport reference point. NOTE 4: The model does not show a specific reference point between Application and Transport on the assumption that a Service layer always exists. In addition to the countermeasures provided at the identified reference points a secure system may have to deploy other countermeasures to protect their assets. Such countermeasures may include billing controls, system auditing and event logging. ETSI ETSI TS 104 103 V1.1.1 (2025-09) 26 Annex A (informative): Considerations for Compliance Obligation A.1 Overview The compliance obligations described in this annex are examples and do not imply that these are the only obligations that apply to communication networks, or that these are the only obligations affected by pervasive encryption in networks. NOTE: A full enumeration of compliance obligations for communication networks is provided in ETSI TR 103 369 [i.10]. A.2 EU GDPR The General Data Protection Regulation (GDPR) [i.28] suggests the use of encryption several times across several articles. For example, Articles 6.4e, 32.1a, 34.3a all suggest use of encryption. It is possible that an over enthusiastic interpretation of these articles leads to a data controller recommending that everything is encrypted at all times. However, such an interpretation does not eliminate the data protection responsibility, as the end points of the communication would normally decrypt the data that has been encrypted whilst in transit. This happens especially with the reception of Personal Identifying Information (PII). Once everything has been decrypted (even if only for temporary processing) obligations for data protection apply in regard to that content. A.3 EU NIS2 Directive The Network and Information Security Directive (NIS Directive) [i.29] applies in particular to two forms of commercial entity: 1) Operators of essential services 2) Digital service providers In ETSI TR 103 456 [i.5] and in the ENISA report on gaps in NIS standardization [i.6] a wide overview of the impact of the NIS Directive [i.29] on networks and on standardization is given. These reports however have not fully addressed the impact on network management from the forms of pervasive encryption addressed by the present document. Many of the goals of the NIS Directive [i.29] may be thwarted, e.g. Article 7 requires this capability as part of a national strategy "defining the strategic objectives and appropriate policy and regulatory measures with a view to achieving and maintaining a high level of security of network and information systems". Many risks may be invisible as a result of the application of end-to-end encryption. Where risks impact essential services provided at the network edge, there is a clear risk of being unable to conform to the core requirements of the NIS Directive [i.29] when end-to-end encryption is deployed. In broad outline the purpose of the NIS Directive [i.29] is to enable EU Member states to provide legal measures that in turn invoke a set of common cyber security technical requirements that include: • structured sharing of information on risks and incidents; • notification of incidents; • outcomes-focused cybersecurity risk management practices and controls to identify and protect assets, detect anomalous analyses and potential incidents, and respond to and recover from incidents that may impact network and information systems; • international cooperation to improve security standards and information exchange, and promote a common global approach to NIS issues through harmonised standards. ETSI ETSI TS 104 103 V1.1.1 (2025-09) 27 The impact of end-to-end encryption of both content and signalling on management of the NIS Directive [i.29] may be such that the necessary analysis to allow information sharing may be severely impeded. A.4 Council of the EU Resolution on Encryption The EU has adopted the "Council Resolution on Encryption - Security through encryption and security despite encryption" [i.33]. The resolution contains several clauses describing objectives, the current use/state of encryption, challenges for ensuring security, striking a right balance, joining forces with the tech industry, a need for a regulatory framework, and innovative investigative examples. Concerning a regulatory framework, the resolution notes: • The need to develop a regulatory framework across the EU that would allow competent authorities to carry out their operational tasks effectively while protecting privacy, fundamental rights and the security of communication could be further assessed. • Potential technical solutions will have to enable authorities to use their investigative powers which are subject to proportionality, necessity and judicial oversight under their domestic legislation, while respecting common European values and upholding fundamental rights and preserving the advantages of encryption. Possible solutions should be developed in a transparent manner in cooperation with national and international communication service providers and other relevant stakeholders. Such technical solutions and standards - and the fast development of technology in general - would also require continually improving the technical and operational skills and expertise of competent authorities to effectively address the challenges of digitalization in their work on a global scale. The present document, and any follow-on work undertaken, may in part address the requirements for further assessment and the transparent development of solutions, outlined in the resolution. A.5 EU Cybersecurity Act The EU Cybersecurity Act [i.32] adopted in April 2019 provides "a framework for the establishment of European cybersecurity certification schemes for the purpose of ensuring an adequate level of cybersecurity for ICT products, ICT services and ICT processes in the Union, as well as for the purpose of avoiding the fragmentation of the internal market with regard to cybersecurity certification schemes in the Union". It tasks ENISA with numerous related functions. The Cybersecurity Act does not, however, explicitly mention or treat encryption except in a legislative history paragraph encouraging ENISA "to promote basic multifactor authentication, patching, encryption, anonymization and data protection advice". A.6 ePrivacy Regulation In 2017, a draft proposal referred to as ePrivacy Regulation [i.3] was introduced in the European Parliament and the Council. Its many provisions lay down rules regarding the protection of fundamental rights and freedoms of natural and legal persons in the provision and use of electronic communications services, and in particular, the rights to respect for private life and communications and the protection of natural persons with regard to the processing of personal data. It has not, however, further progressed. It also has no specific provisions related to encryption except in a legislative history paragraph that mentions that "Service providers who offer electronic communications services should inform end- users of measures they can take to protect the security of their communications for instance by using specific types of software or encryption technologies". ETSI ETSI TS 104 103 V1.1.1 (2025-09) 28 A.7 Radio Equipment Directive The Radio Equipment Directive 2014/53/EU ("RED") [i.2] ensures a single market for radio equipment. In particular, it requires that, before being placed on the market, radio equipment has to incorporate safeguards to ensure that the personal data and privacy of the user are protected. Under the RED [i.2] and the European Standardization Regulation (EU) 1025/2012 [i.4], the Commission is empowered to adopt measures that determine access to markets. For the bulk of the RED [i.2] only radio aspects are considered and the test conditions are specified in Harmonised ENs. However, there are aspects of the RED [i.2] in Article 3.3d/e/f that address security and privacy that may be impacted by end-to-end encryption. The specific impact of persistent and pervasive encryption as addressing RED Article 3.3d/e/f are outlined as follows. "3. Radio equipment within certain categories or classes shall be so constructed that it complies with the following essential requirements: (d) radio equipment does not harm the network or its functioning nor misuse network resources, thereby causing an unacceptable degradation of service;". In this instance the problem statement given in respect of network management stakeholders (see clause 6.3.3 of the present document) applies. "(e) radio equipment incorporates safeguards to ensure that the personal data and privacy of the user and of the subscriber are protected;". Here it may be argued that applying encryption will achieve this. "(f) radio equipment supports certain features ensuring protection from fraud". The following publications address the security aspects of the RED: EN 18031-1 [i.25], EN 18031-2 [i.26], and EN 18031-3 [i.27]. A.8 Lawful access to communication and communications data ETSI TS 102 656 [6] and ETSI TS 101 331 [5] define the broad set of requirements from law enforcement to telecommunications. The expectation is that data is provided en-clair where keys are known to the provider, or in native format when keys are not available. The obvious impact of end-to-end encryption is that less content can be offered en- clair, and if the imposition of end-to-end encryption extends to signalling that too cannot be offered en-clair. A.9 Digital Services Act The DSA [i.34] provides a set of rules regulating the responsibilities of digital services that act as intermediaries within the EU to connect consumers with goods, services and content. In this context, 'digital services' refers to online platforms, such as marketplaces and social media networks, it can also be extended to consider apps that implement such services. The DSA sets out clear due diligence obligations for online platforms and other online intermediaries. This is a different dimension of security than much of existing telecommunications in that it looks at content and its legitimacy under the law. The DSA addresses the requirement to ensure platforms remove fraudulent products and content from being distributed and gives tools to ensure that users can flag illegal or disturbing content, and will also have a clear means of contesting platforms' content moderation. A.10 Cyber Resilience Act The Cyber Resilience Act [i.30] identifies a number of essential requirements to be met prior to placement of things with digital elements on the market (for the EU). The approach and requirements outlined in the present document provide a baseline for achieving the essential requirements for the run-time security of networks, services and applications. ETSI ETSI TS 104 103 V1.1.1 (2025-09) 29 A.11 Artificial Intelligence Act The Artificial Intelligence Act (EU AI Act) [i.31] lays out the foundations for regulating AI in the EU. The AI Act classifies AI according to its risk, as follows: prohibited unacceptable risk; regulated high-risk; limited risk, where deployers and developers are required to ensure that end-users are aware that they are interacting with AI; and unregulated minimal risk. ETSI TS 104 223 [10] defines a set of 13 principles for the provision of secure AI, in addition ETSI TR 104 065 [i.24] maps the requirements of the AI Act [i.31] to the wider output of ETSI. Whilst AI, and the AI Act, do not require data sources or algorithms to be encrypted there is a risk that by abusing encryption maliciously formed data may not be identified and may lead to unintended consequences. Annex B (informative): Encryption tools B.1 Architectures and schemes To a limited extent the form of encryption used by actors will influence the strategies used to mitigate the worst impacts of encrypted traffic and signalling on network operation. In simple terms the success of cryptography is in the management of keys, on the assumption that the underlying algorithms have been tested and proven to achieve their intended security strength. Key management strategies and the matching algorithms fall into a small set of classes: • Symmetric encryption: - Only the end points have access to the key. • Asymmetric encryption: - One key to lock, a matching key to unlock. Often termed public key encryption in that one key of the pair can be made public with close to zero likelihood of an adversary determining the private key. - Widely used in e-commerce and large parts of the internet to protect the content of transactions, and widely built into core protocols. • Functional encryption: - Functional encryption is a generalization of asymmetric encryption. It is seen in two primary forms where the public key element has semantic meaning (e.g. an email address): Identity based encryption (see ETSI TR 103 618 [i.7], ETSI TR 103 719 [i.8]). Attribute based encryption (see ETSI TS 103 532 [3], ETSI TS 103 458 [i.9]). • Homomorphic encryption: - A form of encryption that allows operations on encrypted data without decrypting it first. The result of the computation is in an encrypted form, when decrypted the output is the same as if the operations had been performed on the unencrypted data. In addition, due attention should be paid to the evolution of the above strategies to mitigate the threat from Quantum Computing and to the development of algorithms resistant to Quantum Computing attacks, as defined in the output of ETSI TC CYBER QSC (Quantum Safe Cryptography). ETSI ETSI TS 104 103 V1.1.1 (2025-09) 30 B.2 Additional key management issues From the outline of clause B.1 above, there are two forms of keying, symmetric where both parties have the same key, and asymmetric where the keys are paired with one key used for encryption and one for decryption. A general rule of thumb is that symmetric encryption is fast, and asymmetric encryption is slow. A second rule of thumb is not to over expose a key, so in the same way that users are recommended to use different passwords for every site or purpose, it is conventional practice to use a "session key" for every session. If a new session key is used, and if the new session key cannot be linked to any other session key from the same user, then even if the key for a single session is compromised only that session is compromised, this is common practice in cellular radio. The means by which a session key is derived is an important consideration, recognizing weaknesses that may lead to a loss of perfect forward security has seen a move to ephemeral key agreement schemes. With some key exchange methods identical keys will be generated if the same parameters are used on either side. The role of ephemeral methods is to guarantee that a different key is used for each connection, with the addition of protection against a store and extract attack. Therefore, an attack on any long-term key would not cause all the associated session keys to be breached, halting the attempt to recover data encrypted with those session keys (this offers a guarantee of perfect forward secrecy). Due to the promise of ephemeral keys giving guarantees of perfect forward secrecy they are being embedded in many of the commonly used network protocols. This includes TSLv1.3 for protection of IP traffic, WPA-3 for Wi-Fi® networks, and use of Ephemeral Diffie Helman (EDF) in key exchanges. ETSI ETSI TS 104 103 V1.1.1 (2025-09) 31 Annex C (informative): Case analysis of impact of end-to-end encryption C.1 Criminal activity - general The term Going Dark has been used by law enforcement authorities for several decades and was specifically cited by the U.S. Office of the Director of National Intelligence [i.13] as a major concern for law enforcement. Figure C.1: Law enforcement view of Going Dark Figure C.1 shows the "classic" Going Dark Venn diagram, where A and B intersect, society gets worried. Crimes that fall into the intersection include drug crimes, terrorism and child sexual exploitation. When using telecommunications in aid of criminal activity the criminals effectively hide in plain sight and use technology to mask their activity. The problem of "Going Dark" is that law enforcement are always playing catch up, and the inherent danger is that the gap between what is feasible to gather evidence on and the ability of the criminal to mask their activity is one of many unknown unknowns (the existence of crime is a known unknown, it is known to be going on, but it is often unknown who is involved and where and how to stop the criminal having a technological advantage). The problem that ETI is faced with, is to enable reasonable legal access to data, such as images, to make a value judgement on the content without violating privacy or denying legitimate use of certain data. This is in addition to having reasonable access to each protocol layer's header to allow for network capabilities to be optimized (in practice this is giving some access to Layer N+1, i.e. to the layer above, which means some restricted override of data hiding). If criminal extortion over telecommunications takes place it is unlikely to be limited to a single channel, rather it is more likely to occur over multiple channels belonging to the victim (e.g. social media channels, direct telecommunications links). The problem of pervasive end-to-end encryption makes discovery and investigation hard. C.2 Cryptovirology At one extreme of the rise of pervasive encryption is the specific application of cryptography to malicious ends. The insight that encryption could be used to skirt the protections provided by the network has led to the creation of a new field of study: cryptovirology. This field is devoted to studying the methods in which cryptography may be used to design powerful malicious software. The first cryptovirology attack, "cryptoviral extortion" [i.15], was presented at the 1996 IEEE Security & Privacy conference. This attack featured either a cryptovirus, crypto worm, or cryptotrojan, containing the public key of the attacker, which would then hybrid encrypt the victims' files, the term for this type of attack was later coined to be ransomware. A crimes for which various manifestations of technology pose extreme challenges to law enforcement B crimes for which society demands an exceptionally high level of effective prevention, investigation, and prosecution ETSI ETSI TS 104 103 V1.1.1 (2025-09) 32 C.3 Melissa virus and malicious software distribution Many computer viruses and other forms of malware have to a greater or lesser extent, relied upon encryption to bypass protection and infect computer end points. Examples include the Melissa virus dating from 1999, the Morris worm, which was the precursor to Melissa dating from 1988, and attacks such as those performed by ransomware. The Melissa virus was a fairly simple macro-virus, used to distribute content by inveigling itself to commercial email programs and accessing contact data, then distributing itself through the email program to a subset of the infected account's contacts. Whilst not of itself particularly malicious, Melissa spawned the development of much more malicious code by proving the value of a delivery mechanism, and gave impetus to the study of cryptovirology, the application of cryptography for implementation of malicious software. If payloads can be encrypted, for example using a Melissa-like virus accessing contact lists where the public key of the contact is visible, then a malicious payload can be distributed and protected from observation by the network through end-to-end encryption. Given that the recipient is receiving mail from a known and trusted contact the likelihood of successful transmission of the viral and malicious content is more assured. Whilst whole disk encryption is commonly applied by computer users at end points (minimizes risk in event of loss or theft of the computer or its disk), variations of it may be applied with malicious intent. This is somewhat exacerbated with most modern computers having crypto-acceleration built in, having Hardware Security Modules (HSMs), and having direct access to crypto-libraries from the OS. The generic fields of Ransomware and of Kleptography use attacker-controlled encryption to variously encrypt critical files on a computer with the effect of making the computer unusable. In some instances the attack is reversible (e.g. using an asymmetric key pair with one key used to encrypt the target and the matching paired key to release/decrypt the target), but the attacker is not required to be able to return an attacked system to its previous state (e.g. encrypting a target and not retaining the key to allow decryption). The increasing reliance and acceptance of the end-to-end encryption together with the wider availability of the cryptographic primitives at the end points, increases the risk of malware distribution and attack. This risk may be mitigated by opening and inspecting the nature of the encrypted content. C.4 Coercive control ETSI TR 103 936 [i.17] identifies a large number of concerns regarding the misuse of devices and applications in the context of coercive control within intimate relationships often known as Consumer IoT Enabled abuses or Technology Facilitated Abuse (TFA). TFA behaviours include but are not limited to stalking and omnipresence, surveillance (wiretapping, bugging, videotaping, geolocation tracking, data mining, social media mapping, and the monitoring of data and traffic on the internet), intimidation, impersonation, humiliation, threats, consistent harassment/unwanted contact, sexting, and image-based sexual abuse. Where the attacker is able to use encryption tools this may make it more difficult to identify the source and content of traffic that enables coercive control. It is recognized, and shown in ETSI TR 103 936 [i.17], that many capabilities offered by devices and applications are also legitimate and that not all invocations of them are malicious, rather it may be reasonably suggested that the majority of invocations are benign. The reasonable protection offered by such devices being accessible only over an encrypted connection is, as suggested reasonable. The countermeasure model identified in clause 7.4 of the present document of "{pervasive encryption, integrity, transparency and explicability}" should give assurance that the affected party has knowledge of the affecting party. ETSI ETSI TS 104 103 V1.1.1 (2025-09) 33 Annex D (informative): Application of Security and Privacy Controls D.1 NIST NIST 800-53 [i.23] provides "... a catalog of security and privacy controls for information systems and organizations to protect organizational operations and assets, individuals, other organizations …" from threats and possible attacks. The "consolidated control catalog addresses security and privacy" from two perspectives: a) functionality, which includes control mechanisms; and b) assurance, which includes confidence levels of security and privacy capabilities provided by the control mechanisms. ETSI ETSI TS 104 103 V1.1.1 (2025-09) 34 Annex E (informative): Bibliography • Mbanaso and Cooper: "Conceptual Design of Obligation of Trust Protocol". ETSI ETSI TS 104 103 V1.1.1 (2025-09) 35 History Version Date Status V1.1.1 September 2025 Publication
|
9510b0b4febcc39d76c2a20868042ba3
|
104 102
|
1 Scope
|
The present document defines a set of methods and the overall methodology for incorporating Zero Trust approaches as defined in NIST SP 800-207 [2] into an organization, product or service for the purpose of maximizing the transparency and explicability of the attack surface and to optimize the application of cybersecurity resources to minimize the attack surface. The present document specifies the ZT-Kipling methodology applied to the requirements set out in ETSI TS 104 103 [1] and which addresses the countermeasure framework described in ETSI TS 104 101 [i.8]. NOTE: Whilst the ZT-Kipling methodology and its associated methods can be automated the present document does not directly address how it can be automated and this aspect may be addressed in future standardization
|
9510b0b4febcc39d76c2a20868042ba3
|
104 102
|
2 References
| |
9510b0b4febcc39d76c2a20868042ba3
|
104 102
|
2.1 Normative references
|
References are either specific (identified by date of publication and/or edition number or version number) or non-specific. For specific references, only the cited version applies. For non-specific references, the latest version of the referenced document (including any amendments) applies. Referenced documents which are not found to be publicly available in the expected location might be found in the ETSI docbox. NOTE: While any hyperlinks included in this clause were valid at the time of publication, ETSI cannot guarantee their long-term validity. The following referenced documents are necessary for the application of the present document. [1] ETSI TS 104 103: "Cyber Security (CYBER); Encrypted Traffic Integration (ETI); Problem Statement review and requirements definition". [2] NIST SP 800-207: "Zero Trust Architecture". [3] ETSI TS 102 165-1 (V5.3.1): "Cyber Security (CYBER); Methods and protocols; Part 1: Method and pro forma for Threat, Vulnerability, Risk Analysis (TVRA)".
|
9510b0b4febcc39d76c2a20868042ba3
|
104 102
|
2.2 Informative references
|
References are either specific (identified by date of publication and/or edition number or version number) or non-specific. For specific references, only the cited version applies. For non-specific references, the latest version of the referenced document (including any amendments) applies. NOTE: While any hyperlinks included in this clause were valid at the time of publication, ETSI cannot guarantee their long-term validity. The following referenced documents may be useful in implementing an ETSI deliverable or add to the reader's understanding, but are not required for conformance to the present document. [i.1] ETSI TR 103 305-1: "Cyber Security (CYBER); Critical Security Controls for Effective Cyber Defence; Part 1: The Critical Security Controls". [i.2] Rudyard Kipling's Just So Stories: "The Elephant's child", published in 1902. [i.3] John Kindervag: "No More Chewy Centers: Introducing The Zero Trust Model Of Information Security". [i.4] GSMA™: "FS.37, GTP-U Security". ETSI ETSI TS 104 102 V1.1.1 (2025-09) 6 [i.5] GSMA™: "FS.40, 5G Security Guide", Version 3.0. [i.6] E.M. Hutchins et al.:"Intelligence-Driven Computer Network Defense Informed by Analysis of Adversary Campaigns and Intrusion Kill Chains". [i.7] Recommendation ITU-T X.200: "Information technology - Open Systems Interconnection - Basic Reference Model: The basic model". [i.8] ETSI TS 104 101: "Cyber Security (CYBER); Encrypted Traffic Integration (ETI) Techniques to allow authorized users to identify and access encrypted traffic".
|
9510b0b4febcc39d76c2a20868042ba3
|
104 102
|
3 Definition of terms, symbols and abbreviations
| |
9510b0b4febcc39d76c2a20868042ba3
|
104 102
|
3.1 Terms
|
For the purposes of the present document, the following terms apply: least persistence: means of granting access to an asset for only sufficient time to perform the requested action least privilege: means of granting access to a system asset only to those entities who have a legitimate purpose for access NOTE 1: Thus access to a protected asset is granted to only allow those rights or privileges that are essential to perform the required task. NOTE 2: As defined in NIST SP 800-207 [2].
|
9510b0b4febcc39d76c2a20868042ba3
|
104 102
|
3.2 Symbols
|
Void.
|
9510b0b4febcc39d76c2a20868042ba3
|
104 102
|
3.3 Abbreviations
|
For the purposes of the present document, the following abbreviations apply: 5G NSA 5G Non-Stand Alone 5G SA 5G Stand Alone AV Attack Vector C&C Command and Control CI/CD Continuous Integration/Continuous Delivery CKC Cyber Kill Chain CSC Critical Security Control DoS Denial of Service ETI Encrypted Traffic Integration H-MNO Home Mobile Network Operator IMEI International Mobile Subscriber Identity IMSI International Mobile Equipment Identity IoT Internet of Things IT Information Technology MitM Man in the Middle MNO Mobile Network Operator OSI Open Systems Interconnection PEI Permanent Equipment Identifier PoP Point of Presence RAN Radio Access Network RCE Remote Code Execution SBA Service Based Architecture SUPI Subscription Permanent Identifier TA Threat Agent ETSI ETSI TS 104 102 V1.1.1 (2025-09) 7 TV Threat Vector UE User Edge UPF User Plane Function V-MNO Visitor Mobile Network Operator ZT Zero Trust ZTA Zero Trust Architecture ZT-Kipling Zero Trust Kipling
|
9510b0b4febcc39d76c2a20868042ba3
|
104 102
|
4 Zero Trust security design principles
| |
9510b0b4febcc39d76c2a20868042ba3
|
104 102
|
4.1 Introduction
|
Modern network design, supporting high speed, always-on connectivity with near 100 % availability, has led to a number of paradigms and initiatives that attempt to give assurance of security. These include "secure by design", "secure by default", and, as a stepping stone to Zero Trust (ZT), the principles of "least privilege" and "least persistence". "Never trust, always verify", as originally outlined by John Kindervag [i.3] is the main design principle of ZT, and has been formalized in NIST SP 800-207 [2]. ZT and the Kipling criteria, specified in this document, combine across the entire organization giving transparency and explicability of the security features and making all aspects of the network operation transparent and explicable. That, therefore, reinforces the application of best engineering practice in system provision. Without such attention to detail, the boundary of the system is unknown to its stakeholders and this uncertainty is an opening to the system being attacked. An open, attackable system costs more to maintain, and may lead to over-provisioning that, in turn, further exposes the system to attack. The approach to ZT and the application of the Kipling criteria applies to all aspects of a system, including planning, provisioning, operations, maintenance and security, where security is not optional and is embedded throughout. The Kipling Criteria require that the analyst and designer ask the following questions of each and every system element for each context it is used: What?, Why?, When?, How?, Where?, and Who? NOTE: The Kipling Criteria are so named as they come from a short story by Kipling [i.2] from which the following quote is taken "I Keep six honest serving-men: (They taught me all I knew) Their names are What and Where and When And How and Why and Who". In the context of the present document these 6 questions when asked appropriately of every element, and of how each element is associated to any other element, give a complete picture of the role and purpose of the element. Thus this allows the designer or analyst to be able to demonstrate the validity of the element in the system. In giving an assurance of the security of a network the application of the ZT and the use of the ZT-Kipling methodology ensures that the following principles shall be strictly enforced at every stage of a cyber attack lifecycle: • Minimize the attack surface • Impose a principle of least privilege to allow the use of any asset • Impose a principle of least persistence for the use of any asset The ETI problem statement, in ETSI TS 104 103 [1] suggests that the following steps are taken to address the problem of pervasive encryption thus is consistent with the mandating of the principles above: • transparency and explicability of all elements in the network; • least persistence and least privilege to deploy, access and make use of any element in the network; and • Application of the ZT Kipling methodology as defined in the present document. The ZT-Kipling methodology and its supporting methods defined in the present document shall apply to all elements of a system, and by default, shall include the supply chain. As such, the present document is not a technology to be deployed, but rather a sound approach to the business of effective telecommunications. ETSI ETSI TS 104 102 V1.1.1 (2025-09) 8 Application of the ZT-Kipling methodology, and its associated methods, hereinafter simply referred to as ZT-Kipling, impacts how security technologies, elements, protocols are deployed. In the present document this is extended by application of the Kipling Criteria. ZT-Kipling enforces each stakeholder to answer a small set of questions that result in transparency and explicability of the purpose of the system and all of its functionality. This ensures that the minimum and most effective set of features, including security features, are included in the system, and by default unnecessary ones are removed.
|
9510b0b4febcc39d76c2a20868042ba3
|
104 102
|
4.2 Purpose of ZT in systems
|
As outlined in clause 4.1 above ZT is not a technology, rather it is an approach, formalized in the present document as the ZT-Kipling methodology and supporting methods, to look at systems in order to achieve transparency and explicability of the components or assets of a business system that when combined offer secure services to users. Zero Trust Architecture (ZTA) is based on the assumption that security breaches are inevitable with threat causes inside and outside of a perimeter of a concern, be it an organization, a Data Centre, a service provider infrastructure, or anything else. The methods used to underpin ZT-Kipling are drawn from, and extend, the Critical Security Controls (CSCs) described by ETSI TR 103 305-1 [i.1]. The application of CSC is shown in more depth in clause 5 of the present document but the specific role played by CSC-7 and CSC-10, addressing continuous vulnerability management and malware defences respectively, is outlined below: • CSC-7 (Continuous Vulnerability Management) is to "Develop a plan to continuously access and track vulnerabilities on all enterprise assets within the enterprise's infrastructure, in order to remediate, and minimize, the window of opportunity for attackers. Monitor public and private industry sources for new threat and vulnerability information". • CSC-10 (Malware Defences) is to "Prevent or control the installation, spread, and execution of malicious applications, code, or scripts on enterprise assets".
|
9510b0b4febcc39d76c2a20868042ba3
|
104 102
|
4.3 ZT outline
|
All assets of the system are impacted by ZT-Kipling and are illustrated in Figure 1. Figure 1: Perspective of Zero Trust in Security ETSI ETSI TS 104 102 V1.1.1 (2025-09) 9 To quote from ETSI TS 104 103 [1] "Zero trust … provides a collection of concepts and ideas designed to minimize uncertainty in enforcing accurate, least privilege per-request access decisions in information systems and services in the face of a network viewed as compromised.". For the purposes of the present document the prior definition from ETSI TS 104 103 [1] is refined as follows, ZT is a security strategy (or approach), which is based on no implicit trust (i.e. zero trust) in the digital world, and is designed to detect and prevent breaches, while consistently (or better continuously) verifying all users, all devices, all layers (e.g. OSI layered model Recommendation ITU-T X.200 [i.7]), all applications, across all locations in real time (run-time), and applying continuous integration and continuous delivery (CI/CD) pipeline security, resulting in preventative security from all attack vectors at all stages of the attacks: thus trust becomes explicit. ZT-Kipling consists of five (5) iterative (and recursive) steps in addition to asking the questions of the Kipling Criteria, as Figure 2 illustrates. The steps are repeated continuously for the lifetime of the protected surface. The steps are: 1) Define the protected surface - identify what needs to be protected. 2) Map the transaction flows - how does the traffic flow to, through, and from the protected surface. 3) Build a Zero Trust Architecture (ZTA) - based on the protected surface and the transaction flows, what should ZTA look like? What are its security components and mechanisms? 4) Create Zero Trust security policy - follow Kipling criteria to define the Zero Trust security policy, which adheres to the defined ZTA. 5) Monitor and maintain - maintain and monitor the protected surface. Figure 2: ZT-Kipling methodology The specific actions to be taken for each step are defined in more detail in clause 5. Annex A provides a normative use case of the application of ZT-Kipling. In order to minimize the attack surface and to further develop knowledge of the attack surface the assets of the system or network the security controls defined in ETSI TR 103 305-1 [i.1], in particular CSC-1 and CSC-2, shall be applied. As outlined in ETSI TR 103 305-1 [i.1] CSC-1 is intended to "Actively manage (inventory, track, and correct) all enterprise assets (end-user devices, including portable and mobile; network devices; non-computing/Internet of Things (IoT) devices; and servers) connected to the infrastructure physically, virtually, remotely, and those within cloud environments, to accurately know the totality of assets that need to be monitored and protected within the enterprise. ETSI ETSI TS 104 102 V1.1.1 (2025-09) 10 This will also support identifying unauthorized and unmanaged assets to remove or remediate", and CSC-2 does similarly for software assets as "Actively manage (inventory, track, and correct) all software (operating systems and applications) on the network so that only authorized software is installed and can execute, and that all unauthorized and unmanaged software is found and prevented from installation or execution". As Figure 2 illustrates, Kipling criteria is used in each of the 5 steps leading to the creation of Zero Trust security policies, following Least Privilege and Least Persistence principles.
|
9510b0b4febcc39d76c2a20868042ba3
|
104 102
|
4.4 ZT-Kipling application to achieve Least Privilege principle
|
In building an understanding of the application of ZT-Kipling to the least privilege paradigm the Kipling Criteria apply. In particular, when the use of an asset is determined by multiple criteria (e.g. attribute-based access control) the Kipling criteria provide deep knowledge of the role of an asset and its users, the reasons for the access, and the corresponding behaviours. Drilling into the meaning of each of the questions and what might represent its answer is use case dependent. Table 1 provides an example of how each question might be addressed within a Zero Trust security policy (Step 4) for asset access use case. Table 1: Example of application of Kipling criteria for asset access in Step 4 Question Example for asset access (Least Privilege principle) What What asset(s) are allowed to be accessed by the entity? Why Why is that entity accessing the asset? When When is the asset allowed to be accessed by the entity? How How does the asset know and verify that access is permitted? Where Where is the entity with relation to the asset? Who Who is the entity accessing the asset? 4.5 ZT-Kipling application to achieve Least Persistence principle The security concern of persistent relationships (i.e. still in existence but idle) is that they act as uncontrolled attack surfaces. In maintaining the principle of always minimizing the attack surface, the aim of least persistence is to ensure that the protected surface is always maximized and controlled. Application of the Kipling criteria to the least persistence principle is illustrated in Table 2. Table 2: Example of application of Kipling criteria for asset existence Question Example for asset existence (Least persistence) Comment What What is the asset This is often the semantic or contextual element of an asset's identifier. E.g. border gateway. Why Why is that asset in the system This extends the semantic identifier to address the context in which the asset exists. When When is the asset meant to be available (e.g. is it ephemeral or persistent, if ephemeral how is it invoked and so forth)? The broad assumption should be to minimize the number of persistent elements. How How is the asset operated (e.g. what does it require in order to operate)? Where Where is the asset (logically and geographically)? Who Who owns the asset? This should identify the liability chain including for reporting of any vulnerabilities. ETSI ETSI TS 104 102 V1.1.1 (2025-09) 11
|
9510b0b4febcc39d76c2a20868042ba3
|
104 102
|
5 Applying ZT-Kipling using Critical Security Controls
| |
9510b0b4febcc39d76c2a20868042ba3
|
104 102
|
5.1 Considering Cyber Attack Lifecycle - Cyber Kill Chain
|
The cyber attack lifecycle - also known and referred to in the present document as Cyber Kill Chain (CKC) - is a framework that outlines the stages that a cyber attack typically follows, from initial reconnaissance stage to the final data exfiltration stage (Actions on Objectives). CKC consists of 7 stages, as illustrated in Figure 3 [i.6]. Figure 3: Cyber Kill Chain Incorporating ZT-Kipling into networks' architectures from their inception delivers preventative security posture with the goal to provide proper security measurements as early as possible within the CKC, thereby diminishing the threat landscape. With that, no assumptions shall be made. ZT-Kipling applies for all CKC stages. Annex A provides illustrations of such ZT-Kipling implementations.
|
9510b0b4febcc39d76c2a20868042ba3
|
104 102
|
5.2 Step 1 - Define the Protected Surface
|
The present document identifies means to implement ZT-Kipling using specific Critical Security Controls (CSCs) from ETSI TR 103 305-1 [i.1]. In this case where specific CSCs are identified the present document identifies how they shall be applied in order to satisfy the ZT-Kipling methodology. Step 1 of ZT-Kipling seeks to define or identify the protected surface by determining the attack surface that has to be protected. An analysis of the threats and vulnerabilities should also be carried out using the approach defined in ETSI TS 102 165-1 [3] with ZT-Kipling questions (What, Why, When, How, Where, Who), as illustrated in Figure 2. An attack surface may consist of the following: • Managed assets • Unknown assets • Assets controlled by or maintained by 3rd parties • Ephemeral assets ETSI ETSI TS 104 102 V1.1.1 (2025-09) 12 The building of knowledge of the assets of the system and how they are connected or related shall be addressed by application of the following CSCs: • CSC-1 (Inventory and Control of Enterprise Assets): "Actively manage (inventory, track, and correct) all enterprise assets (end-user devices, including portable and mobile; network devices; non-computing/Internet of Things (IoT) devices; and servers) connected to the infrastructure physically, virtually, remotely, and those within cloud environments, to accurately know the totality of assets that need to be monitored and protected within the enterprise. This will also support identifying unauthorized and unmanaged assets to remove or remediate". • CSC-2 (Inventory and Control of Software Assets): "Actively manage (inventory, track, and correct) all software (operating systems and applications) on the network so that only authorized software is installed and can execute, and that all unauthorized and unmanaged software is found and prevented from installation or execution". • CSC-3 (Data Protection): "Develop processes and technical controls to identify, classify, securely handle, retain, and dispose of data". • CSC-4 (Secure Configuration of Enterprise Assets and Software): "Establish and maintain the secure configuration of enterprise assets (end-user devices, including portable and mobile; network devices, non- computing/IoT devices, and servers) and software (operating systems and applications)." • CSC-5 (Account Management): "Use processes and tools to assign and manage authorization to credentials for user accounts, including administrator accounts, as well as service accounts, to enterprise assets and software". • CSC-6 (Access Control Management): "Use processes and tools to create, assign, manage, and revoke access credentials and privileges for user, administrator, and service accounts for enterprise assets and software". • CSC-7 (Continuous Vulnerability Management): "Develop a plan to continuously access and track vulnerabilities on all enterprise assets within the enterprise's infrastructure, in order to remediate, and minimize, the window of opportunity for attackers. Monitor public and private industry sources for new threat and vulnerability information". • CSC-9 (Email and Web Browser Protections): "Improve protections and detections of threat from email and web vectors, as these are opportunities for attackers to manipulate human behaviour through direct engagement". • CSC-10 (Malware Defences): "Prevent or control the installation, spread, and execution of malicious applications, code, or scripts on enterprise assets". • CSC-12 (Network Infrastructure Management): "Establish, implement, and actively manage (track, report, correct) network devices, in order to prevent attackers from exploiting vulnerable network services and access points". • CSC-15 (Service Provider Management): "Develop a process to evaluate service providers who hold sensitive data, or are responsible for an enterprise's critical IT platforms or processes, to ensure these providers are protecting those platforms and data appropriately". • CSC-16 (Application Software Security): "Manage the security life cycle of in-house developed, hosted, or acquired software to prevent, detect, and remediate security weaknesses before they can impact the enterprise". NOTE: Although the aforementioned CSCs refer to "enterprises", service providers' infrastructures are included into those definitions.
|
9510b0b4febcc39d76c2a20868042ba3
|
104 102
|
5.3 Step 2 - Map the Transaction Flows
|
Mapping the transaction flow step results in an intra-systems, inter-systems, or both flow, which could encompass any number of CSCs, ETSI TR 103 305-1 [i.1], depending on the type of transaction flow. The reflections of which CSCs matter for Step 2 are reflected in Step 1 (define the protected surface), Step 3 (build a Zero Trust Architecture (ZTA)), and Step 4 (create Zero Trust security policy). ZT-Kipling questions (Figure 2) shall be applied accordingly, which normative Annex A elaborates on further. ETSI ETSI TS 104 102 V1.1.1 (2025-09) 13
|
9510b0b4febcc39d76c2a20868042ba3
|
104 102
|
5.4 Step 3 - Build a Zero Trust Architecture
|
Once the protected surface is defined and transaction flows are mapped (Steps 1 and 2, as above) the architecture of the Zero-Trust implementation shall be developed. The CSCs [i.5], corresponding to the protected surface (Step 1) and, following the mapped transaction flows (Step 2), while applying ZT-Kipling questions (Figure 2), shall be used to build the resulting ZTA. Normative Annex A presents an example use case. This step shall consider the protected surface (defined in Step 1), which is the attack surface that needs to be protected. Whilst the threat surface encompasses all the potential threats that can exploit vulnerabilities in the system, an attack surface is a sum of all possible points from which Threat Agents (TAs) can attack: Those points may include various cloud environments, IoTs/OTs/UEs, Internet assets, IT infrastructures, and more. Further, Step 3 shall consider Attack Vectors (AVs) and Threat Vectors (TVs). While the AVs are the methods through which TAs launch attacks, or the "how" of a cyber attack, the TVs include the potential sources and motivations behind them or the "who" and "why" of a cyber attack. ZT-Kipling shall apply for TAs, AVs and TVs identification for any given architecture considering all stages of CKCs.
|
9510b0b4febcc39d76c2a20868042ba3
|
104 102
|
5.5 Step 4 - Create Zero Trust Security Policy (policies)
|
Once ZTA is built (Step 3), ZT security policies shall be created. The CSCs [i.5] corresponding to the protected surface (Step 1), following the mapped transaction flows (Step 2), and resulting ZTA (Step 4), shall be reflected in the security policies. ZT-Kipling questions (Figure 2) shall apply to creation of ZT security policies. Normative Annex A presents an example use case.
|
9510b0b4febcc39d76c2a20868042ba3
|
104 102
|
5.6 Step 5 - Monitor & Maintain
|
Step 5 of ZT-Kipling focuses on monitoring and maintenance of the designed and implemented ZTA and ZT security policies. While addressing the ZT-Kipling questions (Figure 2), the following controls from ETSI TR 103 305-1 [i.1] shall apply: • CSC-8 (Audit Log Management): "Collect, alert, review, and retain audit logs of events that could help detect, understand, or recover from an attack". • CSC-11 (Data Recovery): "Establish and maintain data recovery practices sufficient to restore in-scope enterprise assets to a pre-included and trusted state". • CSC-12 (Network Infrastructure Management): "Establish, implement, and actively manage (track, report, correct) network devices, in order to prevent attackers from exploiting vulnerable network services and access points." • CSC-13 (Network Monitoring and Defence): "Operate processes and tooling to establish and maintain comprehensive network monitoring and defence against security threats across the enterprise's network infrastructure and uses base". • CSC-14 (Security Awareness and Skills Training): "Establish and maintain a security awareness program to influence behaviour among the workforce to be security conscious and properly skilled to reduce cybersecurity risks to the enterprise". • CSC-15 (Service Provider Management): "Develop a process to evaluate service providers who hold sensitive data, or are responsible for an enterprise's critical IT platforms or processes, to ensure these providers are protecting those platforms and data appropriately." • CSC-16 (Application Software Security): "Manage the security life cycle of in-house developed, hosted, or acquired software to prevent, detect, and remediate security weaknesses before they can impact the enterprise." ETSI ETSI TS 104 102 V1.1.1 (2025-09) 14 • CSC-17 (Incident Response Systems: "Establish a program to develop and maintain an incident response capability (e.g. policies, plans, procedures, defined roles, training and communications) to prepare, detect, and quickly respond to an attack". • CSC-18 (Penetration Testing): "Test the effectiveness and resiliency of enterprise assets through identifying and exploiting weaknesses in controls (people, processes, and technology), and simulating the objectives and actions of an attacker". ETSI ETSI TS 104 102 V1.1.1 (2025-09) 15 Annex A (normative): 5G roaming use case for application of ZT-Kipling A.1 Overview Figure A.1 illustrates a use case, which is used to explain the application of ZT-Kipling steps. A high level schema of 5G Non-Stand Alone (NSA) and 5G Stand Alone (SA) infrastructures are depicted, where a roaming 5G SA User Edge (UE), attached to the 5G NSA - Roaming UE, is required to connect to its Destination. Figure A.1: Use Case - 5G SA and 5G NSA Infrastructures with Roaming UEs In relation of ZT-Kipling to 5G roaming use case, ZT is a cybersecurity paradigm focused on: • real time continuous monitoring /visibility for cyber risks, threats and vulnerabilities; • least privileged security policies enforcement to protect from cyber risks, threats, and vulnerabilities; • 5G User-ID (SUPI) and 5G Equipment/Device-ID (PEI) security policies granularity for both monitoring and enforcement; • security across all layers of mobile networks: Application, Signalling, Data, and Management; • security across all exposed locations: Roaming, RAN, Open-RAN, N6/SGi, APIs; • security against all attack vectors: C&C, RCE, botnets, malware, MitM, fraudulent IDs, identity, Ransomware, DoS; and • security across all software lifecycle stages: runtime, CI/CD (shift left), DevOps. NOTE: The aforementioned list applies to all use cases of 5G networks. A.2 Step 1 - Define the Protected Surface Using Table A.1, ZT-Kipling questions apply to define the protected surface. ETSI ETSI TS 104 102 V1.1.1 (2025-09) 16 Table A.1: ZT-Kipling Questions Applied to Step 1 - Define the Protected Surface What What assets [applications, devices, etc.] can the Roaming UE access? Why Why is that roaming UE accessing those assets (Destination)? When When can the roaming UE access Destination? How How does the Destination know and verify that Roaming UE access is permitted? Where Where is the Roaming UE in relation to the Destination? Who Who [what is its IMEI, IMSI, user] is the Roaming UE accessing the Destination? The protected surface for the UE (depicted in Figures A.1) that is roaming through the visitor's 5G NSA network, roaming exchange, its home 5G SA network, and continuing to transit through Internet to Destination, is the sum of the following elements: 1) roaming UE; 2) UE - V-MNO (5G NSA) transit; 3) V-MNO (5G NSA) Core; 4) V-MNO (5G NSA) Core - Roaming Exchange - H-MNO (5G SA) Core transit; 5) H-MNO (5G SA), including Service Based Architecture (SBA) and User Plane Function (UPF), Core; 6) H-MNO 5G SA Core - Internet - Destination transit; and 7) destination. Table A.2 summarizes the applicability of CSCs from ETSI TR 103 305-1 [i.1] for each identified protected surface element. Table A.2: Step 1 - Define the Protected Surface & CSCs Protected Surface Element CSC-# 1. Roaming UE 1, 2, 3, 4, 5, 6, 9, 10,16 2. UE - V-MNO (5G NSA) transit 1, 2, 3, 4, 5, 6, 7, 10, 12 3. V-MNO (5G NSA) Core 1, 2, 3, 4, 5, 6, 7, 10, 12, 15, 16 4. V-MNO (5G NSA) Core - Roaming Exchange - H-MNO (5G SA) Core transit 1, 2, 3, 4, 5, 6, 7, 10, 12 5. H-MNO (5G SA) Core 1, 2, 3, 4, 5, 6, 7, 10, 12, 15, 16 6. H-MNO (5G SA) Core - Internet - Destination transit 1, 2, 3, 4, 5, 6, 7, 10, 12 7. Destination 1, 2, 3, 4, 5, 6, 9, 10, 16 A.3 Step 2 - Map the Transaction Flows Figure A.2 illustrates the transaction flow from the Roaming UE to Destination. Although the figure illustrates one- directional flow, in most cases, the flows are bi-directional, depending on the application used. ETSI ETSI TS 104 102 V1.1.1 (2025-09) 17 Figure A.2: 5G Roaming Use Case. Step 2 - Map the Transaction Flows Although the illustration of the transaction flow in the present document is a high-level view, a deeper perspective is highly recommended. It is like peeling an onion - starting with a high-level architecture and then, digging down to the next level, all the way to physical interfaces and ports, mapping the flows from one element to another one. Therefore, responses to the ZT-Kipling questions might be slightly different, based on the level of the transaction flow considered. For example, Table A.3 illustrates the responses to the ZT-Kipling questions for a high level of transaction flow, while Table A.4 illustrates the responses to the questions for a lower level of the transaction flow details. The aforementioned tables provide a good illustration of how the responses to ZT-Kipling questions might differ. Table A.3: ZT-Kipling Questions Applied to Step 2 - Map the Transaction Flows, high level What What service providers will the Roaming UE use to reach the Destination? Why Why will the Roaming UE use those service providers? When When can the Roaming UE use those service providers ? How How will the transiting traffic traverse through all service providers? Where Where from the Roaming UE can connect to the Destination? Who Who [what is its IMEI, IMSI, user] is the Roaming UE accessing the Destination? Table A.4: ZT-Kipling Questions Applied to Step 2 - Map the Transaction Flows, lower level What What Point of Presence (PoP) of the V-MNO will be used to attach to the network? Why Why will the Roaming UE use that service provider (V-MNO)? When When can the Roaming UE use the V-MNO? How How will the V-MNO and H-MNO verify the Roaming UE? Where Where can the Roaming UE access the V-MNO from? Who Who [what is its IMEI, IMSI, user] is the Roaming UE accessing the Destination? A.4 Step 3 - Build a Zero Trust Architecture Once the protected surface is defined and transaction flows are mapped (Steps 1 and 2, respectively, are completed), ZTA (Step 3) can be built. Before any security architecture can be considered, it is important to understand AVs, TVs and TAs related to all the elements are within the defined protected surface and the mapped transaction flows are valid [2]. Application of ZT-Kipling identifies AVs, TVs, and TAs for all stages of CKCs within the architecture, as illustrated in Table A.5. Further, the CSCs identified in TR 103 305-1 [i.1], in Steps 1 and 2, shall apply by applying the ZT-Kipling method as defined in the present document. ETSI ETSI TS 104 102 V1.1.1 (2025-09) 18 Table A.5: ZT-Kipling Questions Applied to Step 3 Comment What What are the TAs for the defined protected surface? Identify AVs & TVs Why Why could the TAs attack the defined protected surface? Identify TVs When When could the TAs attack the defined protected surface? Identify TVs How How could identified TAs launch attacks on the defined protected surface Identify AVs & TVs Where Where from TAs could attack the defined protected surface? Identify AVs & TVs Who Who are the possible TAs? Identify AVs & TVs Depending on the CKC stage, the TAs could camouflage, using the penetrated elements, which need to be protected in the first place, as extensions of themselves, resulting in the increase of TA attack surface and its impact. Figure A.3 illustrates the seven (7) TAs and AVs for the Roaming UE use case discussed in the present document. Considering all stages of CKC, the TAs and the AVs for the use case are: • TA1 & AV1 - Roaming UE; • TA1 & AV2 - RAN; • TA3 & AV3 - V-MNO, 5G NSA Infrastructure; • TA4 & AV4 - Roaming Exchange; • TA5 & AV5 - H-MNO, 5G SA Infrastructure; • TA6 & AV6 - Internet; and • TA7 & AV7 - Destination. Figure A.3: 5G Roaming Use Case, TAs and AVs The aim of ZTA is to deliver the architecture, which provides defences from any TA and AV at any CKC stage. To achieve this, ZTA includes the following methodologies and technologies: • Identity and Access Management; • Devices; • Networks; • Micro-segmentation; ETSI ETSI TS 104 102 V1.1.1 (2025-09) 19 • Encryption; • Applications and Workloads Validation; • Data; and • Visibility and Analytics. ZTA has three (3) main elements: • Users; • Applications; and • Infrastructure. Applying the aforementioned ZTA methodologies and technologies to those main three elements, while addressing ZT- Kipling, results in a ZTA. Table A.6 provides a cross-reference between the TAs and AVs for the corresponding Protected Surface Elements, the corresponding ZTA Elements, and ZTA methodologies and technologies, which shall be included into ZTA for the discussed Roaming UE use case. Further, applicable CSCs [i.1], as identified in clause 5.2 of the present document, are depicted in Table A.6. Table A.6: Step 3 - ZTA Methodologies & Technologies to be Applied Against Identified TAs TA & AV CSC-# Protected Surface Element ZTA Element ZTA Methodologies & Technologies to be Included 1, 2, 3, 4, 5, 6, 7 1, 2, 3, 4, 5, 6, 9, 10, 16 1. Roaming UE Users Applications Apply principles of least privilege access and identity. Verify the Roaming UE integrity, including IMSI/SUPI, IMEI/PEI validation. Validate the UE data integrity through visibility and analytics [i.4]. 1, 2, 3 1, 2, 3, 4, 5, 6, 7, 10, 12 2. UE - V-MNO (5G NSA) transit Infrastructure Micro-segment networks and possible workloads Devices' management [includes RAN elements, routers, switches, mobile core elements facing RAN, UE] Encrypt traffic transiting through air or in RAN sharing use cases. 1, 2, 3, 4, 5, 6, 7 1, 2, 3, 4, 5, 6, 7, 10, 12, 15, 16 3. V-MNO (5G NSA) Core Infrastructure Applications (related to the core) Micro-segment networks and possibly workloads Devices' management [includes routers, switches, mobile core elements and functions, Operations Support Systems (OSS)/Business Support Systems (BSS) systems, UE] Validate core elements' workloads continuously (CI/CD) Validate UE device integrity, including IMSI/SUPI, IMEI/PEI Visibility and analytics of data integrity [i.4]. 3, 4, 5 1, 2, 3, 4, 5, 6, 7, 10, 12 4. V-MNO (5G NSA) Core - Roaming Exchange - H-MNO (5G SA) Core transit Infrastructure Micro-segment networks and possibly workloads Devices' management [includes routers, switches, mobile core elements and functions] Encrypt traffic transiting through untrusted roaming exchanges. 1, 2, 3, 4, 5, 6, 7 1, 2, 3, 4, 5, 6, 7, 10, 12, 15, 16 5. H-MNO (5G SA) Core Infrastructure Applications (related to core) Micro-segment networks and possibly workloads Devices' management [includes routers, switches, mobile core elements and functions, Operations Support Systems (OSS)/Business Support Systems (BSS) systems, UE] Validate core elements' workloads continuously Validate UE device integrity, including IMSI/SUPI, IMEI/PEI Visibility and analytics of data integrity [i.4]. ETSI ETSI TS 104 102 V1.1.1 (2025-09) 20 TA & AV CSC-# Protected Surface Element ZTA Element ZTA Methodologies & Technologies to be Included 1, 5, 6, 7 1, 2, 3, 4, 5, 6, 7, 10, 12 6. H-MNO (5G SA) Core - Internet - Destination transit Infrastructure Micro-segment networks and possibly workloads Devices' management [includes routers, switches, mobile core elements and functions] Validate the Destination integrity Encrypt traffic transiting through untrusted domains. 1, 6, 7 1, 2, 3, 4, 5, 6, 9, 10, 16 7. Destination Users Applications Apply principles of least privilege access and identity Verify Destination integrity Validate the device integrity, including IMSI/SUPI, IMEI/PEI Validate data integrity through visibility and analytics [i.4]. Figure A.4 illustrates the resulting ZTA for the discussed Roaming UE use case, identifying the positioning of security elements following the ZT principle of securing across all layers of mobile networks (Application, Signalling, Data, and Management), exposed locations (Roaming, RAN, Open-RAN, N6/SGi, APIs), and securing against all TAs and AVs, as identified in Figure A.3 and Table A.6. ZTA includes security control checks along the transaction flow path through the identified protected surface elements. The protection mechanisms and technologies include strict security policies, which Step 4 of ZT-Kipling addresses. Figure A.4: Use Case. Step 3 - Build ZTA A.5 Step 4 - Create Zero Trust Security Policy (policies) Once the protected surface (Step 1) is defined, roaming transaction flow (Step 2) is mapped, and ZTA is built (Step 3), ZT security policies are created using ZT-Kipling. Table A.7 provides the applicable ZT-Kipling questions to be used for roaming ZT security policies use case. ETSI ETSI TS 104 102 V1.1.1 (2025-09) 21 Table A.7: ZT-Kipling questions applies to Step 4 - Create ZT Security Policy What What applications are being used to traverse the networks? Identify and validate signalling and user plane applications. Why Why is the packet accessing a resource? Identify legitimate flows for signalling & user planes. When When is the resource being accessed? Predictable signalling/user planes traffic behaviours. How How does the packet access the protected surface throughout the communications? Visibility into signalling and user planes. Where Where is the packet source and destination? Specify the source/destination. Who Who should be connected to this flow? Validate unique mobile user IDs - IMSIs/SUPIs, IMEIs/PEIs. As ZT security policies apply to every security element along the transaction flow, the responses to ZT-Kipling questions might vary, depending on the security element's position within ZTA. Further, every security element along the transaction flow path shall address the CSCs corresponding to the protected surface element, which was identified in Step 1 of ZT-Kipling (clause 5.2 of the present document) and reflected in Tables A.2 and A.6. A.6 Step 5 - Monitor & Maintain During this step, roaming network monitoring and maintenance shall take place contiguously. CSCs applicable to this step are: 8, 11, 12, 13, 16, 17, and 18, as identified in clause 5.6 of the present document. Table A.8 provides the applicable questions for this ZT-Kipling step 5 for the Roaming use case. Table A.8: ZT-Kipling Questions Applied to Step 5 - Monitor & Maintain What What applications are traversing the networks? Identify and validate all transiting applications without relying on port numbers, as those may be spoofed. Any changes? Why Why is specific traffic transiting? Is it legitimate? When When is the resource being accessed? Does it follow the pattern? What are the differences? Any changes? How How does the packet access the protected surface throughout the communications? Any changes? Where Where is the packet sourced from? Was it validated? Any changes? Who Who is accessing the assets? Are the unique mobile user IDs - IMSIs/SUPIs, IMEIs/PEIs - validated? Depending on the monitored roaming traffic, customers' coverage, involved MNOs' and roaming partners changes, Step 5 shall lead to step 1 recursively, as illustrated in Figure 2. ETSI ETSI TS 104 102 V1.1.1 (2025-09) 22 Annex B (informative): Bibliography • GSA ZTA 3.1: "Zero Trust Architecture (ZTA) Buyer's Guide", Version 3.1. • CISA ZTMM 2.0: "Zero Trust Maturity Model", Version 2. • ETSI TR 103 644 (V1.2.1): "Cyber; Observations from the SUCCESS project regarding smart meter security". ETSI ETSI TS 104 102 V1.1.1 (2025-09) 23 History Document history V1.1.1 September 2025 Publication
|
8e90692f70e1e9a0f58f164e7ad90f06
|
104 047-2
|
1 Scope
|
The present document is one of the parts of the radio conformance test specifications of the DECT-2020 New Radio (NR) radio device. The present document specifies radio device protocol conformance testing including Medium Access (MAC), Data Link Control (DLC) and Convergence Layer (CVG) protocol layers. Further the present document defines the test conditions, test configurations, and requirement for testing functions and a test system, to be used for protocol conformance testing.
|
8e90692f70e1e9a0f58f164e7ad90f06
|
104 047-2
|
2 References
| |
8e90692f70e1e9a0f58f164e7ad90f06
|
104 047-2
|
2.1 Normative references
|
References are either specific (identified by date of publication and/or edition number or version number) or non-specific. For specific references, only the cited version applies. For non-specific references, the latest version of the referenced document (including any amendments) applies. Referenced documents which are not found to be publicly available in the expected location might be found in the ETSI docbox. NOTE: While any hyperlinks included in this clause were valid at the time of publication, ETSI cannot guarantee their long-term validity. The following referenced documents are necessary for the application of the present document. [1] ETSI TS 103 636-1: "DECT-2020 New Radio (NR); Part 1: Overview". [2] ETSI TS 103 636-2: "DECT-2020 New Radio (NR); Part 2: Radio reception and transmission requirements". [3] ETSI TS 103 636-3: "DECT-2020 New Radio (NR); Part 3: Physical layer". [4] ETSI TS 103 636-4: "DECT-2020 New Radio (NR); Part 4: MAC layer". [5] ETSI TS 103 636-5: "DECT-2020 New Radio (NR); Part 5: DLC and Convergence layers". [6] ETSI TS 103 874-1: "DECT-2020 New Radio (NR); Access Profile; Part 1: Overview". [7] ETSI TS 103 874-2: "DECT-2020 New Radio (NR); Access Profile Part 2; Smart Metering, City and Buildings". [8] ETSI TS 103 874-3: "DECT-2020 New Radio (NR); Application Profile; Part 3: IPv6 Profile". [9] ETSI TS 104 047-1: "DECT-2020 New Radio (NR); Conformance Specification; Part 1: Radio Transmission and Reception".
|
8e90692f70e1e9a0f58f164e7ad90f06
|
104 047-2
|
2.2 Informative references
|
References are either specific (identified by date of publication and/or edition number or version number) or non-specific. For specific references, only the cited version applies. For non-specific references, the latest version of the referenced document (including any amendments) applies. NOTE: While any hyperlinks included in this clause were valid at the time of publication, ETSI cannot guarantee their long-term validity. ETSI ETSI TS 104 047-2 V1.1.1 (2025-10) 11 The following referenced documents may be useful in implementing an ETSI deliverable or add to the reader's understanding, but are not required for conformance to the present document. [i.1] ETSI ES 202 553: "Methods for Testing and Specification (MTS); TPLan: A notation for expressing Test Purposes".
|
8e90692f70e1e9a0f58f164e7ad90f06
|
104 047-2
|
3 Definition of terms, symbols and abbreviations
| |
8e90692f70e1e9a0f58f164e7ad90f06
|
104 047-2
|
3.1 Terms
|
For the purposes of the present document, the terms given in ETSI TS 103 636-4 [4], ETSI TS 103 636-5 [5] and the following apply: companion device: functionality of a lower tester implemented within a real test device FT-PT: RD operating in both FT and PT modes Lower Tester FT (LT_FT): tester entity operating strictly in FT mode, providing indirect control and observation of the EUT via the DECT-2020 NR radio interface Lower Tester FT-PT (LT_FT_PT): tester entity operating in both FT and PT modes, providing indirect control and observation of the EUT via the DECT-2020 NR radio interface Lower Tester PT (LT_PT): tester entity operating strictly in PT mode, providing indirect control and observation of the EUT via the DECT-2020 NR radio interface Lower Tester Test Component (LT_TC): tester component within the test system used to observe events in both directions during testing reference time accuracy: As defined in ETSI TS 103 636-2 [2].
|
8e90692f70e1e9a0f58f164e7ad90f06
|
104 047-2
|
3.2 Symbols
|
For the purposes of the present document, the symbols given in ETSI TS 103 636-4 [4] apply.
|
8e90692f70e1e9a0f58f164e7ad90f06
|
104 047-2
|
3.3 Abbreviations
|
For the purposes of the present document, the abbreviations given in ETSI TS 103 636-1 [1], ETSI TS103 636-4 [4], ETSI TS103 636-5 [5] and the following apply: NOTE: An abbreviation defined in the present document takes precedence over the definition of the same abbreviation, if any, in ETSI TS 103 636-1 [1], ETSI TS 103 636-4 [4] or ETSI TS 103 636-5 [5]. ATM Abstract Test Method EUT Equipment Under Test IUT Implementation Under Test PICS Protocol Implementation Conformance Statement PIXIT Protocol Implementation eXtra Information for Testing TD Test Description TP Test Purpose TPLan Test Purpose Language TS Test System TSS Test Suite Structure ETSI ETSI TS 104 047-2 V1.1.1 (2025-10) 12
|
8e90692f70e1e9a0f58f164e7ad90f06
|
104 047-2
|
4 Overview
|
4.1 Handling of conformance requirements in different releases of the core specification The core specifications, ETSI TS 103 636 series, and application specific profiles, ETSI TS 103 874 series, consist of multiple releases, where each release introduces new functionalities, procedures and configurations, that are then carried to the future releases. Therefore, each function or procedure is tested according to the specification release when introduced first time in the corresponding specifications. In case a function or procedure is updated or even removed from a future release the impact to the test purpose(s) and/or test case(s) is indicated in corresponding test purpose(s) and test case(s) clauses. Therefore, when differences between conformance requirements in different releases of the core specifications have an impact on the Pre-test conditions, Test procedure sequence or/and the Specific message contents, the Conformance requirements related to different releases are specified separately in the given test and indicated separately per release. Further, when there is no Release indicated for a conformance requirement text, this should be understood that the conformance requirements in the latest version of the specification of the release are valid and identical to the requirements when the feature was first introduced to the core specifications. The declaration by the apparatus supplier i.e. the Protocol Implementation Conformance Statement (PICS) is used to determine the release of the RD. The PICS pro forma is provided in Annex B. The used configuration for test cases is defined in the supported application profile specifications as defined ETSI TS 103 874-1 [6] and Protocol Implementation Conformance Statement (PICS). In case device does not support any application profile the device vendor should declare a test configuration based on valid parameter values of the core specifications.
|
8e90692f70e1e9a0f58f164e7ad90f06
|
104 047-2
|
4.2 Testing of optional functions and procedures
|
Any function or procedure which is optional, as indicated in the present document, may be subject to a conformance test if it is implemented in the RD. The implementation of the certain optional feature can be determined based on RD's support on given application profile specification as defined ETSI TS 103 874-1 [6], and by a declaration by the apparatus supplier. The declaration by the apparatus supplier i.e. the Protocol Implementation Conformance Statement (PICS) is used to determine whether an optional function/procedure has been implemented. The PICS pro forma is provided in Annex B.
|
8e90692f70e1e9a0f58f164e7ad90f06
|
104 047-2
|
4.3 Implicit testing
|
For some DECT-2020 NR signalling and protocol features conformance is not verified explicitly in the present document. This does not imply that correct functioning of these features is not essential, but that these are implicitly tested to a sufficient degree in other tests.
|
8e90692f70e1e9a0f58f164e7ad90f06
|
104 047-2
|
4.4 Repetition of testing
|
As a general rule, the test cases specified in the present document are highly reproducible and do not need to be repeated unless otherwise stated. However, the rate of correct RD behaviour can be specified statistically, e.g. "at least 90 %", which is defined separately in the specific test case. Additionally, in some of the test cases, HARQ retransmissions are not tolerated, because of characteristics of the test case. For tests that expect the RD behaviour described above and are not fully (100 %) reproducible, if a RD fails the test on its first attempt and the tester suspects the reason is due to the statistical nature of the test, the UE should be re-tested one or more times to confirm the result. When an EUT supports both FT and PT modes, any test which is applicable for both modes may be tested using only either one of the configurations. ETSI ETSI TS 104 047-2 V1.1.1 (2025-10) 13 5 Reference conditions, generic and test procedures, and test parameters 5.1 Reference conditions The reference environmental conditions used by all signalling and protocol tests are as specified in ETSI TS 103 636-2 [2], Annex C or according to the conditions existing at the intended use of the RD declared by the device manufacturer. The generic test architectures are defined in Annex A, and where a test requires an environment or configuration that are different, these will be specified in the test itself. 5.2 Generic and test procedures A set of basic generic procedures for bringing the IUT into a specific initial state, as well as test procedures on comprising of well-defined actions after the IUT enters the specific state are specified in clause 8.4. These procedures are used in numerous test cases throughout the present document.
|
8e90692f70e1e9a0f58f164e7ad90f06
|
104 047-2
|
6 Test Suite Structure and Test Purposes
| |
8e90692f70e1e9a0f58f164e7ad90f06
|
104 047-2
|
6.1 Structure for DECT-2020 NR tests
|
The test grouping is organized according to the structure of the protocols defined in the core specifications ETSI TS 103 636-4 [4] and ETSI TS 103 636-5 [5]. Moreover, test purposes are identified and categorized by a group, subgroup and sequential two-digits number (uniquely assigned upon definition of each test purpose). Table 6.1-1 shows the DECT-2020 NR Test Suite Structure (TSS) defined for conformance testing. ETSI ETSI TS 104 047-2 V1.1.1 (2025-10) 14 Table 6.1-1: Test Suite Structure (TSS) Root Group Sub-group MAC Spectrum management Beaconing transmissions Operating Channel(s) and Subslot(s) selection Power Control Selecting RD for association Broadcast Procedure Random Access Procedure Random Access Transmission Random Access Reception Scheduled Access data transfer HARQ Procedures HARQ Transmission HARQ Feedback Multiplexing and Assembly Mobility Procedures Association Procedure Association Request Association Response Association Release Security Procedures Paging Procedures DLC DLC Retransmissions (ARQ) DLC SDU Lifetime control Routing Services Uplink hop-by-hop routing Downlink flooding CVG Transparent Procedures Multiplexing Procedures Duplicate removal Procedure CVG Segmentation
|
8e90692f70e1e9a0f58f164e7ad90f06
|
104 047-2
|
6.2 Test Purpose definition conventions
| |
8e90692f70e1e9a0f58f164e7ad90f06
|
104 047-2
|
6.2.1 General
|
The TPs are defined according to Table 6.2.1-1. Table 6.2.1-1: Test Purpose definition rules Field Description TP id The TP id is a unique identifier. It shall be specified according to the TP naming conventions defined in the present document. Test objective The test objective gives a brief description of the test to indicate which conformance requirement is intended to be tested in the test purpose. Reference The reference field of the TP header indicates where the tested requirement is expressed. The field may refer to several clauses. Mode The field indicates to which operating mode(s) of the IUT the test purpose may be applied to. The value of this field is either FT, PT, FT-PT, or a set of the preceding. Config The Config field indicates the test system configuration required for testing the current TP. PICS Selection The PICS selection row contains a Boolean expression which may be used to indicate whether the test purpose is applicable according to a completed PICS. An empty row implicitly indicates that the test purpose is applicable for every implementation. Initial conditions The initial conditions define in which initial state the IUT shall be to apply the actual TP. In the corresponding Test Case, when the execution of the initial condition fails, it leads to the assignment of an Inconclusive verdict. NOTE 1: The initial condition is written using the TPLan notation and ETSI style for Programming Language (PL). Expected behaviour The expected behaviour defines the events which are parts of the TP objective, and according to which the IUT are expected to perform in order to conform to the base specification. In the corresponding Test Case, Pass or Fail verdicts can be assigned here. NOTE 2: The expected behaviour is written using the TPLan notation and ETSI style for Programming Language (PL). ETSI ETSI TS 104 047-2 V1.1.1 (2025-10) 15
|
8e90692f70e1e9a0f58f164e7ad90f06
|
104 047-2
|
6.2.2 TP Identifier naming conventions
|
The identifier of a TP is built according to Table 6.2.2-1. Table 6.2.2-1: TP naming convention TP/<root>/<pg>/<gr>/<sgr>/<nn> Abbreviation Description <root> = root DECTNR DECT-2020 NR TP/MAC/<gr>/<sgr>/<nn> MAC Medium Access Control TP/MAC/SPM/<sgr>/<nn> SPM Spectrum Management Procedures TP/MAC/SPM/BEA/<nn> BEA Beaconing Transmissions TP/MAC/SPM/OCS/<nn> OCS Operating Channel(s) and Subslot(s) selection TP/MAC/SPM/RDS/<nn> RDS Selecting RD for association TP/MAC/SPM/PC/<nn> PC Power Control TP/MAC/BC/<nn> BC Broadcast Procedure TP/MAC/RACH/<sgr>/<nn> RACH Random Access Procedures TP/MAC/RACH/TX/<nn> TX Random Access Transmission TP/MAC/RACH/RX/<nn> RX Random Access Reception TP/MAC/HARQ/<sgr>/<nn> HARQ HARQ Operation TP/MAC/HARQ/TX/<nn> TX HARQ Transmission TP/MAC/HARQ/FB/<nn> FB HARQ Feedback TP/MAC/MUXA/<nn> MUXA Multiplexing and Assembly TP/MAC/MOB/<nn> MOB Mobility Procedures TP/MAC/ASS/<sgr>/<nn> ASS Association Procedure TP/MAC/ASS/REQ/<nn> REQ Association Request TP/MAC/ASS/RES/<nn> RES Association Response TP/MAC/ASS/REL/<nn> REL Association Release TP/MAC/SEC/<nn> SEC Security Procedures TP/MAC/PG/<nn> PG Paging Procedures TP/DLC/<gr>/<sgr>/<nn> DLC Data Link Control TP/DLC/SEG/<nn> SEG DLC Segmentation TP/DLC/REA/<nn> REA DLC Reassembly TP/DLC/RETX/<nn> RETX DLC Retransmission TP/DLC/LFT/<nn> LFT DLC SDU lifetime control TP/DLC/ROU/<nn> ROU Routing Services TP/CVG/<gr>/<sgr>/<nn> CVG Convergence TP/CVG/MUX/<sgr>/<nn> MUX Multiplexing TP/CVG/MUX/EP/<nn> EP Endpoint Multiplexing TP/CVG/MUX/MT/<nn> MT Mux Tag Multiplexing TP/CVG/DUP/<nn> DUP Duplicate Removal TP/CVG/SEG/<nn> SEG CVG Segmentation <nn> = sequential number 01 to 99
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.