November 2020 doc IEEE 802 11 201770 r

November 2020 doc. : IEEE 802. 11 -20/1770 r 6 RCM and EDP PAR and CSD comment responses Date: 2020 -11 -05 Authors: Submission Slide 1 Stephen Mc. Cann, Huawei

November 2020 doc. : IEEE 802. 11 -20/1770 r 6 Abstract This document is a composite of all IEEE 802 WG comments and IEEE 802. 11 RCM SG responses on the RCM (802. 11 bh) and EDP (802. 11 bi) PARs and CSDs. Submission Slide 2 Stephen Mc. Cann, Huawei

November 2020 doc. : IEEE 802. 11 -20/1770 r 6 IEEE 802. 11 bh PAR Submission Slide 3 Stephen Mc. Cann, Huawei

November 2020 doc. : IEEE 802. 11 -20/1770 r 6 802. 1 comments on PARs and CSDs Prelude to P 802. 11 bh and P 802. 11 bi comments 802. 1 appreciates the clear intent (as captured in the Scope and Need of the PARs for both proposed projects) to facilitate improvements in user privacy. 802. 1 also appreciates the clear indication of the difficulties (already discussed) that both projects will face in improving user privacy, and in particular the possibility of unacceptable service disruption or limitation when an STA’s MAC address is changed. Response: Thanks to IEEE 802. 1 for their comments Submission Slide 4 Stephen Mc. Cann, Huawei

November 2020 doc. : IEEE 802. 11 -20/1770 r 6 802. 1 comments on PARs and CSDs Both P 802. 11 bh and P 802. 11 bi CSDs CSD Templates • The CSD templates used appear to be out of date, as they refer to “Balanced costs (infrastructure versus attached stations)”. • Use the current CSD template that can be retrieved from https: //www. ieee 802. org/devdocs. shtml. Response: The CSD documents have been updated using the latest template. Submission Slide 5 Stephen Mc. Cann, Huawei

November 2020 doc. : IEEE 802. 11 -20/1770 r 6 802. 1 comments on PARs and CSDs P 802. 11 bh PAR 2. 1 Project Title: • The project title indicates a change in MAC address would be only by randomization. The scope discusses changing addresses. Additionally, this amendment does not enhance the service, but rather preserves the service while addressing privacy concerns with MAC addresses. • Change the title of the amendment to, “IEEE Standard for Information technology Telecommunications and information exchange between systems Local and metropolitan area networks Specific requirements Part 11: Wireless LAN Medium Access Control (MAC) and Physical Layer (PHY) Specifications Amendment: Service preservation with randomized or changing MAC addresses” Response: The title has been changed to: “Operation with Randomized and Changing MAC Addresses” Submission Slide 6 Stephen Mc. Cann, Huawei

November 2020 doc. : IEEE 802. 11 -20/1770 r 6 802. 3 comments on PARs and CSDs P 802. 11 bh PAR General • Neither the PARs nor CSDs clearly distinguish the need for two similar projects. Scopes overlap, and distinct identity is not established between proposed P 802. 11 bh and proposed P 802. 11 bi. • We appreciate the submittal is a. pdf, but the file name being a docx extension caused some problems in opening. Response: Thanks to IEEE 802. 3 for their comments The 802. 11 bh PAR amends the standard to address requirements for managing 802. 11 non. AP STAs that change their MAC addresses. No changes required. Submission Slide 7 Stephen Mc. Cann, Huawei

November 2020 doc. : IEEE 802. 11 -20/1770 r 6 802. 1 comments on PARs and CSDs P 802. 11 bh PAR 5. 2. b. Scope of the project: • 802. 1’s interpretation of the Scope of both PARs is that incorporation of additional information elements in 802. 11 that could further diminish user privacy (to, for example, facilitate continued user identification in the absence of a permanent STA to MAC address association) is Out of Scope. • Please amend the Scope to state explicitly what new user tracking capabilities are within or excluded from the Scope of the PAR. Response: Note that 5. 2. b. states “…without decreasing user privacy…”. An additional explanatory note with respect to 5. 2. b to clarify that the intention is to maintain the IEEE P 802. 11 REVmd D 5. 0 requirements regarding changing a MAC address, e. g. not during an ESS connection, has been added to section 8. 1. Submission Slide 8 Stephen Mc. Cann, Huawei

November 2020 doc. : IEEE 802. 11 -20/1770 r 6 802. 1 comments on PARs and CSDs P 802. 11 bh PAR 5. 2. b. Scope of the project 5. 5 Need for the Project: • 802. 1 is concerned that both Scope and Need of P 802. 11 bh appear to explicitly limit consideration of disruption, and the need for session continuity, to 802. 11 mechanisms. Important 802. 1 Q TSN capabilities, such as bandwidth allocation, depend on flow identification that span both the wired and wireless networks. [Our comments on the CSDs point out that Technical Feasibility appears to be based, at least in part, on erroneous information about the scope of 802. 1 in general and 802. 1 Q in particular]. • Please amend the Scope to explicitly indicate that the mechanisms to enable session continuity recognize the fact that sessions extend to the LAN beyond the medium supported by 802. 11. Response: An additional explanatory note with respect to 5. 2. b to clarify that the intention is to maintain the IEEE Std 802. 11 TM-2020 requirements regarding changing a MAC address, e. g. not during an ESS connection, has been added to section 8. 1. Submission Slide 9 Stephen Mc. Cann, Huawei

November 2020 doc. : IEEE 802. 11 -20/1770 r 6 802. 1 comments on PARs and CSDs P 802. 11 bh PAR 5. 2. b. Scope of the project: • The use of 802. 11 technology is expected to play an important and increasing role in environments that are tightly controlled, e. g. , in flexible factories https: //mentor. ieee. org/802. 1/dcn/20/1 -20 -0026 -00 -ICne-ieee 802 -nendica-report-flexible-factory-iot-use-cases-and-communication-requirements-for-wired-and-wireless -bridged-networks. pdf. 802. 1 is concerned that P 802. 11 bh could restrict the use of some 802. 11 devices in these environments, and thus reduce the use of low latency and prioritization capabilities discussed in the context of 802. 11 be with TSN. • Please amend the Scope to indicate that a conformant 802. 11 device should be configurable to operate without address randomization. Response: The IEEE P 802. 11 REVmd D 5. 0 allows configuration of whether a STA randomizes or not the MAC address (see the use of dot 11 MACPrivacy. Activated). No changes required. Submission Slide 10 Stephen Mc. Cann, Huawei

November 2020 doc. : IEEE 802. 11 -20/1770 r 6 802. 3 comments on PARs and CSDs P 802. 11 bh PAR 6. 1. 2, Registration activity • The PAR does not make it clear that work will only use randomization and changing of MAC addresses within the local address space. Without that restriction, this question should be answered yes. Because the current my. Project PAR code only allows an explanation here if answered yes, either the answer could be changed to yes here citing previous RAC interest in MAC address randomization, or an explanation of the restriction could be added to 8. 1. Response: The answer to 6. 1. 2 has been changed to “yes”. Submission Slide 11 Stephen Mc. Cann, Huawei

November 2020 doc. : IEEE 802. 11 -20/1770 r 6 Changes to the 802. 11 bh PAR • The new document number of the P 802. 11 bh PAR is: 11 -20 -1795 r 1 • Change the title to: “Operation with Randomized and Changing MAC Addresses” • Change 5. 2. b to: “This amendment specifies modifications to the medium access control (MAC) mechanisms to preserve the existing services that might otherwise be restricted in environments where STAs in an ESS use randomized or changing MAC addresses, without affecting user privacy. User privacy includes exposure of trackable information to third parties or exposure of an individual's presence or behavior. This amendment introduces mechanisms to enable session continuity in the absence of unique MAC address-to-STA mapping. For STAs in an ESS that use randomized or changing MAC addresses, this amendment preserves the ability to provide customer support, conduct network diagnostics and troubleshooting, and detect device arrival in a trusted environment. ” • Change 6. 1. 2 to: Is the Standards Committee aware of possible registration activity related to this project? : Yes and add “Explanation: The RAC may want to review for correct and consistent usage of registry terms. ” • In section 8. 1, add the following: “ 5. 2. b: The IEEE Std 802. 11 TM-2020 requirements regarding changing a MAC address will be maintained. A non-AP STA MAC address shall not change during an ESS connection. ” Submission Slide 12 Stephen Mc. Cann, Huawei

November 2020 doc. : IEEE 802. 11 -20/1770 r 6 IEEE 802. 11 bh CSD Submission Slide 13 Stephen Mc. Cann, Huawei

November 2020 doc. : IEEE 802. 11 -20/1770 r 6 802. 1 comments on PARs and CSDs P 802. 11 bh CSD Titles • The association between each CSD and its PAR is not entirely clear. One CSD is entitled “CSD Draft for Privacy Amendment of RCM Study Group” and the other “RCM SG Proposed CSD Draft for 802. 11 RCM Project”. The association between a CSD and a PAR needs to be clear now and 5 years hence when the CSD is being confirmed for project progression to SA Ballot and Rev. Com. After 5 years many will have forgotten subtle distinctions that accompanied the initial discussion and might not even be sure that a CSD belongs to either project. • Include the project designation P 802. 11 bi/P 802. 11 bh and amendment title in the CSD. Response: The title has been changed to: “Proposed CSD for P 802. 11 bh” Submission Slide 14 Stephen Mc. Cann, Huawei

November 2020 doc. : IEEE 802. 11 -20/1770 r 6 802. 1 comments on PARs and CSDs P 802. 11 bh CSD 1. 2. 2 Compatibility: • Both CSDs claim compliance with 802. 1 Q, without qualification. However, this appears to be based on an incomplete understanding of the scope of 802. 1 Q. The referenced presentation (19 0851 00, P 802. 1 CQ MAC Address Assignment Requirements) states: “P 802. 1 CQ is currently the only 802. 1 project dealing with end station behavior … Formerly, protocols for end stations were out of scope for 802. 1”. This is not the case and has not been the case for many years. 802. 1 Q includes end station requirements for participation in the MMRP, MVRP, MSRP, CFM, and other protocols. These protocols are important in Time Sensitive Networking environments requiring reliable and timely frame delivery (such as industrial and professional audio networks) and in service provider networks. The reference to 19 -0851 -00 is misleading. A more complete analysis of the interaction of 802. 1 protocols is in an informative annex (following IEEE Std 802 E recommendations) “Privacy in Bridged Networks” to P 802. 1 AEdk/D 0. 5 (also in first task group ballot). • Remove the 19 -0851 -00 presentation from the list of CSD references. A more detailed response to the 802. 1 Q compliance question is required. Use of a fixed STA/MAC relationship needs to remain possible throughout the duration of an association. Response: The response to 1. 2. 2 is clear and precise. The References have been removed from the CSD. Submission Slide 15 Stephen Mc. Cann, Huawei

November 2020 doc. : IEEE 802. 11 -20/1770 r 6 802. 3 comments on PARs and CSDs P 802. 11 bh CSD 1. 2. 2, Compatibility • Once approved an amendment becomes part of the base standard, therefore, IEEE Std 802 c is part of IEEE Std 802 and should be addressed for this project. Unless a method for restricting the range for random local MAC addresses is specified, randomization is incompatible with the optional SLAP capabilities of Std 802 c. This should be addressed here. 1. 2. 3, Distinct identity • See comments on proposed P 802. 11 bi. There is insufficient information discriminating the work on this project and proposed P 802. 11 bi. Response: For 1. 2. 2, the answer is Yes. IEEE P 802. 11 REVmd D 5. 0 already requires compliance to optional SLAP capabilities of IEEE Std 802 c. There is no intention in 802. 11 bh to change this. For 1. 2. 3, the distinct identity for 802. 11 bh states: “address operational challenges resulting from the use of random or changing MAC addresses”. No changes required. Submission Slide 16 Stephen Mc. Cann, Huawei

November 2020 doc. : IEEE 802. 11 -20/1770 r 6 802. 1 comments on PARs and CSDs P 802. 11 bh CSD References: • It is unclear why the presentations 19 -0884 -00, 19 -1027 -01, 19 -1320 -00 are in either, much less both CSDs. • P 802. 11 bi concerns privacy exposures beyond the use of persistent source MAC Address, and these presentations are very much focused on MAC Address assignment. • The presentations reflect the different and contested views on 802. 11 aq and 802 c on the extent of the randomly chosen MAC address space, but their bearing on the feasibility or otherwise of the proposed PARs is nowhere made clear. Since they promote opposing views it is not clear what conclusion the reader of the CSDs is meant to draw. • Remove these three presentations from the references list of both CSDs. Response: The references have been removed from the CSD. The second paragraph of 1. 2. 4 j) has been removed from the CSD. Submission Slide 17 Stephen Mc. Cann, Huawei

November 2020 doc. : IEEE 802. 11 -20/1770 r 6 Changes to the 802. 11 bh CSD • Change the title to: “Proposed CSD for P 802. 11 bh” • Remove the 2 nd paragraph of 1. 2. 4 j) • Remove the references Submission Slide 18 Stephen Mc. Cann, Huawei

November 2020 doc. : IEEE 802. 11 -20/1770 r 6 IEEE 802. 11 bi PAR Submission Slide 19 Stephen Mc. Cann, Huawei

November 2020 doc. : IEEE 802. 11 -20/1770 r 6 802. 3 comments on PARs and CSDs P 802. 11 bi PAR General • Neither the PARs nor CSDs clearly distinguish the need for two similar projects. Scopes overlap, and distinct identity is not established between proposed P 802. 11 bh and proposed P 802. 11 bi. • We appreciate the submittal is a. pdf, but the file name being a docx extension caused some problems in opening. Response: The 802. 11 bi PAR amends the standard to address user privacy requirements for 802. 11 non-AP STAs. No changes required. Submission Slide 20 Stephen Mc. Cann, Huawei

November 2020 doc. : IEEE 802. 11 -20/1770 r 6 802. 3 comments on PARs and CSDs P 802. 11 bi PAR General • Neither the PARs nor CSDs clearly distinguish the need for two similar projects. Scopes overlap, and distinct identity is not established between proposed P 802. 11 bh and proposed P 802. 11 bi. • We really can’t understand what this PAR is proposed to do. A better Project Scope or more detail in Need could help with this. 2. 1, Title • Capitalization is inconsistent (not title case nor consistent with IEEE style). Recommend “Enhanced service with data privacy protection”. Response: As stated in the section 5. 2. b, the scope of the amendment is: “to specify new mechanisms that address and improve user privacy. ” The title has been changed to “Enhanced Service with Data Privacy Protection” Submission Slide 21 Stephen Mc. Cann, Huawei

November 2020 doc. : IEEE 802. 11 -20/1770 r 6 802. 1 comments on PARs and CSDs P 802. 11 bi PAR 5. 2. b. Scope of the project: • 802. 1’s interpretation of the Scope of both PARs is that incorporation of additional information elements in 802. 11 that could further diminish user privacy (to, for example, facilitate continued user identification in the absence of a permanent STA to MAC address association) is Out of Scope. • Please amend the Scope to state explicitly what new user tracking capabilities are within or excluded from the Scope of the PAR. Response: An additional explanatory note with respect to 5. 2. b to clarify that the intention is not to diminish user privacy has been added to section 8. 1. Submission Slide 22 Stephen Mc. Cann, Huawei

November 2020 doc. : IEEE 802. 11 -20/1770 r 6 802. 1 comments on PARs and CSDs P 802. 11 bi PAR 5. 2. b. Scope of the project 5. 5 Need for the Project: • 802. 1 is concerned that both Scope and Need of P 802. 11 bh appear to explicitly limit consideration of disruption, and the need for session continuity, to 802. 11 mechanisms. Important 802. 1 Q TSN capabilities, such as bandwidth allocation, depend on flow identification that span both the wired and wireless networks. [Our comments on the CSDs point out that Technical Feasibility appears to be based, at least in part, on erroneous information about the scope of 802. 1 in general and 802. 1 Q in particular]. • Please amend the Scope to explicitly indicate that the mechanisms to enable session continuity recognize the fact that sessions extend to the LAN beyond the medium supported by 802. 11. Response: An additional explanatory note with respect to 5. 2. b to clarify that the intention is not to diminish user privacy, has been added to section 8. 1. Submission Slide 23 Stephen Mc. Cann, Huawei

November 2020 doc. : IEEE 802. 11 -20/1770 r 6 802. 3 comments on PARs and CSDs P 802. 11 bi PAR 6. 1. 2, Registration activity • The PAR does not make it clear that work will only use randomization and changing of MAC addresses within the local address space. Without that restriction, this question should be answered yes. Because the current my. Project PAR code only allows an explanation here if answered yes, either the answer could be changed to yes here citing previous RAC interest in MAC address randomization, or an explanation of the restriction could be added to 8. 1. Response: The answer to 6. 1. 2 has been changed to “yes”. Submission Slide 24 Stephen Mc. Cann, Huawei

November 2020 doc. : IEEE 802. 11 -20/1770 r 6 Changes to the 802. 11 bi PAR • • • The new document number of the P 802. 11 bi PAR is: 11 -20 -1801 r 0 Change the title to: “Enhanced Service with Data Privacy Protection” Change 6. 1. 2 to: Is the Standards Committee aware of possible registration activity related to this project? : Yes and add “Explanation: The RAC may want to review for correct and consistent usage of registry terms. ” • In section 8. 1, add the following: “ 5. 2. b: The IEEE Std 802. 11 TM-2020 requirements regarding user privacy will not be diminished by new requirements added in this amendment. ” Submission Slide 25 Stephen Mc. Cann, Huawei

November 2020 doc. : IEEE 802. 11 -20/1770 r 6 IEEE 802. 11 bi CSD Submission Slide 26 Stephen Mc. Cann, Huawei

November 2020 doc. : IEEE 802. 11 -20/1770 r 6 802. 1 comments on PARs and CSDs P 802. 11 bi CSD Titles • The association between each CSD and its PAR is not entirely clear. One CSD is entitled “CSD Draft for Privacy Amendment of RCM Study Group” and the other “RCM SG Proposed CSD Draft for 802. 11 RCM Project”. The association between a CSD and a PAR needs to be clear now and 5 years hence when the CSD is being confirmed for project progression to SA Ballot and Rev. Com. After 5 years many will have forgotten subtle distinctions that accompanied the initial discussion and might not even be sure that a CSD belongs to either project. • Include the project designation P 802. 11 bi/P 802. 11 bh and amendment title in the CSD. Response: The title has been changed to “Proposed CSD for P 802. 11 bi” Submission Slide 27 Stephen Mc. Cann, Huawei

November 2020 doc. : IEEE 802. 11 -20/1770 r 6 802. 1 comments on PARs and CSDs P 802. 11 bi CSD 1. 2. 2 Compatibility: • Both CSDs claim compliance with 802. 1 Q, without qualification. However, this appears to be based on an incomplete understanding of the scope of 802. 1 Q. The referenced presentation (19 0851 00, P 802. 1 CQ MAC Address Assignment Requirements) states: “P 802. 1 CQ is currently the only 802. 1 project dealing with end station behavior … Formerly, protocols for end stations were out of scope for 802. 1”. This is not the case and has not been the case for many years. 802. 1 Q includes end station requirements for participation in the MMRP, MVRP, MSRP, CFM, and other protocols. These protocols are important in Time Sensitive Networking environments requiring reliable and timely frame delivery (such as industrial and professional audio networks) and in service provider networks. The reference to 19 -0851 -00 is misleading. A more complete analysis of the interaction of 802. 1 protocols is in an informative annex (following IEEE Std 802 E recommendations) “Privacy in Bridged Networks” to P 802. 1 AEdk/D 0. 5 (also in first task group ballot). • Remove the 19 -0851 -00 presentation from the list of CSD references. A more detailed response to the 802. 1 Q compliance question is required. Use of a fixed STA/MAC relationship needs to remain possible throughout the duration of an association. Response: The response to 1. 2. 2 is clear and precise. The references have been removed from the CSD. Submission Slide 28 Stephen Mc. Cann, Huawei

November 2020 doc. : IEEE 802. 11 -20/1770 r 6 802. 3 comments on PARs and CSDs P 802. 11 bi CSD 1. 2. 3, Distinct identity There is insufficient information discriminating the work on this project and proposed P 802. 11 bh seems to be a subset of what this project would allow to be done. Response: For 1. 2. 3, the distinct identity for 11 bi specifies requirements to “improve the privacy experienced by users in environments using IEEE Std 802. 11 technology”. No changes required. Submission Slide 29 Stephen Mc. Cann, Huawei

November 2020 doc. : IEEE 802. 11 -20/1770 r 6 802. 1 comments on PARs and CSDs P 802. 11 bi CSD 1. 2. 4 Technical Feasibility • The P 802. 11 bi CSD (20/1346 r 2) states under 1. 2. 4 Technical Feasibility that: “The IEEE 802. 11 Wireless Next Generation (WNG) Standing Committee (SC) and RCM Topic Interest Group (TIG)/Study Group (SG) have reviewed many presentations indicating that enhancements are technically feasible. These contributions outlined techniques related to privacy to enhance current use cases and enable new ones. ” • However, none of these presentations appear to be referenced in the CSD, and we have not identified them in the RCM study group files on Mentor. The CSD references evaluations of the difficulties faced in improving user privacy (19/588 r 2, 19/1313 r 2, 19/1314 r 2). It is not at present clear that it will be possible to materially improve privacy in the face of a determined attacker who is prepared to correlate information from multiple information elements and other characteristics of device behavior. It is not clear that these improvements will outweigh attendant difficulties in service provision. 802. 1 understands that some proprietary attempts to improve privacy related behavior have been beta tested and withdrawn. 802. 1 does appreciate that the technical challenges faced in demonstrating feasibility in large scale deployment might not be reasonably addressed within the limited lifetime of a Study Group. However, that is no reason to overstate Technical Feasibility in the CSD. • Provide credible references and demonstrate Technical Feasibility. The current set of references do not meet that criteria. Response: The first paragraph of 1. 2. 4 j) is sufficient to demonstrate technical feasibility for 11 bi. The second paragraph of 1. 2. 4 j) has been removed from the CSD. Submission Slide 30 Stephen Mc. Cann, Huawei

November 2020 doc. : IEEE 802. 11 -20/1770 r 6 802. 1 comments on PARs and CSDs P 802. 11 bi CSD References: • It is unclear why the presentations 19 -0884 -00, 19 -1027 -01, 19 -1320 -00 are in either, much less both CSDs. • P 802. 11 bi concerns privacy exposures beyond the use of persistent source MAC Address, and these presentations are very much focused on MAC Address assignment. • The presentations reflect the different and contested views on 802. 11 aq and 802 c on the extent of the randomly chosen MAC address space, but their bearing on the feasibility or otherwise of the proposed PARs is nowhere made clear. Since they promote opposing views it is not clear what conclusion the reader of the CSDs is meant to draw. • Remove these three presentations from the references list of both CSDs. Response: The references have been removed from the CSD. Submission Slide 31 Stephen Mc. Cann, Huawei

November 2020 doc. : IEEE 802. 11 -20/1770 r 6 Changes to the 802. 11 bi CSD • Change the title to: “Proposed CSD for P 802. 11 bi” • Remove the 2 nd paragraph of 1. 2. 4 j) • Remove the references Submission Slide 32 Stephen Mc. Cann, Huawei