Session Date/Time: 20 Sep 2022 14:00

LPWAN

Summary

The LPWAN interim working group meeting discussed the status of various drafts, including the YANG data model, NB-IoT, SCHC over Sigfox, and Compound ACK. Key discussions revolved around addressing ISG comments for the YANG model, clarifying the normative/informational status of the NB-IoT draft relative to 3GPP, and exploring a novel approach to LPWAN convergence using SCHC over different radio technologies. Decisions were made to initiate a Working Group Last Call for the SCHC over Sigfox draft and to continue work on outstanding reviews and modifications for other drafts. Re-chartering ideas were solicited for the mailing list.

Key Discussion Points

IETF Boilerplate: The IPR, Anti-harassment, and Code of Conduct policies were reiterated as applicable to the meeting.
Architecture Draft: An attendee noted the architecture draft had expired. Pascal clarified it was updated in June, but welcomed discussion on adding new items, specifically regarding node identification (e.g., RRI). It was suggested to discuss this on the mailing list first.
SCHC over Sigfox Draft: Sergio had published new versions addressing Anna's review comments. Anna expressed that the draft is now ready for a Working Group Last Call.
Compound ACK Draft: Round trips between Dominic and Sergio were completed. Alexander confirmed he had started his shepherd review and plans to publish it by the next interim meeting.
YANG Data Model Reviews (Lahron):
- Examples in Annex: Proposed adding JSON examples of SCHC rules. While XML is traditional for RFCs, JSON is considered more helpful for developers.
- Security Considerations: Acknowledged the difficulty of addressing comprehensive security within this document, as it primarily formalizes RFC 8724. Security aspects related to rule access will be defined in the architecture draft. It was agreed to add standard security text, noting that read access to rules enables traffic decoding.
- References to Architecture Draft: Discussed if the YANG model's publication should wait for the architecture RFC. It was clarified that the architecture document is informational, not normative, allowing for an informative reference to a draft, thus avoiding delays.
- Identifier Features: Reviewer suggested using YANG 'features' (e.g., IPv6, CoAP) to make identifiers conditionally available. Lahron argued this would make the model too heavy, and identifiers are generic numbers. While acknowledging potential benefits for YANG engine validation, the practical impact on implementation was deemed minor. The working group agreed to prioritize moving forward.
- Field ID Testing: A suggestion for more granular tests (e.g., restricting token-length for IPv6 version field ID) was considered too heavy and potentially limiting for future augmentations. This point was deferred for mailing list discussion.
- NoAck/CoWase Anchors: Decision to keep empty noack and cowase nodes in the model as anchors for future augmentations.
NB-IoT Draft (Anna):
- Document Status (Standard Track vs. Informational): Discussion with Eric regarding the draft's status. For use cases involving 3GPP infrastructure (radio link, north access stratum), the document is informational. For end-to-end or application layer use cases where the 3GPP network is transparent, it can be normative/standard track. Anna will clarify this in the draft.
- Third Use Case Title: The title "end-to-end compression" was found confusing. "SCHC over Non-IP Data Delivery (NIDD)" was preferred for its clarity and specificity to 3GPP terminology.
- Terminology: The term "Capillary Gateway" will be defined in the terminology section, and "Wi-Fi" spelling will be corrected.
- Liaison with 3GPP: The process for liaising with 3GPP via Gonzalo (Ericsson) was reviewed. If Gonzalo is unresponsive, the chairs can send an email to 3GPP (CCing Gonzalo and statement@ietf.org) to invite comments on the draft during IETF last call.
LPWAN Convergence Research (Sergio):
- Sergio presented research on converging SCHC fragmentation and reassembly across different LPWAN technologies (Sigfox, LoRaWAN, NB-IoT).
- He highlighted that existing fragmentation modes (ACT, Uplink I-SCHC) share commonalities like tile size and header size.
- The proposed approach allows sending SCHC fragments (tiles) over different LPWANs, leveraging a single code base for fragmentation/reassembly and compression/decompression, with minimal adaptation logic.
- This could enable multi-radio devices, L2 transparency, and reduce code complexity, potentially allowing for a common "profile that works everywhere."
- The work rekindled a previous working group discussion about a "default rule" or "bootstrap profile" for SCHC, which was previously deemed difficult but now appears more feasible due to identified commonalities. It also raised the idea of sending the same fragments over different radio technologies for resilience.

Decisions and Action Items

Decision: The SCHC over Sigfox draft is ready for a Working Group Last Call (WGLC).
Action (Lahron): Add JSON examples (possibly XML) to an annex in the YANG data model draft.
Action (Lahron): Incorporate standard security considerations text from the IETF Wiki template into the YANG data model draft, acknowledging that read access to rules allows traffic decoding.
Action (Lahron): Reference the architecture draft informatively (not normatively) in the YANG data model draft to avoid publication delays.
Action (Lahron): Discuss the "Field ID testing" point on the mailing list.
Action (Lahron): Add tests for ignore, LSB, and matching-list requiring a target value in the YANG data model draft.
Action (Anna): Clarify the normative/informational status of different use cases within the NB-IoT draft, distinguishing between 3GPP infrastructure (informational) and application-layer/end-to-end (normative).
Action (Anna): Update the abstract of the NB-IoT draft (keep concise, correct "may adapt" to "may adopt").
Action (Anna): Rename the third use case title in the NB-IoT draft to "SCHC over Non-IP Data Delivery (NIDD)".
Action (Anna): Define "Capillary Gateway" in the terminology section of the NB-IoT draft and correct "Wi-Fi" spelling.
Action (Anna): Publish version 11 of the NB-IoT draft incorporating the above changes.
Action (Edgar): Contact Gonzalo (3GPP liaison manager) regarding the NB-IoT draft.
Action (Alexander): Complete the shepherd review for the Compound ACK draft and publish it by the next interim meeting.
Action (Sergio): Publish his LPWAN convergence research as a personal submission paper.

Next Steps

Working Group Last Call: Initiate WGLC for the SCHC over Sigfox draft.
Milestone Review: Decide whether to keep or drop the "ping and wake-up" milestone by the next IETF meeting.
Re-chartering: Participants are encouraged to post any ideas for future re-chartering items to the LPWAN mailing list for discussion.
Minutes Review: Attendees are requested to review and edit the meeting minutes on notes.ietf.org.