802.11ax trigger-based UL + TWT (Kestrel / RTL8852) — scheduled contention-free TX, the standards answer to the TX-egress gap

Ouverte
#236 1 commentaire 0 réactions 0 personnes assignées Voir sur GitHub

Les mainteneurs répondent en général sous 1 jour

Personne n'a encore pris cette issue.

Évaluation

Difficulté
5/5
Temps estimé
Plus d'une semaine
Accessibilité débutants
25/100
Type d'issue
Fonctionnalité
Clarté
Plutôt claire
Activité
Calme
Stack technique
c

Piste de recherche

Start by reading docs/timing-accuracy.md, docs/time-distribution.md, and the Kestrel bring-up notes in CLAUDE.md; the issue says the RTL8852 HAL is a prerequisite and currently out of scope. First determine during Kestrel bring-up whether firmware exposes HE Trigger-frame TX and TWT scheduling to the host. Done means establishing feasibility and, only if supported, implementing the scoped trigger-based UL and TWT scheduling; no specific source files or tests are identified.

Rédigé par le modèle d'indexation à partir du texte de l'issue.

Description

enhancement

Goal

This session characterized and measured the fundamental limit of scheduled TX on Realtek 11ac/n silicon: there is no host-reported TX-egress timestamp, and CSMA governs air-departure, so µs-precise scheduled per-frame TX is not achievable (docs/timing-accuracy.md). The only hardware-TX-timed primitive is the beacon engine (auto-TX at TBTT), which is why devourer's uplink timing-advance loop steers the beacon TBTT rather than data frames.

802.11ax's trigger-based UL access + TWT is the standard's mechanism for exactly this — the AP schedules UE transmit opportunities (contention-free, OFDMA UL via Trigger frames), and TWT negotiates wake/TX windows. That is genuinely cellular-like scheduled access, and it is the forward path to the "scheduled-TX half" of the LTE/5G-parity picture. This issue plans it for the Kestrel (RTL8852) family for whoever takes it on.

Depends on

  • A Kestrel HAL (RTL8852 family) — currently out of scope (see CLAUDE.md). This is a forward-looking plan; it presumes Kestrel bring-up exists first.

Why this is the right lever

Faster transports (PCIe, SDIO — #235) lower the submit floor but cannot schedule air-departure, because that is a MAC/firmware property, not a transport one. 802.11ax moves scheduling into the MAC/PHY: the AP's Trigger frame grants each STA a resource unit and a start time (SIFS after the trigger), and the STA's MAC fires the UL-OFDMA transmission at that instant — hardware-scheduled, contention-free. This is the only path to cellular-grade scheduled UL on Realtek Wi-Fi.

Scope / implementation detail

AP side

  • Build + air HE Trigger frames (Basic Trigger for UL-OFDMA): per-STA RU allocation, UL length, target RSSI, and the common info (UL BW, GI/LTF).
  • Own the schedule: which STAs transmit in which trigger, at what cadence.

STA side

  • Respond to a Trigger with UL-OFDMA TX at trigger + SIFS — this is the hardware-scheduled TX (the STA MAC fires at the scheduled instant, not after CSMA backoff).

TWT

  • Negotiate Target Wake Time agreements (individual + broadcast TWT): schedule wake / service periods; the AP owns the timetable. Useful independently for power-scheduled FPV links.

Key question (decides feasibility)

Does the RTL8852 firmware expose trigger-frame TX + TWT scheduling to the host — the way the current silicon exposes TBTT-timed beacon TX via the reserved-page engine?

  • If yes, this delivers the µs-precise scheduled TX that USB/SDIO/PCIe cannot — because it is a MAC/firmware mechanism, transport-independent.
  • If the firmware owns trigger/TWT scheduling opaquely (typical of consumer 11ax), host control may be limited to negotiating TWT parameters rather than µs-steering individual transmissions. This is the first thing to determine during Kestrel bring-up.

Relation to existing work

devourer's uplink timing-advance loop (beacon-engine TBTT steering, docs/time-distribution.md) is a workaround for the missing scheduled TX. 802.11ax trigger-UL would replace it with the native mechanism — and would be the concrete "we closed the scheduled-TX gap vs LTE/5G" milestone.

Out of scope

The Kestrel HAL itself (prerequisite), and full 11ax data-plane (MU-MIMO, 1024-QAM) beyond what trigger-UL/TWT scheduling needs.

Langage dominant
C
Étoiles
56
Forks
29
Merge moyen
3 h 43 min
PR mergées (30 j)
75

Préparer son environnement

Ce projet ne fournit ni conteneur de développement, ni Dockerfile, ni guide de contribution : l'installation est à votre charge. Commencez par son README, et consultez notre guide de la première contribution pour les étapes générales.

Par où commencer

  1. Lisez l'issue en entier, puis le guide de contribution du projet.
  2. Signalez en commentaire que vous la prenez — cela évite que deux personnes fassent le même travail.
  3. Forkez le dépôt et travaillez sur une branche.
  4. Ouvrez une pull request qui référence le numéro de l'issue.

Autres issues de OpenIPC/devourer

Toutes les issues de OpenIPC/devourer

Issues similaires

Plus d'issues C

Recevez les nouvelles issues par e-mail

Un résumé court des issues GitHub adaptées aux débutants.