Source-linked AI summary
What Will Wi-Fi 8 Be? A Primer on IEEE 802.11bn Ultra High Reliability
Lorenzo Galati Giordano, Giovanni Geraci, Marc Carrascosa, Boris Bellalta
TL;DR
Emerging applications place stringent demands on Wi-Fi reliability, particularly for data delivery and latency. The paper reviews 802.11bn UHR and studies multi-link operation with multi-AP coordination; its case study reports that standalone MLO can exceed 100 ms at the 99.9999% delay percentile and discusses CBF as a way to address dense, high-contention scenarios.
Problem
Emerging applications require reliable data delivery and low maximum latency, needs that Wi-Fi must address in unlicensed spectrum subject to uncontrolled interference.
Source: §I · p. 1Method
The paper reviews Wi-Fi 8 use cases and features, then presents a system-level study of multi-link operation combined with spatial-domain multi-AP coordination.
Source: §I · p. 1, §VII · p. 7Results
With standalone MLO, the 99.9999% delay exceeds 100 ms in the studied scenario; the paper presents CBF as addressing the dense-scenario low-delay challenge.
Source: §VI · pp. 6–7Takeaways & Limitations
The study demonstrates how joint interworking of MLO and spatial-domain multi-AP coordination could achieve ultra-high reliability in Wi-Fi 8.
Source: §VII · p. 7
Abstract
from arXiv · showhide
What will Wi-Fi 8 be? Driven by the strict requirements of emerging applications, next-generation Wi-Fi is set to prioritize Ultra High Reliability (UHR) above all. In this paper, we explore the journey towards IEEE 802.11bn UHR, the amendment that will form the basis of Wi-Fi 8. We first present new use cases calling for further Wi-Fi evolution and associated standardization, certification, and spectrum allocation efforts. We then introduce a selection of the main disruptive features envisioned for Wi-Fi 8 and their associated research challenges, resulting from the outcome of the UHR Study Group. Among those, we focus on multi access point coordination and demonstrate that it could build upon 802.11be multi-link operation to make UHR a reality in Wi-Fi 8.
I. INTRODUCTION
Wi-Fi 8 is framed as a shift from pursuing throughput and lower delay toward prioritizing reliability for emerging applications. The paper links that shift to demanding use cases and to the challenge of providing deterministic service in unlicensed spectrum.
I. INTRODUCTION
- Machines may require Gbps communication with at least three nines of reliability for data delivery and maximum latency, while 0.01% excessively delayed holographic packets could trigger nausea and distress.The paper also identifies immersive communications, manufacturing digital twins, e-health, and cooperative mobile robots as emerging use cases.Source: §I · p. 1, A. Emerging Applications and Use cases · p. 1
II. EMERGING USE CASES DRIVING NOVEL STANDARDIZATION EFFORTS
- Unlike Wi-Fi 7’s focus on throughput and delay, emerging Wi-Fi 8 use cases also require higher reliability, prompting the paper’s UHR focus.The paper positions 802.11bn UHR as the amendment that will form the basis of Wi-Fi 8.Source: §II · p. 1, §I · p. 1
A. Emerging Applications and Use cases
- Wi-Fi’s unlicensed operation exposes it to channel contention and uncontrolled interference, so Wi-Fi 8 intends to pursue determinism through coordination and more efficient spectrum use.Wi-Fi’s MAC was originally designed around CSMA/CA to cope with uncoordinated spectrum use.Source: A. Emerging Applications and Use cases · p. 2
B. Novel Standardization Efforts
Standardization efforts are carrying low-latency and reliability recommendations forward from Wi-Fi 7 into Wi-Fi 8, while an AI/ML group examines possible protocol applications.
B. Novel Standardization Efforts
- The 2019 Real Time Applications Topic Interest Group recommended guidelines for low latency and reliability, informing Wi-Fi 7 and Wi-Fi 8 efforts including TSN integration.The recommendations have been considered in Wi-Fi 7 development, including MLO.Source: B. Novel Standardization Efforts · p. 2
IEEE 802.11 AI/ML Topic Interest Group (TIG):
- The AI/ML Topic Interest Group examines use cases and technical feasibility based on existing mechanisms and expected implementation effort.Its scope is applying AI and machine learning directly to Wi-Fi protocols.Source: IEEE 802.11 AI/ML Topic Interest Group… · p. 2
- Examples under discussion include neural-network CSI feedback compression, AI/ML-assisted roaming, deep reinforcement-learning channel access, and AI/ML-driven multi-AP coordination.The passage lists these as examples of potential AI/ML applications to Wi-Fi protocols.Source: IEEE 802.11 AI/ML Topic Interest Group… · p. 2
IEEE 802.11 Integrated mmWave Study Group (IMMW SG):
The paper describes a separate mmWave study effort alongside 802.11bn UHR standardization, whose objectives target throughput, latency, loss, power saving, and peer-to-peer exchange. It identifies seamless connectivity, determinism, and controlled worst-case delay as reliability concerns.
IEEE 802.11 Integrated mmWave Study Group (IMMW SG):
- A dedicated mmWave study group was created to lay the basis for an amendment using existing Wi-Fi PHY/MAC functions to dynamically operate additional mmWave links.The passage notes potential operation across the three sub-7 GHz bands and mmWave, including the 60 GHz and 45 GHz bands.Source: IEEE 802.11 Integrated mmWave Study… · p. 2
IEEE 802.11bn UHR:
- The 802.11bn Task Group formed in November 2023 for a single-release cycle lasting until 2028, following the UHR Study Group’s establishment in July 2022.The figure compares 802.11be (top) and 802.11bn (bottom) standardization, certification, and commercialization timelines.Source: IEEE 802.11bn UHR · p. 2, Figure 1 · p. 3
- 802.11bn objectives include 25% higher throughput, 25% lower 95th-percentile latency, and 25% lower MPDU loss, including with mobility and overlapping BSSs.The listed objectives also include improved AP power saving and enhanced direct peer-to-peer exchanges.Source: IEEE 802.11bn UHR · p. 3
- The reliability investigation focuses on seamless connectivity, determinism, and controlled worst-case delay, each with distinct opportunities and challenges.Source: IEEE 802.11bn UHR · p. 3
III. SEAMLESS CONNECTIVITY VIA DISTRIBUTED MLO
Distributed multi-link operation extends the 802.11be MLD architecture to APs on separate hardware, with the aim of supporting seamless mobility. Implementing it requires coordination, addressing, and reliable interfaces between the coordination entity and APs.
III. SEAMLESS CONNECTIVITY VIA DISTRIBUTED MLO
- 802.11be MLO supports low-signaling, low-delay link switching and make-before-break transitions between APs controlled by the same MLD.Source: §III · p. 3
Opportunities:
- 802.11bn could extend an MLD across physically distributed APs, creating a virtual cell where multiple links can remain active so a mobile device stays connected to at least one.The paper describes this as embedding native roaming support and improving connection reliability.Source: Opportunities · p. 3
Technical challenges:
- Distributed MLO requires coordination among APs and unique link addressing, since 802.11be guarantees different identifiers only for links on the same physical device.The distributed architecture also needs reliable, general interfaces between the coordination entity and APs, supporting wired or wireless communications and vendor interoperability.Source: Technical challenges · p. 3, §IV · p. 3
Potential implementation:
Unexpected, event-driven time-sensitive traffic can face latency from an ongoing transmission and subsequent contention. Wi-Fi 8 is considering MAC changes to serve such packets promptly through reserved resources and secondary-channel access.
IV. DETERMINISM VIA PHY AND MAC ENHANCEMENTS
- Unexpected high-priority packets can wait for an ongoing transmission to finish and then face channel contention.These are the two primary latency sources identified for event-driven, time-sensitive traffic.Source: §IV · p. 3
- 802.11bn may extend EDCA with more priority classes, add RU reservations and preemption to OFDMA, and use idle secondary channels when the primary is occupied.The proposed EDCA extension includes associated channel-access parameters such as backoff.Source: §IV · p. 4
Proposed mechanisms:
- Resource Reservation and Channel Preemption could reserve an RU for low-latency packets while allowing ordinary data to use it when needed.Pre-padding could let a node promptly place an arriving low-latency packet in the reserved RU; this avoids receiver design changes but requires a multidimensional PPDU and effective preemption between PPDUs when the sender is not the TXOP holder.Source: Proposed mechanisms · p. 4, Figure 2 · p. 4
- Secondary Channel Access could allow transmission on an idle secondary channel without depending on the primary, with anticipated gains at medium-to-high load and without excessive complexity.It may extend 802.11be preamble-puncturing functionality.Source: Proposed mechanisms · p. 4
V. CONTROLLED WORST-CASE DELAY VIA MULTI-AP COORDINATION
Contention between basic service sets leaves worst-case Wi-Fi delays unpredictable, even when access points share an administrative domain. Wi-Fi 8 is expected to use multi-AP coordination, supported by channel-state acquisition and protocol upgrades, to reduce contention and improve reliability.
V. CONTROLLED WORST-CASE DELAY VIA MULTI-AP COORDINATION
- Inter-BSS contention makes worst-case delays unpredictable, motivating Wi-Fi 8 multi-AP coordination to improve reliability, particularly in dense, heavily loaded environments.Restricted-TWT scheduling reduces contention within a BSS, but inter-BSS access remains contention-based.Source: §V · p. 4
Channel state information acquisition:
- Multi-AP coordination can rely on OBSS channel sounding: neighboring STAs measure channels and feed information back to both the serving and OBSS APs.The information can support frequency-resource management, transmit-power adjustment, or beamforming to avoid OBSS interference; acquisition overhead may offset performance gains.Source: Channel state information acquisition · p. 4
Protocol upgrades:
- New frames are envisioned for discovering and managing multi-AP groups, sharing channel and buffer states, and triggering coordinated transmissions to reduce inter-BSS collisions.Coordination may use over-the-air and wired signaling.Source: Protocol upgrades · p. 4
- The schemes may range from basic to advanced depending on inter-AP data exchange and implementation complexity, while the standard-versus-implementation boundary remains undecided.The main coordination schemes are still being considered.Source: Protocol upgrades · p. 5
A. Coordinated TDMA/OFDMA
Coordinated TDMA/OFDMA divides time or frequency resources among access points, while coordinated spatial reuse enables parallel transmissions through shared power control. The approaches trade reduced contention or increased reuse against coordination overhead and technical requirements.
A. Coordinated TDMA/OFDMA
- C-TDMA assigns different APs sequential TXOP slots, whereas C-OFDMA allocates different portions of the frequency band.For C-OFDMA, smaller 20 MHz units offer more flexibility and scheduling gain than 80 MHz units but may require PHY-format changes.Source: A. Coordinated TDMA/OFDMA · p. 5
- C-OFDMA can reduce latency by reducing contention, but the sharing AP must collect neighboring APs’ channel and buffer status and schedule their resources.This creates computational burden and overhead.Source: A. Coordinated TDMA/OFDMA · p. 5
B. Coordinated Spatial Reuse
- C-SR coordinates AP transmit power to support adequate SINR at receiving STAs and parallel transmissions on the same time/frequency resources, potentially increasing throughput and reducing queuing delay.Unlike 802.11ax spatial reuse, this approach does not require other APs to reduce power based on one AP transmitting at maximum power.Source: B. Coordinated Spatial Reuse · p. 5
- C-SR needs RSSI measurements for interfering links; beacon measurements may limit overhead because RSSI is relatively static, while beamforming-aware estimates may improve performance at added complexity and overhead.One proposed implementation has a sharing AP gather STA RSSI, other APs’ target SINRs and intended recipients, then convey calculated powers by trigger frame so APs can select modulation and coding schemes.Source: B. Coordinated Spatial Reuse · p. 5
C. Joint Transmission
Joint transmission is a distributed-MIMO approach in which non-co-located access points jointly serve multiple stations. It may combine high throughput and low latency, but depends on coordination, synchronization, and data-sharing arrangements.
C. Joint Transmission
- Joint transmission, or distributed MIMO, has non-co-located APs jointly transmit data to or receive data from multiple STAs.The approach treats neighboring APs as servers rather than potential interferers.Source: C. Joint Transmission · p. 5
- Joint transmission could simultaneously achieve high throughput and low latency by suppressing interference without sacrificing spatial streams.Source: C. Joint Transmission · p. 5
- Joint transmission may require a new distributed CSMA/CA protocol, tight time, frequency, and phase synchronization, and data sharing among all participating APs.An out-of-band backhaul link, such as 10 Gbps Ethernet, is likely needed to limit sharing overhead and avoid increased queueing delay.Source: C. Joint Transmission · p. 5
- A potential implementation exchanges coordination requests and responses, shares data before triggering the joint transmission, and ends with block acknowledgments from receiving STAs.Overhead may be limited by sharing data in advance when possible, or transmitting wirelessly to other STAs while wired data sharing proceeds.Source: C. Joint Transmission · p. 5
D. Coordinated Beamforming
Coordinated beamforming (CBF) uses collaborative access points to suppress interference from neighboring, non-associated stations. Its potential latency and spatial-reuse benefits depend on CSI overhead, nulling accuracy, and limited antenna degrees of freedom.
D. Coordinated Beamforming
- CBF places radiation nulls toward neighboring non-associated stations, helping avoid contention, permit full-power transmissions, and potentially improve worst-case latency.Each AP uses spatial degrees of freedom for both its own spatial multiplexing and interference suppression.Source: D. Coordinated Beamforming · p. 5
- Unlike joint transmission, CBF keeps each station's data processing at a single AP, avoiding data-sharing overhead and off-band backhauling.CSI acquisition overhead still needs consideration, especially as antenna arrays grow.Source: D. Coordinated Beamforming · p. 5
- CBF must balance data-carrying spatial streams, beamforming gain, and nulling accuracy within the antenna array's limited spatial degrees of freedom.The design also calls for opportunistic user scheduling during newly created spatial-reuse opportunities.Source: D. Coordinated Beamforming · p. 6
- A potential CBF design includes AP coordination, CSI acquisition for interference suppression, and dynamic null-steering in which a donor AP grants an OBSS AP a transmission opportunity.The OBSS AP is obligated to null toward the donor AP's served stations; under the right circumstances, CBF could reduce worst-case latency with lower implementation complexity than joint transmission.Source: D. Coordinated Beamforming · p. 6
VI. CASE STUDY: ULTRA HIGH RELIABILITY IN WI-FI 8
In a simulated two-BSS Wi-Fi 7 MLO network, combining MLO with CBF produced delay benefits only when nulling accuracy was sufficient to offset interference and CSI overhead. At 30 dB or higher accuracy, the 99.9999%-tile delay fell by nearly an order of magnitude compared with standalone MLO.
VI. CASE STUDY: ULTRA HIGH RELIABILITY IN WI-FI 8
- The case study simulates two overlapping BSSs with four-antenna APs and two-antenna stations, each AP carrying a 2 Gbps holographic-video stream.Only AP-STA delay is evaluated; other Wi-Fi 7 and potential Wi-Fi 8 features are omitted to isolate CBF gains.Source: §VI · p. 6
- CSI acquisition overheads of 1 ms and 2 ms can nearly outweigh a CBF nulling capability of 20 dB.The figure compares median, 99%-tile, and 99.9999%-tile delays across nulling accuracies and overhead levels.Source: §VI · p. 6, Figure 4 · p. 7
- At 30 dB or more nulling accuracy, MLO with CBF enables the highest MCS (4096-QAM 5/6) and nearly an order-of-magnitude reduction in 99.9999%-tile delay.Standalone MLO's 99.9999% delay exceeds 100 ms; at 10 dB null accuracy, CBF instead worsens performance, with MCS down to 16-QAM 3/4.Source: §VI · p. 6, Figure 4 · p. 7
- Even with Wi-Fi 7 features such as 4096-QAM and MLO, low delay with ultra-high reliability remains challenging in dense, highly contended scenarios.The results suggest CBF can help Wi-Fi 8 address such demands, including for immersive holographic communications.Source: §VI · p. 7
VII. CONCLUSION
The conclusion reports that jointly interworking multi-link operation with spatial-domain multi-AP coordination could deliver UHR in Wi-Fi 8.
VII. CONCLUSION
- A system-level study demonstrates how joint interworking of multi-link operation and spatial-domain multi-AP coordination could achieve UHR.Source: §VII · p. 7
- The paper reviews emerging use cases, standardization, certification, spectrum regulation, and disruptive innovations envisioned to meet Wi-Fi 8 requirements.Source: §VII · p. 7
BIOGRAPHIES
- The biographies span Nokia Bell Labs and Telefónica Research roles, UPF appointments, and research interests in wireless-network optimization, Wi-Fi, and machine-learning-based adaptive systems.Source: BIOGRAPHIES · p. 7