Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Evaluating UORA-Based Polling Mechanism for Latency-Sensitive Uplink Traffic in Wi-Fi Networks

Douglas Dziedzorm Agbeve, Andrey Belogaev, Chris Blondia, Jeroen Famaey

TL;DR

This paper tackles latency-sensitive uplink traffic in Wi-Fi 6 by evaluating UORA as a scalable polling mechanism against SA UL OFDMA, A2P, and EDCA. It implements and compares these schemes in ns-3 (via the open-source UORAns3) to measure uplink delay and total throughput under mixed deterministic and stochastic traffic using 26-tone RUs. The results show UORA can reduce delay and maintain throughput in dense environments with sparse, latency-sensitive transmissions, with performance highly sensitive to the choice of $OCW_{min}$ and the number of RA RUs, exhibiting a plateau beyond certain sparsity. The study demonstrates UORA's viability as a polling alternative, highlighting practical implications for uplink scheduling in future dense Wi-Fi deployments and outlining optimization opportunities for parameter settings.

Abstract

IEEE 802.11ax (Wi-Fi 6) introduced Orthogonal Frequency Division Multiple Access (OFDMA), which enables simultaneous transmissions through centralized resource allocation. However, effective uplink scheduling requires the Access Point (AP) to identify which stations (STAs) have data to transmit. This typically necessitates polling for buffer status reports, a process that becomes increasingly inefficient and unscalable with growing device density. In this paper, we study how the Uplink OFDMA-based Random Access (UORA) feature improves the scalability and delay experienced by latency-sensitive data streams. We show that UORA enables efficient uplink scheduling while opportunistically identifying buffered traffic from unscheduled STAs, striking a balance between coordination and scalability. Performance evaluation of different polling strategies is done by means of simulation in ns-3. The results indicate that UORA-based polling outperforms alternative schemes in densely deployed network environments with heterogeneous uplink traffic patterns. Furthermore, under highly sparse and sporadic traffic conditions, UORA-based polling yields over 40% delay reduction compared to Scheduled Access (SA) OFDMA.

Evaluating UORA-Based Polling Mechanism for Latency-Sensitive Uplink Traffic in Wi-Fi Networks

TL;DR

This paper tackles latency-sensitive uplink traffic in Wi-Fi 6 by evaluating UORA as a scalable polling mechanism against SA UL OFDMA, A2P, and EDCA. It implements and compares these schemes in ns-3 (via the open-source UORAns3) to measure uplink delay and total throughput under mixed deterministic and stochastic traffic using 26-tone RUs. The results show UORA can reduce delay and maintain throughput in dense environments with sparse, latency-sensitive transmissions, with performance highly sensitive to the choice of and the number of RA RUs, exhibiting a plateau beyond certain sparsity. The study demonstrates UORA's viability as a polling alternative, highlighting practical implications for uplink scheduling in future dense Wi-Fi deployments and outlining optimization opportunities for parameter settings.

Abstract

IEEE 802.11ax (Wi-Fi 6) introduced Orthogonal Frequency Division Multiple Access (OFDMA), which enables simultaneous transmissions through centralized resource allocation. However, effective uplink scheduling requires the Access Point (AP) to identify which stations (STAs) have data to transmit. This typically necessitates polling for buffer status reports, a process that becomes increasingly inefficient and unscalable with growing device density. In this paper, we study how the Uplink OFDMA-based Random Access (UORA) feature improves the scalability and delay experienced by latency-sensitive data streams. We show that UORA enables efficient uplink scheduling while opportunistically identifying buffered traffic from unscheduled STAs, striking a balance between coordination and scalability. Performance evaluation of different polling strategies is done by means of simulation in ns-3. The results indicate that UORA-based polling outperforms alternative schemes in densely deployed network environments with heterogeneous uplink traffic patterns. Furthermore, under highly sparse and sporadic traffic conditions, UORA-based polling yields over 40% delay reduction compared to Scheduled Access (SA) OFDMA.

Paper Structure

This paper contains 11 sections, 3 equations, 8 figures, 1 table.

Table of Contents

  1. Introduction
  2. Related Work
  3. Background: SA OFDMA and UORA in Wi-Fi
  4. SA UL OFDMA
  5. MU EDCA Parameter Set
  6. The UORA operation
  7. Implementation
  8. Performance Evaluation
  9. Simulation Setup
  10. Discussion of Results
  11. Conclusion

Figures (8)

  • Figure 1: Diagramatic representation of the A2P algorithm agbeve2025a2p
  • Figure 2: UL OFDMA frame exchange sequence
  • Figure 3: UL OFDMA frame exchange sequence with UORA. Tables in the bottom show the values of OFDMA Back-Off (OBO) for all STAs before and after BSRP TF is received. STAs whose OBO reaches 0 are framed with red.
  • Figure 4: Average delay of stochastic STAs as a function of the number of these STAs across different minimum contention window sizes
  • Figure 5: Average delay of stochastic STAs as a function of the number of these STAs
  • ...and 3 more figures