Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

QoS-Aware State-Augmented Learnable Framework for 5G NR-U/Wi-Fi Coexistence: Impact of Parameter Selection and Enhanced Collision Resolution

Mohammad Reza Fasihi, Brian L. Mark

TL;DR

The paper tackles QoS-aware coexistence of 5G NR-U and Wi-Fi in unlicensed bands by applying a state-augmented constrained RL controller (QaSAL) to dynamically tune MAC parameters. It examines cost scaling, violation modeling, and enhanced collision-resolution CR-LBT, alongside evaluating MAC parameters (CW,AIFSN,MCOT) for QoS and fairness. Key findings show that signed, threshold-invariant cost scaling stabilizes learning and constraint adherence; Contention Window control yields smoother delay compliance; and CR-LBT significantly reduces collisions and improves airtime efficiency, with CW-based control delivering the strongest QoS performance when paired with CR-LBT. The work also discusses deployment considerations, highlighting QaSAL’s relatively low per-step complexity versus primal-dual methods and outlining directions for multi-channel and PHY-parameter extensions.

Abstract

Unlicensed spectrum supports diverse traffic with stringent Quality-of-Service (QoS) requirements. In NR-U/Wi-Fi coexistence,the values of MAC parameters critically influence delay, collision behavior, and airtime fairness and efficiency. In this paper, we investigate the impact of (i) cost scaling and violation modeling, (ii) choice of MAC parameters, and (iii) an enhanced collision resolution scheme for the Listen-Before-Talk (LBT) mechanism on the performance of a state-augmented constrained reinforcement learning controller for NR-U/Wi-Fi coexistence. Coexistence control is formulated as a constrained Markov decision process with an explicit delay constraint for high-priority traffic and fairness as the optimization goal. Our simulation results show three key findings: (1) signed, threshold-invariant cost scaling with temporal smoothing stabilizes learning and strengthens long-term constraint adherence; (2) use of the contention window parameter for control provides smoother adaptation and better delay compliance than other MAC parameters; and (3) enhanced LBT significantly reduces collisions and improves airtime efficiency. These findings provide practical insights for achieving robust, QoS-aware coexistence control.

QoS-Aware State-Augmented Learnable Framework for 5G NR-U/Wi-Fi Coexistence: Impact of Parameter Selection and Enhanced Collision Resolution

TL;DR

The paper tackles QoS-aware coexistence of 5G NR-U and Wi-Fi in unlicensed bands by applying a state-augmented constrained RL controller (QaSAL) to dynamically tune MAC parameters. It examines cost scaling, violation modeling, and enhanced collision-resolution CR-LBT, alongside evaluating MAC parameters (CW,AIFSN,MCOT) for QoS and fairness. Key findings show that signed, threshold-invariant cost scaling stabilizes learning and constraint adherence; Contention Window control yields smoother delay compliance; and CR-LBT significantly reduces collisions and improves airtime efficiency, with CW-based control delivering the strongest QoS performance when paired with CR-LBT. The work also discusses deployment considerations, highlighting QaSAL’s relatively low per-step complexity versus primal-dual methods and outlining directions for multi-channel and PHY-parameter extensions.

Abstract

Unlicensed spectrum supports diverse traffic with stringent Quality-of-Service (QoS) requirements. In NR-U/Wi-Fi coexistence,the values of MAC parameters critically influence delay, collision behavior, and airtime fairness and efficiency. In this paper, we investigate the impact of (i) cost scaling and violation modeling, (ii) choice of MAC parameters, and (iii) an enhanced collision resolution scheme for the Listen-Before-Talk (LBT) mechanism on the performance of a state-augmented constrained reinforcement learning controller for NR-U/Wi-Fi coexistence. Coexistence control is formulated as a constrained Markov decision process with an explicit delay constraint for high-priority traffic and fairness as the optimization goal. Our simulation results show three key findings: (1) signed, threshold-invariant cost scaling with temporal smoothing stabilizes learning and strengthens long-term constraint adherence; (2) use of the contention window parameter for control provides smoother adaptation and better delay compliance than other MAC parameters; and (3) enhanced LBT significantly reduces collisions and improves airtime efficiency. These findings provide practical insights for achieving robust, QoS-aware coexistence control.

Paper Structure

This paper contains 16 sections, 11 equations, 6 figures, 2 tables.

Table of Contents

  1. Introduction
  2. System Model
  3. QaSAL Framework for CPM and Enhancements
  4. CPM as Constrained MDP
  5. State-Augmented RL for CPM and QaSAL
  6. Cost-Scaling and Violation Modeling
  7. Signed, Threshold-Invariant Scaling
  8. Consistent dual update
  9. Collision Resolution for 5G NR-U/Wi-Fi Coexistence
  10. Computational Complexity at Deployment
  11. Simulation Results
  12. Experimental Setup
  13. Effect of Violation Modeling and Cost Scaling
  14. Impact of Coexistence Parameter Selection
  15. Effect of Enhanced LBT Mechanism (CR-LBT)
  16. ...and 1 more sections

Figures (6)

  • Figure 1: Enhanced LBT in NR-U gNBs. Plain LBT transmits a reservation signal (RS) after backoff until the next slot, whereas CR-LBT Loginov:2021 replaces RS with short collision-resolution (CR) slots that reduce collisions while keeping slot alignment.
  • Figure 2: QaSAL without violation modeling and cost scaling; action is CW.
  • Figure 3: QaSAL with violation modeling and cost scaling; action is CW.
  • Figure 4: QaSAL with violation modeling and cost scaling; action is AIFSN.
  • Figure 5: QaSAL with violation modeling and cost scaling; action is MCOT.
  • ...and 1 more figures