Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Sensing-Assisted Adaptive Channel Contention for Mobile Delay-Sensitive Communications

Bojie Lv, Qianren Li, Rui Wang

TL;DR

This work tackles delay-sensitive uplink scheduling in sensing-enabled mmWave networks by formulating a decentralized multi-agent MDP in which mobile agents locally map their state to back-off parameters. A centralized optimization of policy parameters $\mathbf{b}$ and $\bm{\lambda}$ is solved using an unbiased SGD gradient estimator that leverages environment and mobility sensing, including LoS blockage predictions, to minimize a discounted queueing cost. Key contributions include the P1 formulation, an analytical back-off policy model, unbiased gradient estimation via a single trajectory, a provable convergence guarantee for the SLPG algorithm, and simulations showing superior queuing performance and efficiency over SPSA and baseline policies. The approach delivers practical gains in reduced delays and buffer overflow in mobile mmWave uplink scenarios, by integrating sensing-based channel predictions into adaptive contention and scheduling.

Abstract

This paper proposes an adaptive channel contention mechanism to optimize the queuing performance of a distributed millimeter wave (mmWave) uplink system with the capability of environment and mobility sensing. The mobile agents determine their back-off timer parameters according to their local knowledge of the uplink queue lengths, channel quality, and future channel statistics, where the channel prediction relies on the environment and mobility sensing. The optimization of queuing performance with this adaptive channel contention mechanism is formulated as a decentralized multi-agent Markov decision process (MDP). Although the channel contention actions are determined locally at the mobile agents, the optimization of local channel contention policies of all mobile agents is conducted in a centralized manner according to the system statistics before the scheduling. In the solution, the local policies are approximated by analytical models, and the optimization of their parameters becomes a stochastic optimization problem along an adaptive Markov chain. An unbiased gradient estimation is proposed so that the local policies can be optimized efficiently via the stochastic gradient descent method. It is demonstrated by simulation that the proposed gradient estimation is significantly more efficient in optimization than the existing methods, e.g., simultaneous perturbation stochastic approximation (SPSA).

Sensing-Assisted Adaptive Channel Contention for Mobile Delay-Sensitive Communications

TL;DR

This work tackles delay-sensitive uplink scheduling in sensing-enabled mmWave networks by formulating a decentralized multi-agent MDP in which mobile agents locally map their state to back-off parameters. A centralized optimization of policy parameters and is solved using an unbiased SGD gradient estimator that leverages environment and mobility sensing, including LoS blockage predictions, to minimize a discounted queueing cost. Key contributions include the P1 formulation, an analytical back-off policy model, unbiased gradient estimation via a single trajectory, a provable convergence guarantee for the SLPG algorithm, and simulations showing superior queuing performance and efficiency over SPSA and baseline policies. The approach delivers practical gains in reduced delays and buffer overflow in mobile mmWave uplink scenarios, by integrating sensing-based channel predictions into adaptive contention and scheduling.

Abstract

This paper proposes an adaptive channel contention mechanism to optimize the queuing performance of a distributed millimeter wave (mmWave) uplink system with the capability of environment and mobility sensing. The mobile agents determine their back-off timer parameters according to their local knowledge of the uplink queue lengths, channel quality, and future channel statistics, where the channel prediction relies on the environment and mobility sensing. The optimization of queuing performance with this adaptive channel contention mechanism is formulated as a decentralized multi-agent Markov decision process (MDP). Although the channel contention actions are determined locally at the mobile agents, the optimization of local channel contention policies of all mobile agents is conducted in a centralized manner according to the system statistics before the scheduling. In the solution, the local policies are approximated by analytical models, and the optimization of their parameters becomes a stochastic optimization problem along an adaptive Markov chain. An unbiased gradient estimation is proposed so that the local policies can be optimized efficiently via the stochastic gradient descent method. It is demonstrated by simulation that the proposed gradient estimation is significantly more efficient in optimization than the existing methods, e.g., simultaneous perturbation stochastic approximation (SPSA).
Paper Structure (10 sections, 2 theorems, 14 equations, 5 figures, 1 table, 1 algorithm)

This paper contains 10 sections, 2 theorems, 14 equations, 5 figures, 1 table, 1 algorithm.

Table of Contents

  1. Introduction
  2. System Model
  3. Mobility Model
  4. Distributed Channel Contention Model
  5. Uplink Channel and Transmission Model
  6. System Queue Dynamics
  7. Proposed Distributed Channel Contention Framework and Problem Formulation
  8. Proposed SGD-BASED SOLUTION
  9. Simulations and Discussions
  10. Conclusions

Key Result

Lemma 1

Given the local scheduling parameters $(\mathbf{b}, \bm{\lambda})$, $\hat{\mathsf{g}}^{\texttt{b}}_{\kappa,\ell} \left( \mathbf{b}, \bm{\lambda}; \bm{\tau}^{\mathbf{b}, \bm{\lambda}}_{\texttt{MC} } \right )$ and $\hat{ \mathsf{g}}^{\lambda}_{\kappa,\ell} \left( \mathbf{b}, \bm{\lambda}; \bm

Figures (5)

  • Figure 1: Illustration of uplink communication scenarios: mobile agents are navigating in a room with furniture, pillars as signal blockers; the walls are the main scattering clusters.
  • Figure 2: Illustration of proposed framework: centralized optimization of distributed channel contention.
  • Figure 3: Indoor simulation scenario.
  • Figure 4: The empirical cumulative distribution function (CDF) of the total discounted cost in the indoor environment, where average arrival rate $\bar{A}_{k}=0.6$, $\forall k\in \mathcal{K}$.
  • Figure 5: Visualization of an agent's proposed back-off timer parameters corresponding to all local states, where average arrival rate $\bar{A}_{k}=0.6$, $\forall k\in \mathcal{K}$.

Theorems & Definitions (7)

  • Definition 1: Local and Global System State
  • Definition 2: Local Scheduling Policy
  • Remark 1
  • Lemma 1
  • proof
  • Theorem 1: Stochastic Convergence
  • proof