AoI-Driven Queue Management and Power Control in V2V Networks: A GNN-Enhanced MARL Approach

TL;DR

The paper addresses the challenge of maintaining fresh status updates in V2V networks under limited wireless resources by modeling multi-packet status as batched data and jointly optimizing packet dropping and transmit power. It introduces a GNN-enhanced MAPPO framework with CTDE, where GraphSAGE captures topology-induced interference, and agents operate with a hybrid discrete-continuous action space to manage queues and power. A novel AoI-focused training objective integrates an AoI-aware GNN embedding loss with a centralized critic to guide topology-aware decisions, achieving scalable coordination. Experiments across density, channel conditions, and traffic loads demonstrate significant AoI reductions (down to around 1.8 ms) and robust performance against packet length and arrival rate variations, highlighting practical benefits for cooperative perception in dynamic vehicular networks.

Abstract

Queue management and resource allocation play a critical role in enabling cooperative status awareness in vehicular networks. This paper investigates the problem of age of information (AoI)-aware status updates in vehicle-to-vehicle (V2V) communication, where each vehicle's status is represented by multiple interdependent packets. To enable fine-grained queue management at the packet level under resource constraints, we formulate a joint optimization problem that simultaneously learns active packet dropping and transmit power control strategies. A hybrid action space is designed to support both discrete dropping decisions and continuous power control. To exploit the graph-structured interference inherent in V2V topology, a graph neural network (GNN) is introduced to aggregate slowly varying large-scale fading, allowing agents to capture topological dependencies implicitly without frequent message exchange. The overall framework is built upon multi-agent proximal policy optimization (MAPPO), with centralized training and decentralized execution (CTDE). Simulations demonstrate that the proposed method significantly reduces average AoI across a wide range of network densities, channel conditions, and traffic loads, consistently outperforming several baselines.

Figures (10)

• Figure 1: Illustration of queueing model in a V2V communication scenario with batch size $u = 3$.
• Figure 2: Illustration of queue management with batch size $u = 3$, including packet replacement and dropping rules.
• Figure 3: The proposed GNN-enhanced CTDE framework based on MAPPO.
• Figure 4: Return for each training episode with increasing iterations.
• Figure 5: The average AoI for each training episode with increasing iterations.