Jointly Optimizing Terahertz based Sensing and Communications in Vehicular Networks: A Dynamic Graph Neural Network Approach

TL;DR

The paper tackles THz-enabled joint sensing and communications in vehicular networks by formulating a topology-sensitive service-mode optimization that maximizes the number of successfully served vehicles. It introduces a dynamic graph neural network that selects among multiple node and layer aggregation functions to adapt to varying vehicle topologies, enabling joint optimization of SPV service modes and their served service-request vehicles. The approach yields substantial gains over fixed-aggregation GNNs and non-GNN optimization baselines, achieving up to 17–28% improvements and near-optimal performance in dynamic urban scenarios. By leveraging topology-aware features and dynamic aggregation, the method enhances spectral efficiency, link reliability, and interference management in THz vehicular ISAC systems, with practical applicability in fast-changing V2X environments where blockages and activity states vary.

Abstract

In this paper, the problem of vehicle service mode selection (sensing, communication, or both) and vehicle connections within terahertz (THz) enabled joint sensing and communications over vehicular networks is studied. The considered network consists of several service provider vehicles (SPVs) that can provide: 1) only sensing service, 2) only communication service, and 3) both services, sensing service request vehicles, and communication service request vehicles. Based on the vehicle network topology and their service accessibility, SPVs strategically select service request vehicles to provide sensing, communication, or both services. This problem is formulated as an optimization problem, aiming to maximize the number of successfully served vehicles by jointly determining the service mode of each SPV and its associated vehicles. To solve this problem, we propose a dynamic graph neural network (GNN) model that selects appropriate graph information aggregation functions according to the vehicle network topology, thus extracting more vehicle network information compared to traditional static GNNs that use fixed aggregation functions for different vehicle network topologies. Using the extracted vehicle network information, the service mode of each SPV and its served service request vehicles will be determined. Simulation results show that the proposed dynamic GNN based method can improve the number of successfully served vehicles by up to 17% and 28% compared to a GNN based algorithm with a fixed neural network model and a conventional optimization algorithm without using GNNs.

Figures (17)

• Figure 1: Illustration of the considered vehicular network model.
• Figure 2: Visualization of node feature.
• Figure 3: Neighbor sampling process of vehicle $v$.
• Figure 4: Structure of the proposed dynamic GNN model.
• Figure 5: Key explanations to the node and layer aggregation functions.