Synesthesia of Machines (SoM)-Enhanced ISAC Precoding for Vehicular Networks with Double Dynamics

TL;DR

This paper introduces a novel deep reinforcement learning-based precoding method for vehicular ISAC networks that effectively adapts to double dynamics, outperforming traditional approaches in real-time scenarios.

Contribution

It proposes a synesthesia of machine-enhanced precoding framework using DRL and a parameter-shared actor-critic architecture for improved vehicular ISAC performance.

Findings

Outperforms existing precoding methods in dynamic vehicular environments.

Reduces training time with a parameter-shared actor-critic architecture.

Effectively adapts to double dynamic scenarios with high accuracy.

Abstract

Integrated sensing and communication (ISAC) technology is vital for vehicular networks, yet the time-varying communication channels and rapid movement of targets present significant challenges for real-time precoding design. Traditional optimization-based methods are computationally complex and depend on perfect prior information, which is often unavailable in double-dynamic scenarios. In this paper, we propose a synesthesia of machine (SoM)-enhanced precoding paradigm that leverages modalities such as positioning and channel information to adapt to these dynamics. Utilizing a deep reinforcement learning (DRL) framework, our approach pushes ISAC performance boundaries. We also introduce a parameter-shared actor-critic architecture to accelerate training in complex state and action spaces. Extensive experiments validate the superiority of our method over existing approaches.