Joint Resource Allocation for V2X Sensing and Communication Based on MADDPG

Abstract

Vehicle-to-Everything (V2X) communication is expected to play a critical role in enabling Intelligent Transportation Systems (ITS) within sixth-generation (6G) networks. Integrated Sensing and Communication (ISAC) technology is essential for enhancing spectrum efficiency and reducing resource overhead. However, this also demands a more intelligent and efficient resource allocation framework for next-generation vehicular networks. In this paper, we propose a joint resource allocation method for V2X communication and sensing, aiming to optimize both communication rate and sensing performance. We consider both communication and sensing, using the communication rate as the measure of communication and the Cramér-Rao Lower Bound (CRLB) as the measure of sensing accuracy. In addition, a reward function is designed based on the characteristics of the scenario. The power allocation is used as a continuous action space, and we employ the MultiAgent Deep Deterministic Policy Gradient (MADDPG) algorithm to solve this optimization problem to address the challenge of dynamic resource allocation. Simulation results show that the proposed method achieves joint optimization of communication and sensing resources across various scenarios, significantly improving the overall system performance. Compared with the PPO algorithm, the proposed algorithm can improve the communication rate by 60% and achieve the trade-off between communication and sensing performance.

Keywords

• Vehicle-to-everything (V2X)
• integrated sensing and communication (ISAC)
• resource allocation
• deep reinforcement learning