Innovation Series: Advanced Science (ISSN 2938-9933, CNKI Indexed)

Beam Design and Cooperative Positioning Algorithm for Vehicle-mounted Communication and Sensing Integrated System

Zhuo Chen

School of Electronic and Information Engineering, University of Science and Technology Liaoning, Anshan, 114051, Liaoning, China

Corresponding Author: Zhuo Chen

Abstract: With the rapid development of intelligent connected vehicles (ICVs) and 6G vehicular networks, integrated sensing and communication (ISAC) has become the core enabling technology for high-level autonomous driving. However, the practical deployment of vehicular ISAC systems is severely restricted by three key challenges: insufficient dynamic beam adaptation in high-mobility scenarios, degraded cooperative localization accuracy in non-line-of-sight (NLOS) environments, and the difficulty of collaborative optimization for communication and sensing performance. To address these issues, this paper proposes a deep learning-based joint beam design and cooperative localization algorithm framework for vehicular ISAC systems. Specifically, we first develop a Transformer-based time-varying channel prediction and multi-objective beam optimization network, which realizes the joint optimization of communication spectral efficiency and sensing measurement accuracy. Then, a graph neural network (GNN)-driven multi-node cooperative localization model is designed to suppress NLOS errors using high-precision sensing parameters from optimized beams, with a closed-loop optimization mechanism between beam design and localization established. Simulation results show that compared with traditional separate benchmark algorithms, the proposed scheme improves communication spectral efficiency by 22.3%, sensing angle measurement accuracy by 31.5%, and reduces the root mean square error of vehicular cooperative localization by 46.8%, with end-to-end inference latency within 1ms. The proposed method meets the real-time and reliability requirements of high-dynamic vehicular scenarios, providing an effective technical solution for the engineering application of vehicular ISAC systems.

Keywords: Communication and sensing integration; Vehicle-mounted beam design; Cooperative positioning; Deep learning; Intelligent Connected Vehicles

References