Research on Performance Evaluation Strategy of Communication and Perception Fusion in Intelligent Connected Vehicles
Huanyi Liu · 2025
As the core of the future transportation system, the integration performance of communication and perception of intelligent connected vehicles directly affects driving safety and efficiency. However, the current evaluation system has challenges such as an imperfect index system, a single method, and insufficient dynamic optimization, making it difficult to meet the requirements of real-time performance and accuracy in complex scenarios. In response to these problems, this paper proposes a dynamic evaluation strategy based on 5G-V2X (5th Generation Vehicle-to-Everything) and multi-sensor fusion. Through methods such as dynamic time slot scheduling, MIMO (Multiple-Input Multiple-Output) beamforming technology and high-resolution range-Doppler analysis, resource allocation is optimized, communication delay is reduced and sensing accuracy is improved. The simulation experiment results show that this strategy reduces the 90 % quantile delay from 75 ms in the fixed-weight scheme to 53 ms, improving the performance by more than 20 %. Meanwhile, it can still maintain a detection probability of more than 90 % under adverse conditions such as rain and fog. This research provides an effective solution for the collaborative optimization of communication and perception in intelligent connected vehicles, and lays the foundation for the integration of future technologies such as 6G communication and edge computing collaboration.