QoS aware Joint Radar Communication Optimization via Backtrack based Training Network

Joint Radar Communication (JRC) system achieves radar detection and communication transmission using a shared hardware platform, making it more suitable for integration, miniaturization, and efficient spectrum utilization compared to traditional standalone radar or communication devices. Due to the mutual interference between the two subsystems, there is growing interest in reducing negative impacts and optimizing the overall performance of JRC systems. Among various approaches, optimizing resource scheduling strategies within JRC systems to balance the performance of each subsystem in different environments and enhance system Quality of Services (QoS) values has become a significant research direction. In this paper, we propose a deep learning-based optimization framework tailored for JRC systems. By analyzing the current application environment of JRC systems, the proposed method dynamically adjusts the energy resource allocation strategy within the JRC system. This approach aims to balance the subsystems' performance concerning the current system environment and demands, achieving self-optimization to adapt to the current application scenarios. Simulation results indicates that our proposal could improved the JRC average performance under different application scenarios.