Deep reinforcement learning for time allocation and directional transmission in joint radar-communication

• Main Authors: Lee, Joash, Cheng, Yanyu, Niyato, Dusit, Guan, Yong Liang, Gonzalez, David G.
• Other Authors: School of Computer Science and Engineering
• Format: Conference or Workshop Item
• Language: English
• Published: 2022
• Subjects: Engineering::Computer science and engineering; Engineering::Electrical and electronic engineering::Wireless communication systems; Vehicle-to-Everything; Deep Reinforcement Learning; Resource Allocation; Joint Radar-Communication
• Online Access: https://hdl.handle.net/10356/155437
• Institution: Nanyang Technological University

Current strategies for joint radar-communication (JRC) rely on prior knowledge of the communication and radar systems within the vehicle network. In this paper, we propose a framework for intelligent vehicles to conduct JRC, with minimal prior knowledge, in an environment where surrounding vehicles execute radar detection periodically, which is typical in contemporary protocols. We introduce a metric on the usefulness of data to help the vehicle decide what, and to whom, data should be transmitted. The problem framework is cast as a Markov Decision Process (MDP). We show that deep reinforcement learning results in superior performance compared to non-learning algorithms. In addition, experimental results show that the trained deep reinforcement learning agents are robust to changes in the number of vehicles in the environment.