MODEL-BASED ONLINE LEARNING FOR RESOURCE SHARING IN JOINT RADAR-COMMUNICATION SYSTEMS

The ever-increasing congestion in the radio spectrum has made coexistence and co-design for radar and communication systems an important problem to address. The radio spectrum is a rapidly time-frequency-space varying resource, and learning is required to use the spectrum and mitigate the interference. This paper proposes a model-based online learning (MBOL) framework to enable a structured way to formulate efficient online learning algorithms for resource sharing in joint radar-communication (JRC) systems. As an example, we apply the MBOL framework for allocating frequency resources in non-cooperative shared spectrum scenarios. The proposed MBOL algorithm learns a predictive model using online convex optimization (OCO) and chooses the best frequency channels in uncertain interference environments. The algorithm outperforms the considered baseline algorithms in terms of regret that quantifies the cost of learning.