Print Email Facebook Twitter Approximately Optimal Radar Resource Allocation Title Approximately Optimal Radar Resource Allocation: Joint Radar Communication for Automotive Applications Author Jayachandra, Karan (TU Delft Electrical Engineering, Mathematics and Computer Science; TU Delft Microwave Sensing, Signals & Systems) Contributor Schöpe, M.I. (mentor) Yarovoy, Alexander (mentor) Degree granting institution Delft University of Technology Corporate name Delft University of Technology Date 2021-07-12 Abstract Due to the rising number of wireless device users, it is expected that there will be a scarcity in the spectrum. The will especially true for the Automotive Spectrum between 77 and 81 GHz. In this thesis, we apply Sensor Management to the Joint Radar Communication scenario. We develop an algorithm that can allocate resources to both sensing and communication tasks. The resources are across both frequency bands and time division. The proposed solution minimizes the global uncertainty while avoiding interference. The solution is formulated via an optimization problem and can perform a comparison of communication and sensing costs side-by-side. The sensing resources are allocated using a cost minimization and the communication strategy is found by formulating a selection problem. The solution demonstrates that cooperation can lead to higher accuracy for all sensors involved across multiple targets while using resources more efficiently. The algorithm is able to allocate resources to multiple sensors in a non-myopic framework. A proof-of-concept for a distributed version of the algorithm is also shown. Subject Radar Resource ManagementAutomotive radarPOMDPConstrained Optimization To reference this document use: http://resolver.tudelft.nl/uuid:a6d05da8-6b93-4e6c-9a3e-2499ec66f122 Embargo date 2023-07-12 Part of collection Student theses Document type master thesis Rights © 2021 Karan Jayachandra Files PDF Thesis_Report_Karan.pdf 3.84 MB Close viewer /islandora/object/uuid:a6d05da8-6b93-4e6c-9a3e-2499ec66f122/datastream/OBJ/view