The New Jersey Institute of Technology's
Electronic Theses & Dissertations Project

Title:	Learning of radar system for target detection
Author:	Jiang, Wei
View Online:	njit-etd2021-098 (xii, 67 pages ~ 1.8 MB pdf)
Department:	Department of Electrical and Computer Engineering
Degree:	Doctor of Philosophy
Program:	Electrical Engineering
Document Type:	Dissertation
Advisory Committee:	Haimovich, Alexander (Committee chair) Simeone, Osvaldo (Committee member) Abdi, Ali (Committee member) Kliewer, Joerg (Committee member) Michalopoulou, Eliza Zoi-Heleni (Committee member)
Date:	2021-08
Keywords:	Radar detector design Reinforcement learning Supervised learning Waveform constraints Waveform design
Availability:	Unrestricted
Abstract:	In this dissertation, the problem of data-driven joint design of transmitted waveform and detector in a radar system is addressed. Two novel learning-based approaches to waveform and detector design are proposed based on end-to-end training of the radar system. The first approach consists of alternating supervised training of the detector for a fixed waveform and reinforcement learning of the transmitter for a fixed detector. In the second approach, the transmitter and detector are trained simultaneously. Various operational waveform constraints, such as peak-to-average-power ratio (PAR) and spectral compatibility, are incorporated into the design. Unlike traditional radar design methods that rely on rigid mathematical models, it is shown that radar learning can be robustified to uncertainties about environment by training the detector with synthetic data generated from multiple statistical models of the environment. Theoretical considerations and results show that the proposed methods are capable of adapting the transmitted waveform to environmental conditions while satisfying design constraints.