Mr. Saurav Subedi successfully defended his Ph.D. dissertation, “Sparsity-Aware Signal Processing for Passive Radar Systems,” on April 11, 2016.
Passive radars utilize signals of opportunity, making them attractive due to their energy efficiency, low cost, and covert operation. As they do not emit signals, passive radars are increasingly important in addressing frequency spectrum congestion. Saurav’s work primarily focuses on multi-static passive radars within distributed radar networks. From a signal processing perspective, these systems face several challenges surrounding the required high-resolution target localization and tracking while the signals have narrow bandwidth and low signal power. In addition, passive radars are operated in bistatic mode, and multi-static signal fusion requires efficient synchronization and communication between different transmitter and receiver pairs.
To enhance target tracking with minimum data communication traffic between passive radar systems, Saurav employed complex multi-task Bayesian compressive sensing (CMT-BCS) and Gaussian mixture probability hypothesis density (GMPHD) filter. CMT-BCS improves target tracking performance by leveraging statistical relationships between multiple measurements while being less sensitive to dictionary coherence. Meanwhile, GMPHD efficiently handles time-varying target counts, incorporating new target births, disappearances, missed detections, and false measurements. Additionally, he developed a Doppler-only tracking framework that achieves precise target tracking while significantly reducing data traffic within passive radar networks.
Saurav’s work earned him the Best Student Paper Award (Second Place) at the 2015 IEEE International Radar Conference for his paper “Group sparsity based multi-target tracking in multi-static passive radar systems using Doppler-only measurements.”
Beyond passive radar, he also contributed to accurate localization and tracking of passive RFID readers using received signal strength indicators (RSSI). His maximum likelihood-based method compensates for multipath propagation artifacts and non-isotropic antenna patterns, achieving improved localization accuracy over state-of-the-art methods.
Saurav’s Ph.D. research was supported in part by the Air Force Research Laboratory and the Office of Naval Research.