Experiences

• 2021.01-2021.12
• Fast data‑driven adaptation of radar detection
• Proposed a two-stage learning procedure that enables detectors to adapt quickly to the current radar environment. Developed a deep transfer learning-based algorithm and a meta-learning-based algorithm by leveraging model-agnostic meta-learning for the design of fast adaptation of radar detection.

• 2019.01-2020.12
• Deep learning of waveform generation and detection for radar systems
• Formulated a radar system architecture based on the training of the detector and the transmitted waveform. Developed two end-to-end learning algorithms for detection and waveform generation. Extended learning algorithms to incorporate waveform constraints. Provided theoretical results that justify the use of reinforcement learning-based transmitter training.

• 2017.06-2018.12
• Waveform optimization in spectral congestion conditions
• Formulated the joint optimization problem of sensing and backhaul communication functions with spectrum sharing and transmitted waveform power constraints. Developed algorithms based on the block coordinate descent method coupled with the stochastic successive upper-bound minimization method/majorization-minimization and semidefinite programming relaxation.

• 2014.09-2017.05
• Direction of arrival by non‑coherent arrays
• Developed a closed-form expression for the Cramer-Rao bound of non-coherent direction of arrival estimation obtained under certain conditions. Developed closed-form expressions for the maximum likelihood estimates of various unknown parameters obtained from magnitude-only measurements. Derived a closed-form analytical expression for the hybrid Cramer-Rao bound of the spatial frequency difference between two sources in the presence of sensor location errors.