Sparsity Aware Adaptive Radar Sensor Imaging in Complex Scattering Environments
Abstract
In this reporting period, we develop new radar imaging, estimation, and waveform encoding techniques that exploit prior knowledge of the target and its environment to improve system performance through sensing, learning, and exploitation. Our research accomplishment is three fold. First, we develop a variational Bayesian based framework to address the problem of multi-parameter estimation under compound Gaussian clutter in the context of cognitive radar. Results demonstrate an accelerated convergence of the proposed sequential estimation method with an improved asymptotic Cramer Rao bound compared with the conventional expectation-maximization (EM) method and the classic Bayesian approach, especially under small sample size. Second, we develop estimation method for range and Doppler using weighted OFDM waveforms for radar targets. We demonstrates that the proposed weighted OFDM modulation scheme results in a lower Cramer-Rao bounds for delay estimation compared with the classic constant-envelope OFDM modulation while meeting the requirement on the peak to average power ratio. Third, we study impact of waveform encoding on nonlinear electromagnetic tomographic imaging algorithms using multiple simultaneous excitation sources. By numerical simulations, we show that the proposed iterative image reconstruction algorithm using coded multiple source excitation achieves faster convergence and better quality images than the conventional single source excitation imaging.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 15, 2015
- Accession Number
- AD1000124
Entities
People
- Yuanwei Jin
Organizations
- University of Maryland Eastern Shore