TARGET DETECTION IN ULTRA-WIDEBAND NOISE RADAR SYSTEMS
Open Access
- Author:
- Kwon, Yangsoo
- Graduate Program:
- Electrical Engineering
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- August 30, 2012
- Committee Members:
- Ram Mohan Narayanan, Dissertation Advisor/Co-Advisor
Ram Mohan Narayanan, Committee Chair/Co-Chair
William Evan Higgins, Committee Member
Mark Levi, Committee Member
Vishal Monga, Committee Member - Keywords:
- noise radar
target detection
MIMO radar - Abstract:
- Ultra-wideband (UWB) noise radar has been widely considered as a promising technique for covert high-resolution detection of multiple targets due to several advantages such as excellent immunity from jamming and interference, low probability of detection and interception, and relatively simple hardware architectures. The most important advantages of noise radar is immunity of interception by an adversary, since the transmitted noise-waveform is constantly varying and never repeats exactly. In addition, UWB noise radar obtains high-resolution detection of multiple targets due to its high instantaneous bandwidth. Based to the above advantages, UWB noise radar systems are garnering more and more attention recently. In this dissertation, two issues concerning detection and estimation based on information theory and compressive sensing applied to UWB noise radar are investigated in detail. This dissertation explores a target detection method using the total correlation formalism based on information theory which enables the detection of multiple targets at intermediate and low signal-to-noise ratio (SNR) regimes. This approach uses the largest eigenvalue of the sample covariance matrix to extract information from the transmitted signal replica, and outperforms the conventional total correlation detector when reflected signals have intermediate or low SNR values. Additionally, in order to avoid ambiguous target occurrence, an adaptive threshold is proposed, which guarantees the detection performance with the same receiving antenna elements for a given false alarm probability. The threshold is computed from the largest and smallest eigenvalue distributions based on random matrix theory. Numerical simulations show this detection method can be used for a wide range of SNR environments, and the threshold provides definitive target detection. This dissertation also explores an application of compressive sensing for multipleinput multiple-output (MIMO) UWB noise radar imaging. Two schemes to improve the system performance of a sample selection and an adaptive weighting allocation are investigated. The sample selection is based on comparing the norm values of candidates among the received signal, and selecting the largest M samples among N per antenna to obtain selection diversity. Moreover, an adaptive weighting allocation which improves reconstruction accuracy of compressive sensing by maximizing the mutual information between target echoes and the transmitted signals is investigated. Further, this weighting scheme is applicable to both sample selection schemes, a conventional random sampling and the proposed selection. Simulations show that this selection method enhances the multiple target detection probability and reduces the normalized mean square error (NMSE). This dissertation also develops the adaptive weighting allocation of compressive sensing for MIMO UWB noise radar imaging from a practical perspective. For an adaptive weighting allocation scheme in the previous chapter, however, perfect knowledge of the target scene is not available in practice due to the existence of noise at receivers or lack of measurements. Thus, the recovery error provides inaccurate weighting values which can degrade reconstruction accuracies and target detection probabilities. In order to mitigate this problem, an adaptive weighting allocation with reconstruction error by using knowledge of recovery error variances is investigated. Numerical simulations for various scenarios show that this practical scheme improves NMSE and the target detection probability.