Amplitude Comparison Direction Finding Method Based On Compressed Sensing

The conventional Nyquist sampling rate requires that the sampling rate must be at least double of the maximum frequency present in the signal. Unfortunately, in many cases it doesn't meet this requirement because of the large bandwidth. Recently, Donoho and Candes proposed Compressed Sensing theory, which applies the sparsity prior of the signals and can accurately reconstruct original signals from a small quantity of measurements, provided an appropriate optimized procedure. There have been two distinct major approaches to sparse recovery:Basis Pursuit and Orthogonal Matching Pursuit. In this thesis, Compressed Sensing theory is applied to the directions of arrival (DOA) estimation of multiple radar targets present in the mainlobe of a rotating antenna, by Exploiting the Amplitude Modulation.In most modern radar systems the target DOA is estimated by the monopulse technique. However, when multiple targets are present in the range-azimuth cell under test, the monopulse system provides an erroneous DOA measure, somewhere in the direction of the "power centroid" of all the targets, and it is not useful to track any of the targets in the antenna main beam. Some methods such as the Maximum Likelihood and Asymptotic Maximum Likelihood have been proposed to reduce the deleterious effects of target multiplicity. Although they can achieve the DOAs estimation and the accuracy is high, they can not provide an exactly DOA measure when the targets have the same Doppler frequency. The main contributions of this thesis are as follows:1. The Orthogonal Matching Pursuit algorithm suffers from the inter-atom interference (IAI), an adaptive IAI mitigation method in the OMP method was proposed. For instance, in the sparse channel estimation problem, a sensing dictionary is designed adaptively and posterior information is utilized efficiently to prevent false atoms from being selected due to serious IAI. Numeral experiments illustrate that the proposed algorithm can improve the ability of detecting the non-zeros entries and the channel estimation accuracy.2. In order to estimate the DOA of one single radar target, a direction finding method via beam scanning and sparse reconstruction was proposed. The proposed method utilizes the peak characteristic of antenna pattern and sparse property of received data. Unlike the conventional methods based on peak-searching and symmetric constraint, the sparse reconstruction algorithm requires less pulse and takes advantage of compressive sampling.3. In order to estimate the DOAs of multiple stationary targets in the same range-azimuth resolution cell using only one receiving channel, the Cramer-Rao Lower Bounds (CRLB) were derived. Then the DOAs estimation method based on the compressed sensing theory was proposed, the estimation of DOAs was viewed as the sparse vector reconstruction problem, the nonzero entries and their coordinate represent the targets amplitudes information and direction information respectively. BP algorithm was used to reconstruct the sparse vector.4. The tracing algorithm of moving targets was studied. Based on the DOAs estimation of stationary targets, the estimation of Doppler frequencies is step up. In order to estimate the Doppler frequencies, frequency extension to every angle is needed. Then the support of the sparse vector needed to be reconstructed contains the complex amplitudes, Doppler frequencies, and DOA of multiple targets information. When the Doppler frequencies are not well separated, the ML methods can not provid an exact estimation. However, the BP methods estimate the DOAs exactly.