Research On Two-Dimensional Direction Of Arrival Estimation Algorithms Of Noncircular Signals

Array signal processing is a process that enhance the useful signal, suppress the unwanted interference and noise, and pick up the useful signal characteristics by processing the received signal of sensor array. A hot issue of array signals processing is the estimation of Direction of Arrival(DOA).Actually, noncircular signals are extensively used in communication systems, e.g., amplitude modulation(AM) signals and binary phase shift keying(BPSK). In the field of array signal processing,the noncircular signal has attracted a great attention, and the noncircular feature has been widely used to enhance the DOA estimation performance.In this paper,we deeply research the 2D-DOA of noncircular signal for uniform planar array. The main work of this paper is listed as follows:1) a two-dimensional(2D) direction of arrival(DOA) estimation algorithm of Noncircular Signals(NC) for uniform planar array via Estimating Signal Parameters via Rotational Invariance Techniques(ESPRIT) is investigated and the expression of Cramér-Rao Bound(CRB) is derived which provids criterion for a variety of non-circular signal algorithms. Compared to conventional ESPRIT algorithms, the proposed algorithms exploits the property of noncircular signals to double the array aperture. Therefore, the angle estimation performance of our algorithm is superior to that of the ESPRIT. In addition,the proposed algorithm achieves automatically paired two-dimensional angle estimates and works well without spectral peak searching.2) a two-dimensional direction of arrival estimation algorithm based on noncircular- propagator(NC-PM) method for uniform planar array is investigated.The proposed algorithm expand the array aperture through the data expansion of noncircular signal. Therefore, the angle estimation performance of our algorithm is superior to that of the Propagator Method(PM).And,the proposed algorithm achieves automatically paired two-dimensional angle estimates and works well without spectral peak searching.. Simulation results verify the effectiveness and improvement of the proposed algorithms. We also analyzed the complexity and performance of NC-ESPRIT and NC-PM algorithm.3) 2D-DOA estimation algorithm for uniform rectangular array via noncircular-parallel factor(NC-PARAFAC) method is proposed.Based on the traditional parallel factor(PARAFAC) algorithm,NC-PARAFAC double array aperture by exploiting the property of noncircular signals,thus the estimation accuracy is improved. In addition, NC-PARAFAC has better 2D-DOA estimation performance than some conventional algorithms, which include ESPRIT, PM, NC-ESPRIT, and NC-PM.4) The problem of 2D-DOA estimation for two parallel linear array is discussed, and derives Noncircular Signals Propagator Method Based On Euler transformation(Euler-NC-PM).The proposed algorithm expand the array aperture through the data expansion of noncircular signal. Therefore, the angle estimation performance of our algorithm is superior to that of the PM. At the same time, the algorithm, which uses the Euler transformation to convert plural arithmetic of NC-PM to real arithmetic, reduces computational complexity, and its angle estimation performance is close to NC-PM algorithm.