Study On DOA Estimation Algorithm Based On Non-circular Signals

Spatial Spectrum is the signal's energy distribution in all directions in space,when it is estimated the DOA (Direction of Arrival) will be estimated. Spatial Spectrum Estimation is one of the most important research fields of Array Signal Processing, it is also be called DOA Estimation or Direction Finding, and has been used in many civilian and military fields include radar, communication, sonar, earthquake, and biomedical engineering and so on. In recent years,along with the extensive using of the Non-circular Signals (such as AM,PAM,BPSK, MASK, ASK and so on) in communication fields, the DOA estimation techniques based on Non-circular Signals have been the research highlights in Direction Finding fields.In this paper, the main works include:At first, introduces the research status and development history of Spatial Spectrum Estimation, explains both the classical and new methods of coherent signal's DOA estimation in detail, then provides the MUSIC algorithm and ESPRIT algorithm which based on Non-circular Signals.Secondly, it researches the DOA estimation method of Non-circular Signals based on Minimum Redundancy Linear Array (MRL). MRL has the advantages of array aperture is bigger and resolving power is higher. To estimated DOA of Non-circular Signals with non-circular features by MRL, which could enhance the utilization efficiency of array elements and gain bigger array aperture just with fewer array elements number. It also could avoid the phenomenon of grating lobe in visible fields and reduce cross coupling largely,which makes the array get a favorable DOA estimation robustness.Thirdly, it researches the DOA estimation method of Non-circular Coherent Signals based on Eigen-space (ES-DOA) algorithm. To solve the problem of solutions of the coherent of Non-circular Coherent Signals, a new algorithm of Eigen-space is given. It reconstructs the covariance matrix whose information has been doubled, and uses Eigen-space algorithm to solute coherent, which could take advantage of the information from signal subspace and noise subspace and avoid the array aperture loss made by Spatial Smoothing algorithm and the complex calculation made by Maximum Likelihood algorithm. Besides, this algorithm could also estimate signal's power, enhances the success probability for the estimation of lower energy signal.At last, this paper provides a new Maximum Eigenvalue algorithm for the estimation of Non-circular Coherent Signals. It takes the advantage of the linear combination relationship between covariance matrix and steering vector to reconstruct the covariance matrix,decomposes the new covariance matrix and searches spectral peak with MUSIC algorithm.This method could solute coherent just use the information from the biggest Eigenvalue vector, which could reduce the calculation amount and also get better stability and effectiveness.