Research And Application Of DOA Estimation Algorithms Based On The Circular Signal

BPSK, AM, MASK and other similar signals are common in the communication system. The pseudo covariance matrix of such signals is not zero, so they are usually called non-circular signals. For the past few years, how to make use of non-circular signals of their own characteristics to improve the DOA estimation performance has become a research hotspot in the field of DOA estimation. It is equivalent to a virtual extension array, and the number of dealed sources can be more than the number of sensors of the array. What’s more, the estimation precision is improved compared with the traditional DOA estimation algorithm. So the DOA estimation algorithm based on the non-circular signals has broad prospects in more and more complicated modern electronmagnetic environment. In this paper, the methed of DOA estimation based on non-circular signals is researched deeply.Based on the traditional DOA estimation model, from the perspective of the incoming signal classification, when incoming signals are all non-circular signals, when incoming signals are mixed with circular and non-circular signals and when incoming signals are coherent non-circular signals, the research of the DOA estimation algorithm are discussed in this paper. NC-MUSIC, NC-ESPRIT, mixMUSIC, mix-ESPRIT and conjugate algorithms which can deal the coherent signals are included. From theory to simulation, it is demonstrated that adding the information of non-circular to DOA estimation not only extends the size of array, but also has more superior performance. Finally, a demodulation method of radar echo under the real background suited for decorrelation algorithm based on noncircular signals is presented. The conjugate algorithms are applicated here and the results consistent with the theory.