摘要：This paper presented a feature vector based spatial differential correlation DOA estimation algorithm. The algorithm uses the matrix properties, structure contains only spatial differential matrix of the correlation signal, and its characteristic value eigenvalue decomposition vectors, the value of each feature vector signal corresponds to a group of related, since the set of feature vectors contain all of the information signal. Respectively, the respective feature vector signal group for matrix modeling, address relevant signal loss caused by the matrix rank, to achieve the relevant signal part of DOA estimates. The same number of array elements in the case of the modified algorithm can be estimated up to correlate signal, increasing the number of signals can be estimated to improve the processing capacity of the array element, especially when the signal to noise ratio is less than zero, still has a better estimate performance, and need not divide the array, pore size will not cause the array
