Study On The DOA Estimation For Coherent Signal Identification

The DOA(Direction-of-Arrival) estimation of coherent signals is a research hotspot and important research subject of array signal processing ,it is also one of the basic task of radar, sonar communication areas and so on. In order to solve the problem of the DOA estimation of coherent signals, researchers all over the world proposed a lot of algorithms, all the algorithms can be departed two types: the first type can decrease the relativity of coherent signals by sacrificing the number of arrays, that is preprocessing the coherent signals then estimate the precise DOA with the common algorithms such as the spatial smoothing, forward backward prediction projection and data matrix decomposition algorithm.. The other type needn't decrease the aperture of the arrays, which shift the arrays or make use of frequency smoothing to decrease the relativity. Aim at the former, the author proposes an improved weighted spatial smoothing algorithm for DOA of coherent signals on the basis of others' results which can fully utilize the correlations of the array output and produce a more stable estimate of the covariance matrix. In the common sense, the resolution probability is higher compared with the common algorithm, especially when the SNR(signal noise ratio) is lower. There are a lot of simulations in this paper, simulation results are provided to verify the theoretical prediction.