Weak signals' DOA estimation based on RBFNN
Zhong Zi-fa · Jisuanji yingyong yanjiu · 2011
This paper proposed a novel intelligent technique for weak signals' DOA estimation in the presence of strong jamming or signal,which transfered the problem of DOA estimation into a large mount of data intelligent learning and recognition problem.Firstly,obtained the relationship between eigenvalue of the correlation matrix and signal power according to the signal subspace theory.Extracted the upper triangular half of the correlation matrix and the eigenvalue of the correlation matrix of knowing direction signals to form training set together.Then constructed the weak signals' DOA estimation model based on RBFNN.Compared with the other methods of weak signals' DOA estimation using algebra calculation,the proposed approach needn't to acquire the direction of the strong signals and attenuate them.In addition,the new method has less computing burden and higher estimation accuracy.The experiments demonstrate its effectiveness and feasibility.