A New Blind Beamforming for Non-Gaussian Signals with Arbitrary Kurtosis

Hong-sheng Li,You He,Ri-jie Yang,Xiao-ming Tang
2006-01-01
Abstract:A new blind beamforming algorithm for non-Gaussian signals with arbitrary kurtosis is presented in this paper. Based on Kurtosis Maximization Algorithm (KMA), a new cost function is defined by introducing the cross-correlation of two signals. This method can estimate the weights of beamformers blindly by maximizing the new cost function, so as to separate the multitargets and find the direction. At the same time, a Genetic Algorithm (GA) with complex encoding is used to compute the weight vectors, which not only avoids the local extremum but also improves the computing speed. Simulation proves correctness of this algorithm.
What problem does this paper attempt to address?