An Adaptive Beamforming Method Based on Klms Algorithm

In this paper, we present an adaptive beamforming method based on the kernel least mean square (KLMS) algorithm and its sparsification processing. The proposed method regresses the mapping from array signal to desired signal by applying the Least Mean Square (LMS) algorithm in a high-dimension feature space using kernel methods. This enables the array to adaptively steer toward the signal source direction while suppressing noise and interferences. To reduce the complexity of the algorithm, we employ a sparsification process which only adds a new sample into the kernel representation when it doesn't satisfy approximate linear dependence (ALD) condition. The performance of the algorithm is analyzed and its effectiveness in adaptive beamforming applications is demonstrated.