A novel denoising method for acoustic target classification in wild environment

The acoustic recognition technology in wireless sensor surveillance network in wild environment is facing the challenge of the complicated and strong acoustic noise, especially the wind noise. Kernel Independent Component Analysis (KICA) is a non-linear method for blind source separation (BSS) technology which was wildly used in signal preprocessing. Considering the high computational complexity of KICA, an improved KICA algorithm is proposed based on the Renyi quadratic entropy estimator. A series of simulation experiment show that the improved KICA algorithm can well maintain the separating performance while reduce the computational complexity of KICA and the algorithm could be well utilized in denoising for the target classification system.