作者: Geert Rombouts , Marc Moonen
DOI: 10.1016/S0165-1684(03)00107-5
关键词: Signal 、 Median filter 、 Mathematics 、 Speech enhancement 、 QR decomposition 、 Adaptive filter 、 Speech processing 、 Distortion 、 Algorithm 、 Noise reduction
摘要: We describe a new adaptive filtering algorithm based upon QR-decomposition for optimal multichannel with an "unknown" desired signal, as well its application to multi-channel acoustic noise reduction. A recursively calculated filter then optimally estimates the speech component in noisy signal. The complexity of this is about 7 times lower than that existing related algorithms, which are mainly GSVD-decompositions, while performance kept at same level. Finally, it shown how parameter can be introduced tuned tradeoff reduction signal distortion.