作者: K. Gerlach
DOI: 10.1109/7.272281
关键词: Electronic engineering 、 Rate of convergence 、 Probability distribution 、 Sample matrix inversion 、 Noise power 、 Covariance matrix 、 Active noise control 、 Control theory 、 Noise reduction 、 Mathematics 、 Gaussian noise
摘要: Convergence results for the sample matrix inversion (SMI) canceler algorithm in nonstationary noise are presented. Exact given convergence rate of average output power residue normalized to quiescent a two-input (one auxiliary), and lower upper bounds derived cancelers with two or more inputs under assumption that there is no internal noise. These function number independent samples processed per channel (main auxiliary input channels, external environment. The environment modeled as single interfering source conditionally Gaussian, level specified at each sampling time instant. Furthermore, this model generalized sense joint probability distribution defined levels over processing batch. This leads capability modeling evaluating SMI variety interference scenarios. >