Synthesis of Algorithms of Adaptive Signal Processing for Tracking Meters Using Nonlinear Blocks with Feed-Back

摘要

We considered the issues of the synthesis of algorithms for adaptive nonlinear signal processing using the feed-back blocks of nonlinear transformation under the influence of non-Gaussian broadband noise and band-pass noise with unknown density of distribution of instantaneous values or their envelopes. It is shown that probabilistic iterative algorithms and algorithms with measuring the coefficient of noise suppression can be used to plot the adaptive feed-back blocks of nonlinear transformation. Note, that to implement probabilistic iterative algorithms, the ability to differentiate the target function on settings is required. We proposed the algorithm of measuring the actual coefficient of noise suppression using the pilot signal in the blocks of nonlinear transformation with arbitrary amplitude characteristics. The conditions of consistency of the estimator of the mentioned coefficient are defined.