Adaptive signal processing

We describe a novel modular learning strategy for detection of a target signal of interest in a nonstationary environment, which is motivated by the information preservation rule. The strategy makes no assumptions on the environment. It incorporates three functional blocks: time-frequency analysis, feature extractions, and pattern classification, the delineations of which are guided by the information preservation rule. The time-frequency analysis implemented using Wigner-Ville distribution, transforms incoming received signal into a time-frequency image and accounts for the time-varying nature of the received signal's spectral context. This image provides a common input to a pair of channels, one of which is adaptively matched to interference acting alone, and the other is adaptively matched to target signal plus interference. Each channel of the receiver consists of a principal components analyser (for feature extraction) followed by a multilayer perceptron (for feature classification), which are implemented using self-organized and supervised forms of learning in feedforward neural networks, respectively.