Resolving the components of transient signals using neural network and subspace inhibition filter algorithms

The authors compare the use of a multilayer perceptron neural network with the use of a subspace inhibition filter algorithm for achieving enhanced local resolution of the arrival times and frequencies of the components of a transient signal. Both the neural network algorithm and the subspace inhibition filter algorithm provide enhanced local resolution of the arrival times and frequencies of the components of a transient signal. Both algorithms allow for detection of signal components in the subregions in the presence of interference from neighboring regions. Both algorithms allow for this detection without the restriction that signals be centered on the Gabor lattice. For comparison, a binary classification example has been constructed in which there is no prior knowledge that one transient signal in noise is present in the subregion. Probability of false alarm and probability of detection are computed for a series of test signals presented to both detectors. It is shown that the neural network is comparable in performance to the subspace inhibition filter algorithm