This paper presents a set of extensive experiments with alternative neural network learning algorithms. These neural network configurations were tested on the problem of discriminating signals generated by autoregressive moving-average ( ARMA) linear systems driven by white noise. These ARMA signals model a wide variety of signals arising in the ocean environment. We tested the various network models for their classification accuracy and speed of learning. The models investigated were back propagation, quickprop, Gaussian node networks, radial basis functions, the modified Kanerva method, and networks without hidden units. For comparison, nearest-neighbor classifiers were also tested. Classification performance and learning time results are presented.