Practical Neural Network Recipies in C++

Evaluating Performance of Neural Networks

359

Signal-to-Noise Ratio

One application of neural networks is filtering noisy signals. Some

aspects of this problem are discussed in Chapter 3, Autoassociation,

especially on page 24 and following. This section will present a

popular means of assessing the effectiveness of filters that are

designed to separate a signal from noise. Note that this generalizes

beyond the autoassociative filters discussed earlier. Other applications

will be discussed at the end of this section.

We often are faced with a problem in which a message signal,

v(t),

is contaminated with random noise, n(t). In the simplest case ...