Introduction to neural networks and adaptive filtering: three illustrative examples

Increasingly, the applications of signal processing and neural networks are gaining developmental momentum and application within the scientific and engineering communities. The current paper provides a quick review of the history, and the definition of neural networks in general. It also provides an overview of the current methodologies involved and the essential mathematical formulation for three different applications and their implementations. Numerical solutions for the design of an adaptive FIR filter using the LMS algorithm are presented. Such filters can be used in many digital signal processing applications. They include: (1) up-sampling, (2) noise canceling, and (3) system identification