ESTIMATION OF A SYSTEM PULSE TRANSFER FUNCTION IN THE PRESENCE OF NOISE

Accurate measurements of system characteristics from limited amounts of data are important in many applications. Some adaptive control systems carry out measurements of plant parameters during normal operation. In communication systems an analogous situation arises in the utilization of time-varying channels 1. Often the application of a special test signal is undersirable so the information must be obtained from ordinary in put output data. The presence of random noise and instrumentation errors can render many meas urements schemes ineffective. Statistical stimation theory provides powerful methods for dealing with this type of problem. Some of these methods are applied to the estimation of the pulse transfer function of a linear system.