ABSTRACT
This chapter begins by discussing the amplitude distributions of random signals and discusses power spectral estimates. There are two basic sources of the most commonly used statistical properties: the amplitude distribution of the random signal and the autocorrelation function or, equivalently, the power density spectrum. A stationary signal is a signal with local statistics that are invariant over the entire duration of the signal. Thus, a periodic signal is a stationary signal, but a transient signal that occurs locally in a long time domain is not stationary. One could argue that no process in the natural world is truly stationary. But, there are many situations in engineering where the assumption of stationarity can be very useful. The power spectrum and the amplitude distribution together contain the basic signal statistics that are normally required for analysis, filtering, signal compression, and other types of processing. The chapter discusses a method for estimating the power spectrum of a stationary random sequence.
