Power Spectrum of Hard-Limited Gaussian Processes

In this paper we examine the power spectral density at the output of an ideal hard limiter when the input is driven by a collection of independent gaussian processes. This work is motivated by the fact that in spectrum analysis, it is often convenient from the point of view of signal processing to precede the analyzer with a hard limiter. In order to determine the effect of the limiter on analyzer performance, it is of interest to compare the power spectral density at the limiter output with that at the limiter input. With this goal in mind, the ideal limiter to be ana3031 3056 TI-IE BELL SYSTEM TECHNICAL J O U R N A L , NOVEMBER 1969 lyzed is shown in Fig. 1. It is assumed that the limiter input is driven by the signal *(t) y(t) Fig. 1 -- Ideal hard limiter.