An Approach to Simulation of Stationary and Non-stationary Processes in the Harmonic Basis

Abstract

Environment simulation is a part of the development and research of various real-time systems. Environmental influences are modeled as stationary and non-stationary processes. Useful signals are commonly simulated as stationary processes. Simulation of non-stationary processes is required to represent signal-noise mixtures, fluctuations, turbulence. The authors propose to simulate a non-stationary stochastic process in the spectral area with a modification of the algorithm for simulating stationary processes. A feature of this approach is the discrete random signal representation in the harmonic basis using canonical expansions. The authors propose a change in stationarity conditions to obtain an algorithm for simulating non-stationary signals in the trigonometric basis. The described approach allows a simple transition from the stationary process simulation to the simulation of non-stationary processes in the same basis.