DocumentCode
1028407
Title
Constructions of particular random processes
Author
Johnson, Glenn Eric
Author_Institution
Anal. Sci. Corp., Reston, VA, USA
Volume
82
Issue
2
fYear
1994
fDate
2/1/1994 12:00:00 AM
Firstpage
270
Lastpage
285
Abstract
This paper reviews how to construct sets of random numbers with particular amplitude distributions and correlations among values. These constructions support both high-fidelity Monte Carlo simulation and analytic design studies. A variety of constructions is presented to free engineering models from “white or normal” limitations embodied in many current simulations. The methods support constructions of conventional stationary and normally distributed processes, nonstationary, nonnormal signal and interference waveforms, nonhomogeneous random scenes, nonhomogeneous volumetric clutter realizations, and snapshots of randomly evolving, nonhomogeneous scenes. Each case will have specified amplitude statistics, e.g., normal, log-normal, uniform, Weibull, or discrete amplitude statistics; and selected correlation, e.g., white, pink, or patchy statistics, clouds. or speckles. Sets of random numbers with correlation, nonstationarities, various tails for the amplitude distributions, and multimodal distributions can be constructed. The paper emphasizes aspects of probability theory necessary to engineering modeling
Keywords
correlation methods; probability; random number generation; random processes; signal processing; amplitude distributions; amplitude statistics; analytic design; correlations; engineering models; high-fidelity Monte Carlo simulation; interference waveforms; multimodal distributions; nonhomogeneous random scenes; nonhomogeneous volumetric clutter; nonnormal signal waveforms; nonstationary processes; normally distributed processes; probability theory; random number set construction; randomly evolving nonhomogeneous scenes; stationary processes; Clouds; Control systems; Detectors; Layout; Probability distribution; Random number generation; Random processes; Signal processing; Statistical distributions; Statistics;
fLanguage
English
Journal_Title
Proceedings of the IEEE
Publisher
ieee
ISSN
0018-9219
Type
jour
DOI
10.1109/5.265353
Filename
265353
Link To Document