Representation and Estimation Techniques for Cyclostationary Random Processes
Abstract
Many communication and control systems employ signal formats that involve some form of periodic processing operation. Familiar examples are signals produced by samplers, scanners, multiplexers, or modulators. Very often these signals can be modelled as cyclostationary processes, i.e., processes whose statistical properties, such as mean and autocorrelation, fluctuate periodically with time. Filters designed to extract signals of this type from a noise background can exhibit dramatically improved performance when the periodic nature of the statistics are taken into account, rather than using the more conventional 'time-average' statistical approach. Some techniques for solving for the optimum filter and a video signal example are discussed.
Document Details
- Document Type
- Technical Report
- Publication Date
- Nov 01, 1972
- Accession Number
- AD0753132
Entities
People
- L. E. Franks
Organizations
- University of Massachusetts Amherst