Spectral Estimation of Continuous Time Processes Using Poisson Distributed Time Samples
Abstract
Report compares two discrete time spectral estimation techniques for estimating the spectral density function of a continuous time process - periodic spectral estimation and Poisson spectral estimation. It shows the Poisson to excel when the spectral density is not bandlimited; when the shape of the spectral density is not known a priori; and when the sampling rate is constrained to be below that of Nyquist. The Poisson theory applies to other areas of signal processing. The Poisson technique could be used in applications such as cross spectral estimation, estimation of the coherence function, and antenna array beamforming.
Document Details
- Document Type
- Technical Report
- Publication Date
- Apr 07, 1978
- Accession Number
- ADA054903
Entities
People
- C. Mirabile
- D. Klamer