Radar Signal Detection and Estimation Using Time-Frequency Distributions.
Abstract
This report addresses the issues related to the applications of time frequency distributions in radar signal detection and parameter estimation problems. The Wigner-Ville distribution, the ambiguity function, and the time-frequency correlation function are the time-frequency distributions considered. The analysis considers a number of alternative radar receiver structures based on discrete time-frequency distributions. Various forms of the discrete ambiguity function and the discrete Wigner-Ville distribution are studied and their properties analyzed. New interpolation formulas are developed. Derivations of receivers based on the discrete time-frequency distribution display the potential for aliasing. The effects of aliasing on the optimality of these receivers are analyzed. Sampling criteria to prevent aliasing are determined and schemes to reduce the sampling requirement are devised. Since digital devices are of fixed register length, the signals processed are quantized and the arithmetic performed is of finite precision. The effect of the related quantization error from these devices in discrete time-frequency distribution based receivers is analyzed. Issues concerning the effect of coefficient quantization due to either storage or in-place computation in the fast Fourier transform (FFT) are discussed and clarified. A bound on the mean-square magnitude of the discrete ambiguity function is developed. Simulations are utilized to verify and validate the results.
Document Details
- Document Type
- Technical Report
- Publication Date
- Oct 01, 1995
- Accession Number
- ADA304818
Entities
People
- Donald D. Welner
- Pramod Varshney
- Tzeta Tsao
Organizations
- Syracuse University