Blind Time-Frequency Analysis for Source Discrimination in Multisensor Array Processing
Abstract
This report includes results on the applications of time frequency distributions in blind source separation and direction finding problems. A novel approach based on time-frequency distributions (TFDs) for separating signals received by a multiple antenna array is developed. The sources have different time-frequency signatures and are instantaneously mixed at the array sensors. The proposed approach provides a significant improvement in performance over the recently introduced spatial time-frequency distributions, specifically for signals with close time-frequency signatures. Spatial averaging of the TFDs of the sensor data is performed to eliminate the interactions of the sources signals in the time-frequency domain, and as such restores important properties of the source TFD matrix. We have also improved nonstationary source signal estimation by performing the blind source separation using ambiguity functions, and as such, avoid the inclusion of cross-terms in the estimation process. Our third contribution to this area is the introduction of the Time-Frequency MUSIC as a new array signal processing method based on time-frequency signal representations. This report also includes contributions to problems of fast computation TFDs, spatial processing for frequency diversity spread spectrum communications as applied to partial jamming mitigation, and subband array processing to combat fading and multipath.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 1998
- Accession Number
- ADA354093
Entities
People
- Moeness Amin
Organizations
- Villanova University