Blind Time-Frequency Analysis for Source Discrimination in Multisensor Array Processing
Abstract
The research efforts over 1998/1999 phase focused on nonstationary signals and the applications of time-frequency signal representations to multi-antenna receivers. We have clearly demonstrated, through analysis and simulations, the offerings of time-frequency distributions in solving key problems in sensor array processing, including direction finding, source discrimination, and blind signal recovery. Significant progress has been made in understanding how the time-frequency signatures of the signals impinging on an array of sensors can be utilized to improve array performance. The major contributions over the fiscal year ending September 30th 1999 are: 1) Regress analysis of eignestructure methods employing time-frequency distributions which has demonstrated their superiority over those used in conventional high resolution methods based on data covariance matrices, 2) Evaluating the performance of the spatial time-frequency distributions in low SNR and coherent signal environments, and introducing maximum likelihood techniques based on the sources' time-frequency signatures, 3) Identifying the role of time-frequency cross-terms in spatial signal processing, and presenting a prudent way to use those terms for enhanced performance, 4) Providing a novel approach for incorporating the revolutionary power spectra into source discrimination and angle of arrival estimation, and examining their advantages over quadratic time-frequency distributions. We have also made significant advances in the solutions of the problem of suppressing nonstationary jammers in broadband signal environment, using both projection techniques and frequency diversity methods. In the following, we summarize the above key contributions, leaving the details to the appendix, which also includes other contributions we have made from the research funding of this project.
Document Details
- Document Type
- Technical Report
- Publication Date
- Oct 01, 1999
- Accession Number
- ADA368941
Entities
People
- Moeness Amin
Organizations
- Villanova University