Blind Beamforming for Collaborative Array Processing in Sensor Networks
Abstract
The detection, localization, tracking, and identification of a single target by acoustical/seismic measured data are fairly well understood. Many of the methods considered in the SensIT program proven in various ways for a single target in an open-air environment, will not be applicable to multiple targets. In the proposal, we advocated a new algorithm based on an efficient computational Approximate Maximum-Likelihood (AML) method using alternate projection to tackle the multiple target cases. The idea is that instead of performing the AML search in high dimensions for M targets, we first perform the ML estimate for the strongest target, then by fixing that target, we perform the ML estimate for the second strongest target, until the M-th target, and then iterate with the first target again.
Document Details
- Document Type
- Technical Report
- Publication Date
- May 01, 2004
- Accession Number
- ADA424487
Entities
People
- Kung Yao
Organizations
- University of California, Los Angeles