Date: January 20, 2006

Time: 3:00 p.m.

Location: Centergy One 5186

Speaker(s): Martin Tobias

Title: Using the PHD for Multitarget, Mulitsensor Tracking with Passive Radar

Abstract:
Ronald Mahler's probability hypothesis density (PHD) provides a promising technique for the passive coherent location (PCL) of targets observed via multiple bistatic radar measurements. A particle filter implementation of the Bayesian PHD filter is used to locate targets with range and Doppler measurements from a receiver that exploits non-cooperative FM radio signals. The PHD filter is attractive for use in multitarget tracking, since it avoids the need for any target number prediction logic. The expected number of targets is simply provided by taking the integral of the PHD. Furthermore, the PHD filter provides an easy way to fuse multiple types of data observations, and it does not perform any explicit report-to-track data association. Instead, a peak extraction algorithm is required. In this talk, the results of applying the PHD particle filter to a simulated, realistic PCL scenario are presented, as well as the challenges involved in making the PHD filter a real-time solution for multitarget tracking with passive radar.

Bio:
Martin Tobias is a graduate student in the School of Electrical and Computer Engineering at the Georgia Institute of Technology. He received his undergraduate degree from Harvard University in 1999, and worked at start-ups IronBridge Networks and Sandburst Corp. before beginning graduate school at Georgia Tech in the fall of 2002. He received the M.S. degree from Georgia Tech in Dec. 2003 and is currently pursuing his Ph.D under the supervision of Dr. Aaron Lanterman.

Date:	January 20, 2006
Time:	3:00 p.m.
Location:	Centergy One 5186
Speaker(s):	Martin Tobias
Title:	Using the PHD for Multitarget, Mulitsensor Tracking with Passive Radar
Abstract:	Ronald Mahler's probability hypothesis density (PHD) provides a promising technique for the passive coherent location (PCL) of targets observed via multiple bistatic radar measurements. A particle filter implementation of the Bayesian PHD filter is used to locate targets with range and Doppler measurements from a receiver that exploits non-cooperative FM radio signals. The PHD filter is attractive for use in multitarget tracking, since it avoids the need for any target number prediction logic. The expected number of targets is simply provided by taking the integral of the PHD. Furthermore, the PHD filter provides an easy way to fuse multiple types of data observations, and it does not perform any explicit report-to-track data association. Instead, a peak extraction algorithm is required. In this talk, the results of applying the PHD particle filter to a simulated, realistic PCL scenario are presented, as well as the challenges involved in making the PHD filter a real-time solution for multitarget tracking with passive radar.
Bio:	Martin Tobias is a graduate student in the School of Electrical and Computer Engineering at the Georgia Institute of Technology. He received his undergraduate degree from Harvard University in 1999, and worked at start-ups IronBridge Networks and Sandburst Corp. before beginning graduate school at Georgia Tech in the fall of 2002. He received the M.S. degree from Georgia Tech in Dec. 2003 and is currently pursuing his Ph.D under the supervision of Dr. Aaron Lanterman.