Speaker: Volkan Cevher

Title: Acoustic Node Calibration

Date: January 16, 2004

Time: 3:00 pm

Location: GCATT Room 325

Abstract:
An acoustic node is a collection of omnidirectional microphones whose relative positions are known with respect to each other. More than one stationary node is needed to track a target in the x-y space due to the observability of the target position vector given the acoustic signals. Hence, the node positions as well as their orientations should be estimated before doing any tracking or localization. We first present the maximum-likelihood (ML) estimator for an acoustic observation model used commonly in the literature. The calibration is based on using a calibrating target carrying a global positioning system (GPS) on board. The effects of the GPS errors on the node position estimates are
considered next. The related Cramer-Rao lower bound is derived. We then present a Newton based search algorithm and different initializations of it to avoid calculating the whole ML surface. One of these initialization methods is based upon calculating the node position and orientation using a synthetic aperture created by the calibrating target. Simulations are done to demonstrate the accuracy of the methods proposed.

Biography:
Volkan CEVHER was born in Ankara, Turkey, in August 1978. He received his B.S. Degree in Electrical Engineering from Bilkent University, Ankara, Turkey in 1999. He is currently studying towards the Ph.D. Degree in Electrical Engineering in Georgia Institute of Technology under the supervision of James H. McClellan. His current research interests include Monte Carlo Markov chain methods, target tracking models, adaptive filters, time frequency distributions, fractional Fourier transform, and array signal processing.

Slides:

Speaker:	Volkan Cevher
Title:	Acoustic Node Calibration
Date:	January 16, 2004
Time:	3:00 pm
Location:	GCATT Room 325
Abstract:	An acoustic node is a collection of omnidirectional microphones whose relative positions are known with respect to each other. More than one stationary node is needed to track a target in the x-y space due to the observability of the target position vector given the acoustic signals. Hence, the node positions as well as their orientations should be estimated before doing any tracking or localization. We first present the maximum-likelihood (ML) estimator for an acoustic observation model used commonly in the literature. The calibration is based on using a calibrating target carrying a global positioning system (GPS) on board. The effects of the GPS errors on the node position estimates are considered next. The related Cramer-Rao lower bound is derived. We then present a Newton based search algorithm and different initializations of it to avoid calculating the whole ML surface. One of these initialization methods is based upon calculating the node position and orientation using a synthetic aperture created by the calibrating target. Simulations are done to demonstrate the accuracy of the methods proposed.
Biography:	Volkan CEVHER was born in Ankara, Turkey, in August 1978. He received his B.S. Degree in Electrical Engineering from Bilkent University, Ankara, Turkey in 1999. He is currently studying towards the Ph.D. Degree in Electrical Engineering in Georgia Institute of Technology under the supervision of James H. McClellan. His current research interests include Monte Carlo Markov chain methods, target tracking models, adaptive filters, time frequency distributions, fractional Fourier transform, and array signal processing.
Slides: