Title :
Design of quantizers for decentralized estimation systems
Author :
Lam, Wai-Man ; Reibman, Amy R.
Author_Institution :
Thomson Consumer Electron., Indianapolis, IN, USA
fDate :
11/1/1993 12:00:00 AM
Abstract :
The authors consider parameter estimation in decentralized systems with distributed processors. They restrict the local processors to be quantizers and consider the optimal design of the systems to minimize the estimation error. They present necessary conditions for the optimal system based on the Bayes distortion functions and Fisher´s information. The numerical results compare the resulting quantizers obtained by different distortion criteria
Keywords :
Bayes methods; distributed processing; information theory; parameter estimation; sensor fusion; vector quantisation; Bayes distortion functions; Fisher´s information; decentralized estimation systems; distributed processors; distributed sensors; estimation error; local processors; necessary conditions; optimal design; optimal system; quantizers design; Communications Society; Estimation error; Iterative algorithms; Parameter estimation; Probability density function; Radar applications; Radar tracking; Sensor fusion; Sensor systems; Target tracking;
Journal_Title :
Communications, IEEE Transactions on
