Title :
Optimal quantization for distributed estimation via a multiple access channel
Author :
Duman, Tolga M. ; Salehi, Masoud
Author_Institution :
Dept. of Electr. & Comput. Eng., Northeastern Univ., Boston, MA, USA
Abstract :
Quantizer design algorithms for decentralized estimation are presented. Scalar quantizer design for the problem of multiple descriptions over a multiple-access channel is also studied. The importance of the initial index assignment is explained and an algorithm (based on simulated annealing) for choosing a good initial index assignment is derived
Keywords :
estimation theory; multi-access systems; quantisation (signal); simulated annealing; telecommunication channels; decentralized estimation; distributed estimation; initial index assignment; multiple access channel; multiple-access channel; optimal quantization; quantizer design algorithms; scalar quantizer design; simulated annealing; Algorithm design and analysis; Distortion measurement; Entropy; Lagrangian functions; Particle measurements; Quantization; Rate-distortion; Sensor fusion; Statistics; Transmitters;
Conference_Titel :
Information Theory, 1995. Proceedings., 1995 IEEE International Symposium on
Conference_Location :
Whistler, BC
Print_ISBN :
0-7803-2453-6
DOI :
10.1109/ISIT.1995.550427