Title :
Numerical study on joint quantization and control under block-coding
Author :
Tan, Xiaobo ; Xi, Wei ; Baras, John S.
Author_Institution :
Dept. of Electr. & Comput. Eng., Michigan State Univ., East Lansing, MI, USA
Abstract :
This paper addresses the joint quantization and control problem for hidden Markov chains with variable-length block-coding. The aim is to understand the impact of communication bandwidth and information delay (due to block-coding) on the control performance. A heuristic algorithm is developed to solve the dynamic programming (DP) equation through the introduction of a metric on the discrete observation space. Numerical results are presented demonstrating the attention division in block-coding and the tradeoffs between control performance and communication bandwidth.
Keywords :
block codes; discrete systems; dynamic programming; hidden Markov models; quantisation (signal); attention division; communication bandwidth; control performance; discrete observation space; dynamic programming equation; heuristic algorithm; hidden Markov chains; information delay; joint quantization and control problem; metric; numerical study; variable length block-coding; Bandwidth; Communication system control; Control systems; Delay; Dynamic programming; Equations; Heuristic algorithms; Hidden Markov models; Optimal control; Quantization;
Conference_Titel :
Decision and Control, 2004. CDC. 43rd IEEE Conference on
Conference_Location :
Nassau
Print_ISBN :
0-7803-8682-5
DOI :
10.1109/CDC.2004.1429462
