Title :
Lossy distributed source coding using graphs
Author_Institution :
Dept. of Electr. Eng. & Comput. Sci., Univ. of Michigan, Ann Arbor, MI
fDate :
4/1/2009 12:00:00 AM
Abstract :
In this letter, lossy distributed source coding using graphs is considered. This corresponds to the source coding part of the graph-based framework for transmission of analog correlated sources over the multiple-access channel (MAC). Consequently, it is shown that a pair of analog correlated sources can be reliably represented into a bipartite graph by allowing certain amount of distortion. An achievable rate-distortion region for this problem is also provided. Therefore, it can be concluded that, for transmission of any (both discrete and continuous) set of correlated sources over MACs, graphs can be used as discrete interface between source coding and channel coding modules.
Keywords :
access protocols; channel coding; graph theory; source coding; MAC; bipartite graph; channel coding; distributed source coding; multiple-access channel; rate-distortion region; Bipartite graph; Broadcasting; Channel coding; Decoding; Maintenance; Performance loss; Probability distribution; Rate-distortion; Source coding; Transmitters; Distributed source coding, correlated messages, random graphs, multiple-access channels.;
Journal_Title :
Communications Letters, IEEE
DOI :
10.1109/LCOMM.2009.081825