Title :
Information integrity in lossy source coding with side information
Author :
Graves, E. ; Wong, T.F.
Author_Institution :
Army Res. Lab., Adelphi, MD, USA
fDate :
Sept. 30 2014-Oct. 3 2014
Abstract :
In this paper we investigate how to achieve information integrity in two common models of source coding with side information. We first look at the classical Wyner-Ziv problem, where the side information is available only to the decoder, with an adversary whom may arbitrarily change the “bin” index sent to the decoder. Interestingly, it is easy to show that standard Wyner-Ziv coding suffices to guarantee no adversary attack strategy may achieve arbitrarily small probability of fooling the decoder if the side information is correlated with the source. Subsequently, we consider the case where the side information is available also at the encoder, and hence joint source encoding can be performed. Here we show the result of Wyner-Ziv problem also extends to joint encoding without needing the side information to be correlated with the source. Moreover, no loss in the optimal compression rate is incurred in order to guarantee information integrity in either case.
Keywords :
codecs; data compression; source coding; Wyner-Ziv problem; Ziv problem; bin index; decoder; information integrity; joint source encoding; lossy source coding; optimal compression rate; side information; standard Wyner-Ziv coding; Decoding; Indexes; Random variables; Rate-distortion; Source coding; Zinc;
Conference_Titel :
Communication, Control, and Computing (Allerton), 2014 52nd Annual Allerton Conference on
Conference_Location :
Monticello, IL
DOI :
10.1109/ALLERTON.2014.7028601