摘要: Let X and Y be two random variables with probability distribution p(x,y), joint entropy H(X,Y) conditional entropies H(X \ Y) H(Y X) . Person P/sub x/ knows person Y/ Y. They communicate over a noiseless two-way channel so that both know It is proved that, on the average, at least + bits must exchanged 2 are sufficient. If p(x.y) > 0 for all (x.y), then communicated average. However, if p (x,y) uniform its support set, average number of needed close to H (Y X). Randomized protocols can reduce amount communication considerably but only when some error acceptable.
stanford.edu 
sci-hub.se 参考文章(8)
CE Shennon, Warren Weaver, A mathematical theory of communication Bell System Technical Journal. ,vol. 27, pp. 379- 423 ,(1948) , 10.1002/J.1538-7305.1948.TB01338.X
Andrew Chi-Chih Yao, Some complexity questions related to distributive computing(Preliminary Report) symposium on the theory of computing. pp. 209- 213 ,(1979) , 10.1145/800135.804414
Andrew Chi-Chin Yao, Probabilistic computations: Toward a unified measure of complexity 18th Annual Symposium on Foundations of Computer Science (sfcs 1977). pp. 222- 227 ,(1977) , 10.1109/SFCS.1977.24
David Huffman, A Method for the Construction of Minimum-Redundancy Codes Proceedings of the IRE. ,vol. 40, pp. 1098- 1101 ,(1952) , 10.1109/JRPROC.1952.273898
D. Slepian, J. Wolf, Noiseless coding of correlated information sources IEEE Transactions on Information Theory. ,vol. 19, pp. 471- 480 ,(1973) , 10.1109/TIT.1973.1055037
Alon Orlitsky, Abbas El Gamal, Communication with secrecy constraints symposium on the theory of computing. pp. 217- 224 ,(1984) , 10.1145/800057.808684
Robert G. Gallager, Information Theory and Reliable Communication ,(1968)
T. Cover, A proof of the data compression theorem of Slepian and Wolf for ergodic sources (Corresp.) IEEE Transactions on Information Theory. ,vol. 21, pp. 226- 228 ,(1975) , 10.1109/TIT.1975.1055356