[1]  Algoet  P H, Cover T M.  A sandwich proof of the Shannon-McMillan-Breiman theorem.  Ann Probab,  1988, 16:  899--909

[2]  Barron A R.  The strong ergodic theorem for densities: generalized Shannon-McMillan-Breiman theorem.  Ann Probab, 1985, 13: 1292--1303

[3]  Breiman L.  The individual ergodic theorem of information theory. Ann Math Statist,  1957, 28:  809--811

[4]  Chung K L.  The ergodic theorem of information theory.  Ann Math Statist, 1961, 32:  612--614

[5]  Doob J L. Stochastic Processes.  New York: Wilely,  1953

[6]  Gray R M.  Entropy and Information Theory.  New York: Springer-Verlag, 1990

[7]  Liu  W, Yang W G. An extension of Shannon-McMillan theorem and some limit properties for onhomogeneous Markov chains.  Stochastic Process Appl, 1996, 61:  129--145

[8]  Liu W,  Yang  W G. The Markov approximation of the sequences of N-valued random variables and a class of small deviation theorems.  Stochastic Process Appl,  2000, 89:  117--130

[9]  McMillan B. The basic theorems of information theory.  Ann Math Statist, 1953, 24:  196--216

[10]  Shannon C. A mathematical theory of communication. Bell System Tech J,  1948,  27:  379--423, 623--656

[11]  Yang W G, Liu W.  The asymptotic equipartition property for $m$th-order nonhomogeneous Markov information sources. IEEE Trans Inform Theory, 2004, 50:  3326--3330