Abstract：{Xn, n≥0}was assumed to be a sequence of random variables taken in finite alphabet set, and to be an asymptotic circular Markov chain and a homogeneous Markov chain under a probability measure P and a probability measure Q, respectively. Divergence rate plays an important role for entropy, relative entropy and mutual information. The divergence rate of asymptotic circular Markov chains was investigated. The divergence rate for asymptotic circular Markov chains related to homogeneous Markov chains was achieved by AEP for asymptotic circular Markov chains and uniform integrability of random variable sequences. As corollary, a known result was generalized.
2012, Vol. 33 
