Asymptotic Approximation to the Entropy Rate of Binary Hidden Markov Processes
Shuangping Chen, Haoran Zheng · Dianzi xuebao · 2006
Based on the convergence of bounds for the entropy rate of binary hidden Markov processes,a numerical approach is advanced.The algorithm can approximate to true value of the entropy rate below a predefined error,and the accuracy can also be estimated.Since the logarithm of the algorithm′s complexity is linear to the logarithm of error,the cost of the algorithm is acceptable for practical use in engineering fields.It casts light to solve the problem of computing the entropy rate of more generalized kinds of HMPs.