摘要
设{Xn,n≥0}是字母集为S={1,2,…,m}的任意信源,其联合分布为p(x1,…,xn),利用相对于非齐次马氏信源熵密度偏差的概念,研究任意离散信源的极限性质,得到了一类用不等式表示的强极限定理(称之为强偏差定理).
Let {Xn, n≥0} be arbitrary information sources taking vlaues in the alphabet S={1, 2, … m} on the measurable space ( Ω,F) and with the joint distribution p(x1, … ,xn). A class of strong deviation theorems on arbitrary information sources are established by using the notion of relative entropy with respect to a non-homogeneous Markov information sources.
出处
《河北工业大学学报》
CAS
2003年第6期108-113,共6页
Journal of Hebei University of Technology
