摘要
研究一类多阶段动态规划问题,给出了求最优解的方法.将这种多阶段动态规划应用于Turbo译码中,不仅可以减少运算量,还可以避免传统的Turbo译码算法需要进行指数运算以及其随着迭代次数的增加容易出现的数据溢出问题,因此是一种十分有效的方法,是对系统工程理论应用领域的拓宽.
This paper deals with a particular variety dynamic programming with multi phases and a method resolving this problem is given in this paper. The application result in decoding turbo code shows that this method can not only simplify computation, but also avoid data overflow caused by exponential operation and iteratlve operation with more times in traditional turbo decoding. So the method given in this paper is effective for decoding turbo code, which generalized the applied field of system engineering theory.
出处
《数学的实践与认识》
CSCD
北大核心
2008年第19期84-88,共5页
Mathematics in Practice and Theory
基金
国家社科基金项目资助(07XJL002)
国家自然科学基金(70672052)
西安交通大学"985二期"07基地项目资助
关键词
多阶段动态规划
最优解
决策
效益
dynamic programming with multi phase
optimal solution
decision
benefit
