基于LTE-DRX的超时策略梯度估计算法  

Gradient estimation algorithm of timeout policy based on LTE-DRX

作　　者：宋捷[1] 李方伟[1] 朱江[1] 

机构地区：[1]重庆邮电大学,重庆400065

出　　处：《重庆邮电大学学报（自然科学版）》2013年第3期319-323,共5页Journal of Chongqing University of Posts and Telecommunications(Natural Science Edition)

基　　金：国家自然科学基金(61071116);重庆市发改委高技术产业项目(20091537)~~

摘　　要：针对长期演进(long term evolution,LTE)系统的非连续接收(discontinuous reception,DRX)终端省电机制中不适当的状态切换造成能量开销较大的缺陷,提出了一种基于超时策略的梯度估计算法。通过构建半马尔可夫控制模型将状态切换问题转换为策略优化问题,结合在线梯度估计从而得到状态切换时间的最优解。仿真实验表明,在不同数据到达率的情况下该算法在平均时延以及节能性上都有更好的表现。An adaptive terminal power-saving algorithm based on the timeout strategy is proposed for energy consumption caused by inappropriate state switching in the energy saving mechanism of long term evolution-discontinuous reception （LTE-DRX）. By establishing semi-Markov model, it is combined with the gradient estimation online so as to obtain the op- timal switching time. Simulation results show that the performance of the proposed algorithm has a great improvement in terms of average delay and power efficiency in the case of different data arrival rate.

关 键 词：LTE-DRX 动态电源管理 超时策略 半Markov控制过程 在线优化 

分 类 号：TN914.53[电子电信—通信与信息系统]