认知无线网络中基于时间序列预测的冲突分解算法  

Collision resolution algorithm based on time-series forecasting for cognitive wireless networks

作　　者：杨双懋[1] 郭伟[1] 唐伟[1] 

机构地区：[1]电子科技大学通信抗干扰技术国家级重点实验室,四川成都611731

出　　处：《通信学报》2011年第11期51-58,共8页Journal on Communications

基　　金：国家科技重大专项(2010ZX03005-002);国家重点基础研究发展计划("973"计划)基金资助项目(2009CB320405)~~

摘　　要：针对认知无线网络提出了一种新的基于业务模型预测的冲突分解算法,该算法利用认知节点对用户业务的感知能力对业务进行建模,使用模型预测冲突报文数,同时结合树形分解算法来分解报文冲突。理论分析和仿真结果都表明,在真实的自相似业务背景下,该算法能提高认知网络的系统吞吐率和降低平均分解周期,从而使得系统的整体性能得到改善。A novel prediction-based collision resolution algorithm for self-similar traffic in cognitive wireless networks was formulated.The packet collision was resolved by combining the tree splitting scheme and the prediction theory.Through theoretical analysis and simulation experiments,the proposed prediction-based collision resolution algorithm performs better than the binary-tree splitting algorithm in terms of network throughput and collision resolution period.

关 键 词：认知网络 冲突分解 时间序列 媒体接入控制 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]