泛在协同环境下最大似然估计时间同步算法  

Time synchronization algorithm based on maximum likelihood estimation in ubiquitous cooperative environment

作　　者：张晖[1] 赵鹏[1] 

机构地区：[1]南京邮电大学江苏省无线通信重点实验室,南京210003

出　　处：《仪器仪表学报》2016年第10期2373-2381,共9页Chinese Journal of Scientific Instrument

基　　金：国家自然科学基金(61471203);江苏省"青蓝工程"(2016);南京邮电大学1311人才计划(2015);江苏省计算机信息处理技术重点实验室开放课题(KJS1518)项目资助

摘　　要：泛在接入环境下,多个终端相互协同,共同为用户提供个性化的业务保障,而时间同步系多终端协同的关键技术之一,终端间时间同步与否直接决定了基于多终端协同的各类应用的性能优劣。本文提出了一种基于最大似然估计的新型时间同步算法,通过结合多终端协同场景特点,建立单网络最大似然估计模型,并提出多网络同步参数融合模型,以实现一般情况下目标终端之间的精准同步;进一步地,分析了时间样本子集正交和共有邻居节点稀疏等情况,提出相应的解决办法,以实现特殊情况下目标终端之间的精准同步。仿真结果验证了该算法的有效性,实测数据亦证明了该算法的实用性。In ubiquitous access environment,multiple terminals cooperate with each other to provide personalized services for users. Obviously,as the key technology of multi-terminal cooperation,time synchronization determines whether the corresponding services perform better or not.In this paper,a novel time synchronization algorithm based on maximum likelihood estimation is proposed. According to the characteristics of multi-terminal cooperation scenario,a maximum likelihood estimation model for single network and a fusion model of synchronization parameters for multi-networks are established respectively,so as to achieve the accurate time synchronization among target terminals in general case.Furthermore,two special cases （i.e.time sample subsets are orthogonal and mutual neighbor nodes are sparse）are analyzed,and then the corresponding solving methods are proposed,so as to achieve the accurate time synchronization among target terminals in special case.Simulation results verify the effectiveness of the proposed algorithm,and actual measured data also verify its practicality.

关 键 词：泛在接入环境 多终端协同场景 时间同步 最大似然估计 

分 类 号：TN915[电子电信—通信与信息系统] TH124[电子电信—信息与通信工程]