报告时间：9月20日（周四）上午10:30

报告地点：电气学院A101     

报告题目：Distributed Estimation inResource-limited Sensor Networks    

报告人：朱善迎 副教授 上海交通大学  

报告人简介：

   朱善迎，上海交通大学自动化系副教授。2013年获上海交通大学工学博士学位。2013年至2015年，在新加坡南洋理工大学以及伯克利新加坡教育联盟(BEARS)开展博士后研究工作。主要研究领域为无线网络中的分布式估计、优化和多智能体协调控制及在工业中的应用。在IEEE系列会刊、Automatica，CDC、IFAC-WC，ICASSP，GLOBECOM等国际知名期刊和会议发表论文50余篇，合作出版英文专著一部。获得2016年度教育部自然科学一等奖和2017年度上海市技术发明一等奖。目前为IEEE会员，IEEE-IES工业信息学技术委员会委员，中国自动化学会青工委委员，TCCT多自主体控制学组委员，并担任多个国际学术会议的会务共同主席和技术委员会委员。

摘要：

    Recent advances insensing and wireless technologies have enabled rapid development of distributedsensor networks and heir wide applications both in military and civil areas.Due to the large size of these networks and the presence of geographicallyspread resources, distributed information processing and optimizationtechniques are gaining prominence. In a typical estimation problem in ad hocnetworks, nodes make noisy measurements of variables of interest. The mainconcern is how to utilize the samples to produce a desired estimate by onlyexchanging data between neighboring nodes. A key challenge in such distributedprocessing involves the lack of global model information at the local nodelevel. Moreover, the systems under consideration are generally resource-limitedand suffer from communication constraints. In this talk, I will introduce twonew distributed estimation algorithms based on consensus strategies and examinethe impacts of channel fading and quantization on these algorithms,respectively. In addition, rigorous theoretical analysis regarding theunbiasedness and consistency properties of the proposed algorithms will beprovided. Some possible future research topics will also be envisioned in theend of the talk.