物联网环境信息协同感知与处理研究

The rapid development of Internet of Things (IoT) and the expansion of its application area, enable it to sample more kinds and more amount of data from its surroundings. The current methods to process big data cannot be directly applied to find specific information sources, such as source of pollution or lost property, among the big and complex data from IoT. The project adopts the ways of collaborative sensing and networked process to decrease the complexity of data processing, and to improve the quality of information source monitoring.In details, the project researches the following three aspects. Firstly, in various applications of IoT, each sampled data contains relatively simple information so that it is hard to take advantage of the information relevance among the data from different sensors. Therefore, the project will analyze the information relevance among the data from different sensors to decrease the complexity of the information processing by adopting the collaborative sensing based on the improved system structure of sensing layer. Secondly, each single data contains relatively low information value when the amount of data becomes big. Meanwhile, the data packet loss cannot be avoided during sensing and transmission. Therefore, the project adopts the method of networked processing to design high efficient and real-time strategy to monitor information source by the spatio-temporal information relevance. Thirdly，when the scale of IoT and the amount of its data become huge, the project analyzes the scalability of the method of collaborative sensing. The project will achieve new theoretical and technical methods for the environmental information processing, and expand the application area of IoT, and improve the service quality of IoT.

物联网的快速发展和应用范畴不断扩展，使其从环境中采集的信息种类变多，数据量变大。要从庞杂数据中高效地监测特定信息源，如污染源、遗失物，现有大数据处理方法难以应对。本项目采用信息协同感知及网络化处理方式，降低数据处理复杂度，提高信息源监测效率。具体从三个方面开展研究：1.针对物联网各类应用系统采集信息相对单一，难以充分挖掘信息关联性，本项目拟采用信息协同感知方式，通过改变物联网感知层的系统架构，使各类信息间建立关联性，从而降低信息处理复杂度；2.针对庞杂数据中单数据信息含量低，及信息传感时存在丢包，本项目采用网络化处理的方式，利用信息的时空关联性，设计高效实时的信息源监测策略；3.当物联网规模及其数据量庞大时，本项目研究了信息协同感知与处理的可扩展性。本项目的研究将为物联网环境信息处理提供新理论和方法，拓展其应用范畴，提高其服务质量。

物联网信息系统感知与网络化处理是具有很高实际应用价值的重要研究方向。我们项目组针对物联网各类应用系统采集信息相对单一缺乏关联性挖掘、采集数据信息含量低、网络规模大等问题开展了深入研究。项目组取得了多项研究成果，包括信息协同感知及分布式处理的机制、城市环境信息（交通信息）的采集与处理方法、利用信息的时空关联性设计高效实时的信息源监测策略、基于城市公交自行车网络的资源共享方法;构建了基于RFID、传感器网络、多媒体网络的物联网信息采集与处理示范系统。. 截止2019年2月1日，项目组已经在国内外期刊和学术会议上发表论文27篇，其中SCI论文15篇及EI收录9篇，国际一流刊物IEEE/ACM Transactions及JSAC等上发表论文6篇,国际一流会议IEEE INFOCOM/SECON/ACM MOBIHOC发表学术论文3篇。申请国家发明专利17项，其中授权2项，获得软件著作权12项。2名教师晋升为教授，2名教师晋升为副教授，1人入选浙江省中青年学科带头人。培养了一批研究生并获得国家奖学金等多项荣誉。