WSN Related Algorithm Research Based On Granular Computing

In recent years, with the rapid development of micro-electromechanical systems, low-power embedded technology, wireless communications, multi-sensor technology, wireless sensor network technology gradually into the horizon, and its low-cost, low-power, distributed and the characteristics of self-organization has brought a revolution in the field of information perception. Wireless sensor networks has become a popular and promising technologies which have broad applications in military, medical, aerospace and other fields. And it has attracted many researchers.Wireless sensor network data is a typical uncertain stream data, which contains a great deal of knowledge and useful information. The stream data has the characteristics of dynamic, real-time and fast which the common data do not have, and because of the wireless sensor network node resources are limited, during the study of wireless sensor stream data mining, it is necessary to fully consider the entire network of battery power, processing ability, storage capacity, communication bandwidth,lifetime and other factors. Therefore, extracting useful knowledge and information from a large number of incomplete, uncertain WSN data has become a new challenge in the field of data mining, which is also one of the new research topics.Granular computing is a new concept and new method in the field of artificial intelligence, mainly used for uncertain, imprecise, incomplete information processing, large-scale mining of massive data and complex problem solving. Therefore, according to the research and technology of wireless sensor network data mining development status in China and abroad, this paper studies two typical problem(cluster routing protocol and missing data tackle) in WSN data mining using granular computing method. The results show that the ideas and methods of granular computing for wireless sensor networks Data Mining have a good theoretical basis and practical value. The innovation and research results of the paper are as follows:1, Creating WSN granularity model based on quotient space, and proposing WSN dynamic topology quotient space cluster routing algorithm(QSRA), while based on the research of QSRA, WSN dynamic topology quotient space multi-hop cluster routing algorithm(QSRA-M) is proposed;2. The granular computing time series is useful for WSN stream data and missing data mining, by analyzing the time series and uncertain stream data it can better deal with WSN missing data forecast and imputing. The experiments show that the selection of a suitable granule is helpful to achieve better results.