Association And Implementation Of Wireless Multimedia Sensor Network

Today's cyberspace contains a lot of objects and information of man-machine-things in ternary space.Wireless multimedia sensors and the data created by them have become an important part of these objects in cyberspace.A lot of correlations between different sensors exists in the mountain of data and cannot be got before processed.With computing power of wireless multimedia sensor networks(WMSNs),it is possible to analyse and associate sensors and data created by them in wireless multimedia sensor networks(WMSNs)before transmission which can improve the efficiency of resources in WMSNs.It also can mining the association rules from sensor data which can make correlations in cyberspace.The large size of data and the demand of real-time make it hard to mining association rules by traditional mining algorithm.It is necessary to design a appropriate method to mining association in WMSNs.In this paper,we pay attention to associating sensors of WMSNs.The main idea of the paper can divide into three part:· We design a novel wireless multimedia sensor network system and give an introduction to different parts and mechanisms of the system.This system is the base of our study.· We analyse the spatial correlation between wireless sensors and use correlation of sensor data to represent it.Based on the spatial correlation,we design a distributed algorithm to cluster sensors which can run on sensors.This algorithm cluster sensors with strong spatial correlation together.· Based on the cluster algorithm,we learn the association rule between sensors.In order to mining the association rules from sensing data,we propose a new data structure named SWMVPT to store sensing data and mining association rules from the data structure.