A Data Aggregation Strategy Based on Confidence Interval in Wireless Sensor Networks
Feng Ren-jian · Beijing Youdian Xueyuan xuebao · 2009
For reducing energy consumption is essential for the life of wireless sensor networks,a confidence interval banding taboo-level of data aggregation protocol for long-lived sensor network is proposed.Considering that the intra-cluster data taking on Gaussian distribution characteristics,confidence interval banding taboo-level of data aggregation(CBTA) makes data sampling confidence interval compared with the total model,that can be effective in reducing the amount of data being sent.In order to avoid hot collected in cluster,taboo algorithm is used for sampling data to calculate the taboo-level.Compared with the previous related research,CBTA adopts new data sending model using spatial correlation of data,that will handle the energy efficiently better.Experiments show that CBTA can satisfy the desired accuracy given by system and the life of network is proved significantly lengthened.