摘要

The algorithm applies weight adjustment to fullfill the classification for the correlative data with combination of the prediction. And the classification of the correlative data is mapped to the partition for the correlative sub-areas in detecting region. Representative nodes are selected to participate in data gathering in networks. Based on the temporal correlation of the sampling data, the state of the data transmission is controlled by AR(p) predictive model at the representative nodes. The effectiveness of the algorithm is testified by the simulations results.