Research on Event-driven Data Collection Protocol of Wireless Sensor Networks

Abstract

This paper carries out research on two types of event-driven data collection protocols of wireless sensor networks: EDDGP and ACBP protocols. In the EDDGP protocol, an adaptive evaluation model is put forward to select the cluster head, including the node classification, the fuzzy grade classification, the evaluation function and the parent grid resolution. Once an event of interest occurs, the member nodes will send the collected information to the cluster heads which will then integrate the data and send a request packet for data collection to the sink. After receiving the request packet, the sink will quickly move to the cluster head for collecting information. If the sink receives the request packet for data collection from other cluster heads of information source in the moving process, it will save their grid ID. The data from the other cluster heads of information source are collected in turn from the nearest ones to the farthest ones. The theoretical research and simulation results show that EDDGP can reduce the energy consumption of location updates of the sink so as to balance the energy consumption of nodes and prolong the lifetime of the network.