Hybrid Data Aggregation Technique in Wireless Sensor Network through Classification of Fruitful Messages

摘要

The wireless sensor network is spatially disseminated independent sensors nodes to monitor wild environment monitoring, fire detection in forest, and militia surveillance. The sensor nodes are deployed to monitor themselves to communicate with each other. The sensor node collects the data from the network is shared between all the sensor nodes. Each sensor nodes of interpretation to be forwarded to base station, probably via other intermediary nodes, before the base station process the received data. The data aggregation aspires to unite response from various sensors into an only message. Data aggregation method minimize number of transmission and amount of network traffic in network, which significantly reduce the energy utilization in each sensor nodes. The data aggregation is a technique which actions to improve the localized congestion problem and it attempts to gather useful information from the sensor neighboring the event. The cooperative nature of ant colony optimization can be used for aggregating the data through shortest path. And the Support Vector Machine based classification remove anomaly message, their by only fruitful message is aggregated. As a result the sensor nodes broadcast only the useful data to the end point in reducing the congestion problem, which sequentially improves the network life time.