SVM based data redundancy elimination for data aggregation in Wireless Sensor Networks

The data aggregation is most important in Wireless Sensor Networks (WSN) due to constraint of resources. There is lot of data redundancy in WSN due to dense deployment. So, it is necessary to minimize the data redundancy by adopting suitable aggregation techniques. To resolve this problem, Support Vector Machine (SVM) based Data Redundancy Elimination for Data Aggregation in WSN (SDRE) has been proposed in this work. First, we build aggregation tree for the given size of the sensor network. Then, SVM method was applied on tree to eliminate the redundant data. The Locality Sensitive Hashing (LSH) is used minimize the data redundancy and to eliminate the false data based on the similarity. The LSH codes are sent to the aggregation supervisor node. The aggregation supervisor finds sensor nodes that have the same data and selects only one sensor node among them to send actual data. The benefit of this approach is it minimizes the redundancy and to eliminate the false data to improve performance of WSN. The performance of proposed approach is measured using the network parameters such as Delay, Energy, Packet drops and Overheads. The SDRE perform better in all the scenarios for different size network and varying data rate.