Developing an optimal MapReduce function for wireless sensor networks in a secure way.

In this research work, security risk in wireless sensor networks is addressed and an efficient approach to overcome security issues and to reduce the energy depletion in WSN was proposed. Here, sensors are grouped into three different clusters and cluster head is elected for each on the basis of sensing ability, battery power and memory. The nodes find the shortest path, and then inform the network to send the sensed data to CH in an encrypted form in a peer to peer manner. The CH will decrypt and aggregate the data received from the cluster before sending it to the BS to reduce the energy consumption and CH will generate a signature by using three types of keys to ensure the authenticity. The communication from CH to BS is broadcast. Now, the BS verifies the sign and the large dataset (observed, historical) are processed parallel in map reduce paradigm for query processing.