In clustering approach the sensor nodes are grouped to form a cluster. The nodes of a clustering network have low powered battery capability and limited processing capabilities. These nodes continuously exchange the data to cluster head which is in turn transforming the data to its base station. Few of these nodes in network may be faulty or may not support life time processing data due to its low power battery. All these sensor nodes measure the temperature, humidity, sound and pollution from environment and collected data is send to cloud for further processing. The fault tolerance mechanism of these nodes is solved by applying genetic algorithm by implementing chromosome technique to identify and avoid fault nodes in the network. This proposed research work increases detection of fault nodes in a network, increase network efficiency, lifetime and reach energy optimization results in Internet of Things (IoT) concept. The performance evaluation shows that the data accuracy in Genetic Algorithm (GA) is higher when compared with Direct Diffusion (DD) Algorithm and Ad-hoc on demand Distance Vector (AODV) Algorithm.
Read full abstract