Variants of neural network Learning models for network energy Optimization of internet of things in Wireless sensor networks

Abstract

Wireless Sensor Networks (WSN) are a prominent area in the growth of technology possessing significant potential in the cloud based internet of things. There exist few open problems that have to be handled so as to effectively operate and get the most benefit of this WSN technology. Over the decades it has been well noted that a foremost challenge is the energy consumption with respect to network utilization. Numerous algorithms have been developed over the years to save the energy consumed and one of the most hopeful alternate algorithms are the multi-objective evolutionary computational intelligent algorithms. These computational intelligent algorithms tend to provide a set of solutions and results in achieving the required rate of energy minimization. In respect of this, several algorithms have been developed over the past years for optimizing the energy consumed in the wireless sensor network models. newlineThe main contribution in this thesis work is to model and develop novel variants of neural network architectural model for improving the sensor network performance and as well to carry out optimization of network overhead that is present between the cloud storage space and the wireless sensor network model. Also, the developed neural network variants aim to optimize the network energy utilization by selecting the optimal nodes in the wireless sensor environment. newlineThe developments made in this thesis are on the third generation neural networks and their applicability for increasing the network lifetime and thereby to transmit data with better accuracy rate in WSN model. The significant research findings made in this thesis are as given below newline