Agent-Based Improved Neuro-Fuzzy for Load Balancing in Sensor Cloud

Abstract

Sensor cloud paradigm is a trending area for most of the applications. It collects the information from physical sensors and stores it in cloud servers, and it can be accessed anywhere. Energy optimization is one of the crucial issues in sensor cloud as sensed information are unprocessed and directly saved into cloud server thereby increasing energy consumption and delay which leads to unbalancing in the network. In this paper, agent-based improved neuro-fuzzy optimization is proposed to avoid transmission of redundant information into cloud along with load balancing among all nodes for equal energy consumption. The agents work on behalf of node, migrate to each node in the cluster, collect information, and submit to CH minimizing node energy consumption. Neuro-fuzzy along with weights is used to improve information accuracy and reducing energy consumption to improve overall network lifetime. Result shows that less energy is consumed along with minimum delay and information with great accuracy is saved into cloud server.