AN ENERGY EFFICIENT MATHEMATICALLY MODIFIED GLOWWORM SWARM OPTIMIZATION FOR ROUTING IN WSN

Abstract

The wireless sensor networks (WSN) is limited with the energy utilization efficiency because the lifetime of the network should be as long as possible. This paper presents the enhanced routing algorithm based on the combination of Integer Linear Programming (ILP) with a modified Glowworm Swarm Optimization (GSO) technique to improve the energy consumption and scalability in flow-changing networks. The ILP model directly selects the energy-efficient transmission paths accounting for the least total energy expenditure satisfying strict constraints and fairly distributing the energy usage among the nodes. The proposed GSO algorithm improves these decisions by periodically comparing the available routing paths with reference to energy consumption and transmission time values. The combination of data mining with the ILP-GSO framework promotes the usage of past and real-time WSN data for routing. Clustering detects energy efficient groups of nodes whereas classification determines energy status of nodes for equal load routing. Association rule mining reveals bundle usage patterns to assist path decision, and predictive analytics detects energy exhaustion to allow path redirection. The integration of the data mining enhances scalability, flexibility, and power optimization; it makes routing choices based on information and does not deplete network energy in volatile, energy-limited conditions. Using these strategies in the proposed method, a huge cutting down in energy consumption is realized whereby the proposed method transmits data using an average of 0.213J energy. Performance comparison with other current approaches reveals better efficiency in terms of energy consumption, shorter transmission time, higher data rates through the ILP-GSO framework. The hybrid model is flexible in its performance criteria and can maintain reliable efficiency regardless of the size of the network and its changes. This approach not only prolong the life cycle of the network to avoid early exhaustion of nodes but also enhances the ability of the network to deliver high-quality performance in terms of energy efficiency and lifetime compared to other typical approaches in the related literature. This work can offer a scalable and efficient approach towards achieving energy-aware network routing.