Improving energy efficiency and routing reliability in wireless sensor networks using modified ant colony optimization

Wireless sensor networks (WSNs) are essential in a wide range of applications, but the challenges of energy efficiency, load balancing, and optimal routing remain critical for ensuring long-term network reliability. In this study, we introduce a Modified Ant Colony Optimization Algorithm (MACOA) to address these challenges. The proposed MACOA lies in several key innovations to address the limitations of existing ACO-based and bio-inspired routing protocols. First, MACOA applies a multi-objective heuristic function to simultaneously optimize power consumption while ensuring reliability, bandwidth, and short path distances to achieve an efficient routing solution. Second, it introduces an adaptive pheromone decay mechanism that dynamically adjusts based on network conditions, such as node energy levels and link reliability, to prioritize energy-efficient paths. Third, MACOA incorporates a load-balancing factor that prevents the overloading of certain nodes, thus extending the network lifetime. Finally, it regulates the exploration–exploitation trade-off dynamically by promoting early-stage exploratory behavior and later-stage exploitative behavior during optimization. Together, these innovations enable MACOA to be an efficient routing protocol that outperforms current state-of-the-art algorithms. We compare the performance of the proposed MACOA with existing state-of-the-art techniques, such as Genetic Algorithms, Particle Swarm Optimization, Artificial Bee Colony, Deep Reinforcement Learning, and Energy Reliable ACO Routing Protocol (E-RARP) in terms of network lifetime, network stabilization time, energy efficiency, load balancing, and throughput. Extensive results demonstrate that the proposed method outperforms the compared techniques. They state the adaptability of the proposed MACOA to dynamic network conditions and its robustness to node failures, which make the proposed MACOA a promising solution for WSNs and qualify it as a potential solution to large-scale and power-limited WSNs.