AI-Driven Hierarchical Routing Protocol for Energy-Efficient Wireless Sensor Networks: Review

Wireless Sensor Networks (WSNs) are widely used in various fields including environmental monitoring, healthcare monitoring, industrial process monitoring, smart infrastructure and others. However, in spite of their popularity, the limited energy of the sensor nodes is still a significant challenge for sustainability and effectiveness of the networks. The main drawback of the conventional hierarchical routing approaches such as LEACH and HEED is the use of fixed or probability-based cluster-head selection solutions that result in the unbalanced energy consumption of the network nodes. To resolve this problem, this study proposes an Artificial Intelligence–Driven Hierarchical Routing Protocol (AI-HRP) which combines Deep Reinforcement Learning (DRL) for intelligent cluster-head selection and route management. The proposed framework is based on a Deep Q-Network (DQN) model which is used to select the optimal cluster heads based on different network parameters, such as, residual node energy, cluster density, link reliability, traffic condition etc. Moreover, a multi-level hierarchical structure is used to mitigate communications overhead while promoting scalability. The results of the simulations show that the proposed AI-HRP is superior to various existing hierarchical routing techniques in terms of network longevity, packet delivery performance, and overall energy utilization.