Towards Energy-Efficient and Trust-Aware Routing in IoT-WSN: Problem Identification and Comprehensive Literature Review

The operation of Internet of Things (IoT) Wireless Sensor Networks (WSN) is essential to critical infrastructure in smart cities, industrial automation, and healthcare monitoring, however it is faced with three co-existing and interrelated problems, namely, energy depletion under limited battery capacity, trust issues from malicious node behaviour, and routing congestion caused by dynamic traffic fluctuations. However, at this point, each of the routing protocols discussed — from classical LEACH to AODV to the newer reinforcement learning-based protocols — solved at best two of these problems, so a new research paradigm is called for — a unified adaptive framework. In this paper, a comprehensive problem statement towards the proposed EDL-IoT-Net framework is presented, which is a Hybrid Deep Reinforcement Learning architecture where the Deep Q-Network (DQN), Temporal Convolutional Network (TCN) and Echo State Network (ESN) are combined. The systematic literature review includes 25+ papers from 2021 to 2025, highlighting the main methodological issues, and justifying the need of the joint energy-trust-congestion optimisation with temporal state modelling. The identified gaps are used as the basis for the design of the EDL-IoT-Net, and is described in the full length manuscript that is attached.