Published Paper Details:

Wireless Sensor Networks (WSNs) are widely used in applications such as environmental monitoring, healthcare, industrial automation, and smart infrastructure. However, the limited energy resources, dynamic network conditions, and vulnerability of sensor nodes create significant challenges for reliable and secure communication. Conventional routing protocols generally rely on static routing decisions and may not effectively respond to changing network conditions or malicious behaviour. This paper proposes an adaptive trust-aware routing framework that integrates trust evaluation with Deep Reinforcement Learning (DRL) to improve routing decisions in wireless sensor networks. The proposed approach continuously evaluates node behaviour based on communication reliability and network characteristics and incorporates the resulting trust information into the routing decision process. A DRL-based agent is designed to learn suitable routing actions according to the current network state, thereby enabling adaptive selection of reliable communication paths. The methodology focuses on improving secure and efficient data transmission while reducing unnecessary routing overhead and energy consumption. The proposed framework is designed to be evaluated using performance measures such as packet delivery ratio, end-to-end delay, energy consumption, routing overhead, and network lifetime. The study provides a methodology-oriented approach for combining intelligent decision-making and trust management in WSN routing and establishes a foundation for future simulation-based validation under different network and security conditions.