Blockchain-Aided Intrusion Detection in Marine Tactical Network Using Reinforcement Learning

Marine Tactical Networks (MTNs) are essential for secure maritime operations, but are highly susceptible to cyber threats. Traditional Intrusion Detection Systems (IDS) often struggle to adapt to the dynamic and complex nature of MTNs. This paper introduces a Blockchain-Aided Intrusion Detection System (BAE-RL), which integrates reinforcement learning (RL) and blockchain technology to improve threat detection and security. The BAE-RL framework is unique in its use of multi-agent adversarial RL, where a defender agent learns to detect attacks by interacting with a simulated attacker agent. This adversarial setup enhances the system’ s ability to identify novel and evolving threats. Additionally, blockchain integration ensures the integrity and immutability of detection data, preventing tampering and ensuring transparency. Experimental results show that the proposed framework outperforms traditional IDS, achieving 80.16% and 95.9% accuracy on the NSL-KDD and AWID datasets, respectively. The BAE-RL framework offers a robust, adaptive, and secure solution for intrusion detection in MTNs.