Adaptive Routing and Security for Heterogeneous Networks Using Quantum Key Distribution and Bat Optimized Recurrent Neural Network

Abstract

In contemporary heterogeneous networks, the reliance on robust and secure communication protocols is increasingly critical due to the rising sophistication of intruding techniques and diverse attack vectors. The dynamic nature of routing in these networks, coupled with nodes of varying computational capabilities, poses a risk of routing attacks, which significantly compromise network security and performance. To address these challenges, this paper introduces an advanced framework combining Post-Quantum Cryptography (PQC) with Bat Optimization Algorithm (BOA) based Adaptive Quantum Routing RNN (AQR-RNN) to enhance security and routing efficiency. Quantum Key Distribution (QKD) is employed to secure communications, thus providing a robust defense against threats. Simultaneously, BOA-AQR-RNN is utilized to optimize routing efficiency, inspired by the echolocation capabilities of bats. This approach leverages AQR-RNN architectures to adaptively learn and predict routing paths, enhancing decision-making and optimization processes. The synergy between QKD and BOA-AQR-RNN approach not only strengthens the security framework of heterogeneous network routing protocols but also achieves superior Quality of Service (QoS) by dynamically optimizing routing strategies. The proposed methodology demonstrates significant potential for advancing secured communication in Internet of Things (IoT) environments and other complex network architectures.