A Cognitive Fuzzy-Neural Cheetah Algorithm for Resilient Network Routing Decisions

Dynamic communication networks such as Mobile Ad Hoc Networks (MANETs), Vehicular Ad Hoc Networks (VANETs), and heterogeneous Internet of Things (IoT) systems exhibit frequent topology changes, uncertain link conditions, and energy constraints. Conventional routing protocols and existing bio-inspired optimization approaches struggle to achieve fast convergence, stable routing, and energy efficiency simultaneously under such conditions. This paper proposes a Cognitive Fuzzy–Neural Cheetah Optimization Framework (CFN-COF) for adaptive and resilient network routing. The framework synergistically integrates the Cheetah Chase Algorithm (CCA) for rapid path exploration, a Fuzzy Logic System (FLS) for uncertainty-aware decision-making, and an Artificial Neural Network (ANN) for experience-driven learning and adaptation. Unlike existing hybrid routing methods, the proposed framework introduces a closed-loop cognitive feedback mechanism, wherein neural learning dynamically tunes fuzzy inference parameters that guide the acceleration and directional behavior of the CCA during route discovery. Extensive simulations conducted using NS-3 across MANET, VANET, and IoT scenarios demonstrate that CFN-COF achieves superior performance in terms of packet delivery ratio, end-to-end delay, convergence speed, and energy consumption compared with AODV, DSR, and Particle Swarm Optimization-based routing. The results confirm that the proposed cognitive hybrid framework provides a scalable and robust routing solution for next-generation dynamic networks.