Adaptive Routing in Urban Wireless Ad Hoc Networks Using AI-Assisted Algorithms

Modern smart city applications as well as vehicular communication and emergency response systems heavily depend on Urban Wireless Ad Hoc Networks (UWAHNs). Conventional network routing procedures face multiple challenges in handling abrupt route topology changes and delayed performance and consume excessive energy while operating in unsecure environments. A proposed routing procedure incorporates artificial intelligence through reinforcement learning predictive models and hybrid optimization with PSO-GA to improve network performance characteristics. Real-time network condition analysis allows the algorithm to dynamically choose the best routing paths which enhances packet delivery and maximizes transmission effectiveness. The proposed routing design supported by AI assistance exceeds traditional methods according to extensive simulation results. Experimental results demonstrate the algorithm's delivery of 94.7% packet success rates while decreasing end-to-end delay by 40% with 30% reduced energy usage to extend the network's operational lifespan. The anomaly detection module powered by AI defends networks from black hole and wormhole and Sybil attacks while protecting users from cyber security threats. The convergence of edge computing and federated learning results in enhanced decision-making capabilities as well as adaptable network system features. Study data shows AI-driven adaptive routing constitutes an adaptable efficient and protected technical solution for UWAHNs. Future investigation includes blockchain security research along with artificial intelligence model enhancements and real-world network installation to boost large urban systems' efficiency.