Enhancing DDoS Detection and Mitigation in 5G Networks Using LSTM and Harris Hawk Optimization

High-demand services like autonomous factories and smart cities depend on 5G networks, but they also present security risks like Distributed Denial of Service (DDoS) attacks.These attacks reduce the viability of conventional detection and mitigation techniques by interfering with network resources, interfering with communication, and compromising system dependability.In order to identify and neutralize DDoS assaults, this study proposes an intelligent and adaptable system that can include Harris Hawk Optimization (HHO) and Long Short-Term Memory (LSTM).While the HHO algorithm is used to fine-tune the mitigation approach dynamically and narrow down the model parameters, LSTM may learn the temporal traffic patterns and, as a result, accurately identify DDoS activity.Additionally, the Fox-style Optimization Algorithm is used to reroute traffic to trusted and energy-efficient nodes when an attack has been detected.By giving priority to trust value, energy level, and network speed, this algorithm reduces the amount of disturbance to the service.The effectiveness of the proposed framework is demonstrated experimentally with a 95.63% detection rate, 94.56% specificity, 95.02% sensitivity, and 94.45% accuracy.Additionally, with a margin of improvement of 5.4%, 7.3%, 33.8%, and 11.5%, respectively, the Harris Hawk Optimization strategy outperforms other optimization strategies such as DSO, EVO, GFROA, and PCOA.Through reliable, low-latency communication even during attacks, the framework not only improves the security posture of 5G networks but also boosts their performance.