Enhancing Network Security with VGG-RF and Grey Gorilla Optimization: A Case Study on Aegean Wi-Fi Intrusion Dataset

These days, WI-Fi (Wireless Fidelity) is enabling an explosion of wireless gadgets, many of which need frequent maintenance and authorisation. Although modern wireless technology depends on Wireless Sensor Networks (WSN), they are susceptible to a wide spectrum of security concerns including attacks, illegal access, and suspicious activities and offer reasonably priced solutions for a range of monitoring needs even if they are fundamental to the technology. These vulnerabilities seriously impact the performance and dependability of WSNs, hence early identification and resolution of them is quite important. Intrusion Detection Systems (IDS) are absolutely essential in shielding WSNs from these hazards. Many studies aiming on enhanced Intrusion Detection Perfect Accuracy, reduced loss, higher Detection Rate, and lower False Alarm Rate show how to eliminate the repetitious aspect of the dataset. This work proposes a high-tech Network Intrusion Detection System (NIDS) to defend Wi-Fi-based WSNs from common cyber threats including injection attacks, flooding, and impersonation. This work presents a hybrid model aiming at enhancing intrusion detection by combining the VGG-RF technique with GGOA-based feature selection. The proposed approach chooses the most pertinent features from the Aegean Wi-Fi Intrusion Detection dataset to lower computing cost using GGOA without compromising classification accuracy. By use of the Random Forest (RF) classifier and the VGG network, which together extract deep hierarchical information, attack detection gains additional strength. In all four measures accuracy, precision, recall, and false positive rates the hybrid VGG-RF with GGOA model shows far better than conventional intrusion detection systems. Combining feature optimisation with hybrid classification methods increases network security noticeably, according to results. Future studies on real-time optimisation and deployment for circumstances with restricted resources can help to strengthen sensor-based network defences against cyber threats.