GF-CWAO-GAN: A Generative-Adversarial and Optimization-Based Framework for Enhanced IoT Botnet Prediction and Network Security

Novel frameworks addressing distinct challenges in technology and network optimization. Firstly, GAN-based framework enhances Botnet prediction using advanced image processing techniques such as S-GF filtering, GDA tone mapping, and IGM for boundary enhancement. The framework includes feature matching and clustering with KL- HM-KMA to predict Botnet effectively, validated through comprehensive experimental analysis. Secondly, in the realm of sensor network applications with mobile sinks, a new framework employs DFS-A* and A-BOTNET algorithms. It optimizes data gathering by dynamically considering mobile obstacles, leveraging SM-D-C3H clustering, RTMDR-BSP for cluster head selection, PFBTP for localization, and AP-MTE for energy-efficient path selection. Simulation results demonstrate enhanced network lifetime and reduced transmission delay. Data gathering using mobile sinks introduces new challenges to sensor network applications. To better benefit from the sink’s mobility, many research efforts have been focused to minimize data gathering time. But, the methods work by considering obstacles as static and has limited distance and failed to address the co-linearity issues. With the motivation to improve the network lifetime and the stability of the network by addressing these issues, in this paper, a novel framework using DFS-A* and A-BOTNET algorithms are proposed for WSN. Initially, the WSN model along with the energy consumption model is designed to initiate the proposed model. After that, clustering is performed using the SM-D- C3H algorithm in which the cluster head for each cluster is selected using RTMDR-BSP. Thereafter, the sensor nodes are localized in the deployed area using the PFBTP algorithm and for the localized nodes, trajectory planning is executed to discover the available path in the presence of obstacles. Finally, the energy-efficient paths are selected using the AP-MTE protocol for data transmission. The simulation results showed that the proposed algorithms improved the network lifetime and minimized the delay in the network.