An End-to-End Deep Learning Neural Network Framework for Proactive DDoS Attack Detection in Cloud Platforms

In a Distributed Denial of Service (DDoS) assault, hackers sneakily insert malicious data packets into client networks' internet traffic, causing a catastrophic outage for end users in a cloud compute system. This is a major issue with the internet of things as detecting and mitigating intrusions is a difficult operation that will impact the whole architecture's functioning. To keep the server safe from cybercriminals, we have implemented a plethora of internet safety safeguards. The goal is to discover any gaps or flaws and then create these algorithms. The development of a model using deep learning for successful intrusion categorization allows for better management of large data sets with a faster convergence rate. Automatic encoding and decoding are the basis of the chosen learning method. There is a problem with over-fitting, even if deep learning models are great at dealing with large datasets. A experimenting approach is used to correct concealed layers and unconscious neurons in order to conquer this. The study of the findings showed that the suggested deep learning technique outperformed the competition in security activities when it came to DDoS tracking. With a accuracy of 97.55%, the suggested Deep Learning Neural Network (DLNN) approach surpasses all competing models. If you employ Harris Hawks Optimization (HHO) and Particle Swarm Optimization (PSO) separately, you'll see worse results; HHO is less sensitive and accurate, while PSO is somewhat better. When HHO and PSO are combined with DLNN, the results are much better. The combination of HHO and DLNN yields a significant improvement in both measures. The suggested DLNN model, however, outperforms all of the other approaches.