Enhancing DDoS Attack Detection and Prevention using CNN-LSTM Hybrid Model for Improved Network Security

DDoS attacks are the great compromise for network security as they are vulnerable to a great extent an efficient Intrusion Environment (IDS) plays a key role in fighting agains and detecting such attacks. This paper proposes a new CNN-LSTM hybrid model-based intrusion detection and prevention system for distributed denial of service (DDoS) attacks. Since it is a complex data, convolutional neural network (CNN) model is used to extract spatial characteristics, while long short-term memory (LSTM) model is utilized for temporal dependence. Accordingly, the dataset used for the training and testing our model contains a variety types of network traffic. With an F1score of 99.91%, an accuracy of 99.86%, a precision of 99.9%, and a recall of 99.7%, the suggested CNN-LSTM model outperformed the competition when it came to identifying DDoS attacks. These measurements show that the model is quite accurate at detecting patterns of distributed denial of service attacks with few false positives and negatives. By accurately detecting and preventing DDoS attacks in a timely manner, an intrusion detection system (IDS) may greatly improve network security, guaranteeing the availability and stability of network services