Cross-layer DDoS attack detection in wireless mesh networks using deep learning algorithm

Abstract Wireless mesh networks (WMNs), owing to its decentralized design and resource limitations, are susceptible to several security vulnerabilities, including distributed denial of service (DDoS) attacks. Traditional DDoS detection techniques are usually unable to effectively mitigate such attacks in WMNs due to their dynamic and complex nature. In this work, we show the capability of a Deep Convolutional Neural Network (DCNN) algorithm at the cross-layer of the network protocol stack to accurately and robustly detect Distributed Denial-of-Service (DDoS) attacks in WMNs. DDoS attack assessment and recognition use a practical dataset varying standard actions such as end-to-end delay, energy consumption, packet delivery ratio, mean packet latency, detection ratio, and packet loss rate when using the CICDDoS2019 dataset. The result shows the proposed method’s strong performance compared to previous detection methods. The simulation results show DCNN-DDoS has a better detection ratio metric than D-ConCReCT, SVM-DoS, FSO-LSTM, HeltIoT-CNNIDS, and AIDS-HML, which grew by 78.12%, 38.54%, 22.8%, 16.33%, and 15.67% respectively. DCNN-DDoS has exhibited superior performance compared to other essential methods, which is evident from the empirical results, which have higher levels of accuracy.