The Convolutional Neural Networks (CNN) is a deep learning method enhances the Network Intrusion Detection System (NIDS) capability to detect the class of normal and abnormal data.
Traditional Deep learning models are infeasible for constrained wireless devices... customized deep learning can be used for your case... you can check "Enabling Deep Learning on IoT Devices"
Luay Abdulwhid Shihab where is your Artical where you proposed this model ( We use Self-taught Learning (STL), a deep learning based technique, on NSL-KDD ) kindly send me i want to study.. Thanks