An optimized deep networks for securing 5g communication system

Nowadays, cellular applications rule the digital world with their betterment. However, security is the primary concern for a better communication range during the communication process. If the messages are hacked, data overhead and collisions occur. So, the present research work has aimed to design the novel Buffalo-based Autoencoder Security Framework (BbASF) developed in the Orthogonal-Frequency-Division- Multiplexing (OFDM) channel. Consequently, the function of the designed model is checked with the Denial of Service (DoS)-CICIDS dataset. The planned model is tested in the python environment. After that, the communication parameters were validated and compared with other schemes. The presented approach has earned the finest outcome in all performance assessments than the compared models. Furthermore, it has reduced the packet loss to the desired level, 2.2e-6. In addition, the presented novel BbASF is good at forecasting malicious behavior; it has earned the best attack detection score of 99.6%. Hence, the present model efficiently predicts malicious features and enriches communication facilities.