Detecting Jamming Attacks Using Machine Learning Models

A wireless network is the target of jamming attacks, which cause an undesired denial of service. Despite its robustness due to the utilisation of millimetre wave bands, 5G is susceptible to these attacks. Random jamming has the ability to disrupt wireless networks and stop conversations. Traditional jamming detection systems and solutions try to use fixed-threshold signal evaluation techniques or software-defined radios. Because of their relatively rigid behaviour, high false alarm frequency, and resource-intensive detections, traditional tactics frequently lose their effectiveness when faced with clever or adaptable threats. Fixed-threshold approaches are free from the need to integrate expensive radio frequency hardware and vast amounts of processing, whilst SDR-based implementations can only be fixed threshold mechanisms. RSSI, SNR, BER, and packet loss rate are considered as the detection model metrics in machine learning-based jamming detection systems. By responding to the jamming enquiries quickly, flexibly, and in a hardware- independent manner.