Detection of jamming and interference attacks in wireless communication network using deep learning technique

Abstract

. The Jamming and interference attacks aim to disable a wireless network, inducing a denial of service. Despite the resilience offered 5G is prone to these regarding the impact to the use of millimetre wave bands. In the last decade, several jamming detection techniques have been developed, including fuzzy logic, game theory, channel surfing, and some others statistical modeling. The plurality of these strategies are inadequate at detecting smart jammers. As a response, efficient and quick jamming and interference high-accuracy detection systems are all still in great demand. The usefulness of many deep learning models in detecting jamming and interference signals is analyzed in this paper. The types of signal features that could be used to diagnose jamming and interference signals are investigated, and a large dataset was created using these parameters. Deep learning algorithms are being kitted, tested, and sorely tested using this dataset. Logistic regression and naïve bayes are representations of these algorithms. The probability of detection, probability of false alarm and accuracy are being used to verify and validate the performance of these algorithms. The simulation results show that a logistic regression algorithm based on jamming detection and interference can detect jammers with perfect seating, a high possibility of detection, and a minimal probability of false alarm.