Abstract:

Although wireless networks allows reliable communication in a diversity of fields, they are vulnerable to intruder jamming attacks that degrade the performance of the network. Secure and stable communication highly relies on the precise detection and effective localization of jamming attacks. The existing methods do not perform well because of low detection rates and high false alarms during the localization of jamming sources. To overcome these limitations, the proposed framework integrates spectrum sensing and time-of-arrival analysis with machine learning (ToA-ML) techniques to identify and accurately locate jamming sources. The proposed models apply in scenarios such as mobile wireless sensor networks, military border enforcement communications, and smart infrastructure systems in which communication is nonstop and must remain uninterrupted. Through experimental assessment, the model proved to greatly enhance jamming detection with lower false alarms, improved accuracy of target localization, and strengthened reliable wireless connectivity systems against intruder interference.