Analysis and Attack Detection in GSM Mobile Network With an Intelligent Jammer Using ANFIS Classifier

A Mobile Ad hoc Network (MANET) is an autonomous system comprising mobile nodes that self-organize and connect via wireless networks, without reliance on a predefined infrastructure. These nodes are inherently susceptible to jamming, a form of denial-of-service attack that renders mobile services unavailable in the affected area. In this work, we introduce an intelligent jammer, constructed based on the Received Signal Strength Index-based Transmission Power Control (RSSITPC) Algorithm. This algorithm leverages Received Signal-Strength Indicator (RSSI) data to ascertain the optimal transmission powers for neighboring nodes and dynamically adjust these powers. The design of the intelligent jammer system incorporates a circuit interface, power unit, power detector, and GSM scanner. It employs a DAC-centered RSSITPC Algorithm to differentiate the jamming signal from legitimate signals by comparing the voltage in the received signal. Following the design phase, a Jamming Attack (JA) analysis is conducted, utilizing metrics such as the Packet Send Ratio (PSR) and Packet Delivery Ratio (PDR). Subsequently, a Hybrid Cross-layer Rate Adaptation (CLRA) Scheme is implemented to enhance JA detection and improve Wireless Link Utilization. The Adaptive Neuro-Fuzzy Interference System (ANFIS) classifier is then used to categorize data as either attack or regular data. For regular data, the Control Channel Attack Prevention (CCAP) algorithm is applied as a preventive measure. The proposed system's effectiveness is validated through comparative performance analysis with other widely used systems. Additionally, considerations are made for the adaptability of these methodologies to evolving intrusion techniques and changing network environments, as well as their scalability in larger, more complex networks.