A Novel Approach to Detect Malicious User Node by Cognition in Heterogeneous Wireless Networks
Cognitive Networks are characterized by their intelligence and adaptability. Securing layered heterogeneous network architectures has always posed a major challenge to researchers. In this paper, the Observe, Orient, Decide and Act (OODA) loop is adopted to achieve cognition. Intelligence is incorporated by the use of discrete time dynamic neural networks. The use of dynamic neural networks is considered, to monitor the instantaneous changes that occur in heterogeneous network environments when compared to static neural networks. Malicious user node identification is achieved by monitoring the service request rates generated to the cognitive servers. The results and the experimental study presented in this paper prove the improved efficiency in terms of malicious node detection and malicious transaction classification when compared to the existing systems.
