In Mobile Ad hoc Networks (MANETs) there are some security problems because of portability, element topology changes, and absence of any framework. In MANETs, it is of extraordinary significance to identify inconsistency and malignant conduct. With a specific end goal to recognize malignant assaults by means of interruption identification frameworks and dissect the information set, we have to choose some components. Thus, highlight determination assumes basic part in recognizing different assaults. In the writing, there are a few recommendations to choose such elements. For the most part, Principal Component Analysis (PCA) breaks down the information set and the chose highlights. In this paper, we have gathered a list of capabilities from some cutting edge works in the writing. Really, our reproduction demonstrates this list of capabilities identify inconsistency conduct more precise. Likewise, interestingly, we utilize PCA for investigating the information set. In contrast to PCA, our results show Sequential pattern mining (SPM) cannot be affected by outlier data within the network. The normal and attack states are simulated and the results are analyzed using NS2 simulator.
A Projection Pursuit Based Risk Assessment Method in Mobile Ad hoc Networks