This paper presents a design framework based on a centralized scalable architecture for effective simulated aerial threat perception. In this framework data mining and pattern classification techniques are incorporated. This paper focuses on effective prediction by relying on the knowledge base and finding patterns for building the decision trees. This framework is flexibly designed to seamlessly integrate with other applications.
The results show the effectiveness of selected algorithms and suggest that more the parameters are incorporated for the decision making for aerial threats; the better is our confidence level on the results. To delve into accurate target prediction we have to make decisions on multiple factors. Multiple techniques used together helps in finding the accurate threat classification and result in better confidence on our results.