Threat Assessment Based on Adaptive Intuitionistic Fuzzy Neural Network

Abstract

This paper proposes a method for threat assessment(TA) based on Adaptive Intuitionistic Fuzzy Neural Network(AIFNN). Firstly, intuitionistic fuzzy proposition is defined and the concept of intuitionistic fuzzy reasoning is discussed and Takagi-Sugeno Kang intuitionistc fuzzy model is developed. Secondly, a model for TA on AIFNN based on Takagi-Sugeno Takagi-Sugeno Kang intuitionistc fuzzy model is established, the attribute functions, ie. membership and non-membership functions, and the inference rules of the system variables are devised with computational relations between layers of input and output and a synthesized computational expression of system outputs ascertained. Thirdly, a learning algorithm of neural based on the extended kalman algorithm is designed. Finally, the validity of the technique is checked and rationality of constructed model is verified by providing TA instances with 400 typical targets. The simulated results show that this method can enhance creditability of TA and improve quality of assessment with precision of synthetic values in reasoning output.