Fuzzy BP neural network in radar intelligence quality evaluation

Abstract

This paper presents a novel approach of radar station intelligence quality evaluation which based on fuzzy Backpropagation neural network (BPNN). Firstly, the index system of the radar station intelligence quality evaluation is established according to the analysis of the process, the characteristics, and the main influencing factors of the radar station intelligence production. And then the factor set, comment set and the membership matrix are structured, the fuzzy BPNN for evaluating the quality of the radar station intelligence is designed referring to the index system. Finally, the experiment shows that the accuracy and stability can be improved effectively by using fuzzy BPNN to evaluate the radar station intelligence quality

Construction of a Smart City Network Information Security Evaluation Model Based on GRA-BPNN

“…To promote the development of network information security in smart cities, the study of network information security evaluation methods has become the focus topic nowadays, for which researchers at home and abroad have also conducted in-depth research. Liu et al proposed a radar station intelligence quality evaluation method based on fuzzy back propagation neural network (BPNN), through the analysis of the radar station intelligence production process, characteristics and main influencing factors, established a radar station Intelligence quality evaluation index system, then constructed factor set, evaluation set and affiliation matrix, and designed a fuzzy BP neural network for radar station intelligence quality evaluation with reference to this index system [6]. Lin and his experimental partners proposed a new tandem evaluation method, hierarchical classification model tandem back propagation neural network (BPNN) method, for multi-metal sensors for predicting rice taste quality, which extracts the characteristic current and potential arrays and processes them appropriately to obtain potential phase plane values and current phase plane values as the current phase plane values used as input variables for the rank classification model, and uses the classification and rank values of the sample outputs and potential phase plane values for score prediction in BPNN [7].…”

“…In Equation ( 5), ∂ ij is the relative difference between the alternative i and the ideal solution under the attribute j, and ρ ∈ [0, 1] is the discrimination coefficient. The correlation between the decision information series of each alternative and the ideal solution is calculated and multiplied with the weight value of each indicator to obtain its weighted correlation value r i , as shown in Equation (6).…”