Feature Detection Algorithm of Chaotic Data in Distributed Networks

Abstract

The chaotic data in the distributed network is affected by the disturbance information of the network nodes, which leads to the poor convergence of the data feature detection. A new algorithm for the chaotic data feature detection in the distributed network is proposed based on fuzzy C-means clustering. According to the association rule attributes of the chaotic data in the distributed network, the principal component analysis is carried out, and the semantic attributes and fuzzy clustering features of the chaotic data in the distributed network are extracted. The joint association rule analysis method is used to deal with the interference filtering of the chaotic data in the distributed network, and the classified fuzzy set of the chaotic data in the distributed network is constructed based on the sensor fusion tracking measure method. In the fuzzy data set, the mixed weighting and adaptive block matching of chaotic data in distributed network are carried out, and the association rule feature quantity of chaotic data in distributed network is extracted. The extracted features are input into the fuzzy C-means clustering classifier for data classification and recognition, and the optimized feature detection of the chaotic data in the distributed network is completed. The simulation results show that the proposed method has good data convergence, strong convergence and anti-interference in the process of feature detection in distributed networks.