Reinforcement learning based access control architecture of Internet of Things big data peak clustering model

In order to improve the accuracy of Internet of Things access control and the clustering of data transmission, a big data peak clustering method based on reinforcement learning is proposed. A big database management model of Internet of Things access control architecture is established by adopting global data pattern. Based on heterogeneous parameters among big data sources of Internet of Things access control architecture, combined with structural feature analysis of data sources, a big data interference filtering model of Internet of Things access control architecture is established by adopting the feature analysis method of blockchain fusion control and association rule mining, and feature extraction of big data peaks of Internet of Things access control architecture is carried out through reinforcement learning algorithm. According to the change of Internet of Things access mode, cluster analysis and pattern recognition of Internet of Things access control architecture big data peak are realized. By constructing the spatial-temporal distribution model of Internet of Things access control architecture big data and Internet of Things transmission channel, the spectral density cluster analysis method is adopted, according to the quantitative parameter analysis of real-time Internet of Things access control architecture data stream, the quantitative recursive analysis method is adopted, and the online Internet of Things access control architecture big data cleaning is used to realize the identification and cluster analysis of Internet of Things access control architecture big data peak features, so as to improve the Internet of Things access control ability. The simulation results show that this method is highly reliable for peak clustering of big data in the access control architecture of the Internet of Things, and has strong dynamic analysis and recognition ability for Internet of Things access and data scheduling, good convergence of data clustering, and low error rate.