A Multivariate Time Series Anomaly Detection Method Based on Clustered Particle Swarm Optimization

Abstract Due to the advent of 5G and integration sensors, sensor nodes need collect data from multiple sources simultaneously.Traditional anomaly detection methods for single attribute time series detect multivariate data, low accuracy, significant node energy consumption. To avoid these problems, this paper provides an improved FCM clustering method based on sliding window. Based CPSO optimal weight, reconstruction error realizes detection. Experiments different datasets demonstrate effectiveness algorithm. Moreover, compared with three classical methods, results show that proposed algorithm has higher accuracy accuracy.