Quantification Analysis of Time Series Based on 2D Visibility State Plot
Chunyu Zhao, Ming Zeng, Erhong Wang, Jingjing Han, Qing‐Hao Meng · 2018
Time series analysis is of significant importance with various engineering applications. Inspired by the idea of recurrence quantification analysis (RQA), a novel quantification time series analysis method based on 2D visibility state plot is proposed. In particular, if the two arbitrary data satisfy the visibility criterion, and consequently become two connected nodes of the associated visibility state plot (VSP). Next, In order to quantitatively analyze the VSP, six measures, i.e., visibility rate (VR), degree of certainty (DC), mean diagonal length (DL), visibility entropy (VE), layered (LAY), vertical/horizontal line length (VHL), are calculated. Finally, to verify the effectiveness of the proposed method, we compare the performance of our approach to the RQA method using wind speed signals of two different environments. The results show that the quantitative measures of VSP can distinguish indoor and outdoor signals, but the classical RQA method fail to work.