Bradley Elizabeth, Mantilla Ricardo
Department of Computer Science, University of Colorado, Boulder, Colorado 80309-0430.
Chaos. 2002 Sep;12(3):596-600. doi: 10.1063/1.1488255.
A recurrence plot is a two-dimensional visualization technique for sequential data. These plots are useful in that they bring out correlations at all scales in a manner that is obvious to the human eye, but their rich geometric structure can make them hard to interpret. In this paper, we suggest that the unstable periodic orbits embedded in a chaotic attractor are a useful basis set for the geometry of a recurrence plot of those data. This provides not only a simple way to locate unstable periodic orbits in chaotic time-series data, but also a potentially effective way to use a recurrence plot to identify a dynamical system. (c) 2002 American Institute of Physics.
递归图是一种用于序列数据的二维可视化技术。这些图很有用，因为它们能以一种肉眼易见的方式呈现出所有尺度上的相关性，但其丰富的几何结构可能使其难以解读。在本文中，我们认为嵌入混沌吸引子中的不稳定周期轨道是这些数据递归图几何的一个有用基集。这不仅提供了一种在混沌时间序列数据中定位不稳定周期轨道的简单方法，还提供了一种利用递归图识别动力系统的潜在有效方法。（c）2002美国物理研究所。
