Abstract
This paper is concerned with estimating the dimension of an attractor. The Grassberger-Procaccia procedure for estimating the correlation dimension is reviewed, and some alternative estimators considered. A particularly important problem is to estimate dimension in the presence of noise in the data. In pursuit of this, a modification of the Grassberger-Procaccia procedure is proposed, and illustrated on both artificial and real data sets.
Users
Please
log in to take part in the discussion (add own reviews or comments).