Early Abandon to Accelerate Exact Dynamic Time Warping

Dynamic time warping is one of the important distance measures in similarity search of time series; however, theexact calculation of dynamic time warping has become a bottleneck. We propose an approach, named early abandon dynamictime warping, to accelerate the calculation. The method checks if values of the neighbouring cells in the cumulative distancematrix exceed the tolerance, and if so, it will terminate the calculation of the related cell. We demonstrate the idea of earlyabandon on dynamic time warping by theoretical analysis, and show the utilities of early abandon dynamic time warping bythorough empirical experiments performed both on synthetic datasets and real datasets. The results show, early abandondynamic time warping outperforms the dynamic time warping calculation in the light of processing time, and is much betterwhen the tolerance is below the real dynamic time warping distance