Tight lower bounds for dynamic time warping

Dynamic Time Warping (DTW) is a popular similarity measure for aligning and comparing time series. Due to DTW's high computation time, lower bounds are often employed screen poor matches. Many alternative have been proposed, providing range of different trade-offs between tightness computational efficiency. LB Keogh provides useful trade-off in many applications. Two recent bounds, Improved Enhanced, substantially tighter than Keogh. All three the same worst case complexity - linear with respect series length constant window size. We present four new DTW class. Petitjean Improved, only modest additional overhead. Webb more efficient while bound. always The parameter free usually Enhanced. A parameterized variant, further Webb*, some constrained distance functions. In extensive experiments, proves be very effective nearest neighbor search.