A Novel Data Association Algorithm for Unequal Length Fluctuant Sequence

Citations of this article
Mendeley users who have this article in their library.


There are quantities of such sensors as radar, ESM, navigator in aerospace areas and the sequence data is the most ordinary data in sensor domain. How to mine the information of these data has attracted a great interest in data mining. But sequence data is easily interfered and produces some fluctuant points. When dealing with these sequences, traditional sequence similarity measurement such as Euclidean distance arises large error, especially for unequal length fluctuant sequence. A novel average weight 1-norm unequal length fluctuant sequence similarity measurement algorithm based on dynamic time warping (DTW) is proposed to solve this problem. It constructs an absolute distance matrix based on DTW firstly, then weight average weight 1-norm and modify it with modifying factor to measure the distance of unequal length fluctuant sequence. It solves the fluctuation sensitivity of maximum distance measurement algorithm. Finally transform distance to similarity as the index of the association, associate the sequence data according to the maximum similarity association rule. Simulation results show the effectiveness of the proposed algorithm when associating unequal length fluctuant sequence, association rate is above 70% and simulate the effect of variation of the sequence length, fluctuant rate and processing time to the proposed algorithm.




Guan, X., Sun, G., Yi, X., & Guo, Q. (2015). A Novel Data Association Algorithm for Unequal Length Fluctuant Sequence. In Procedia Engineering (Vol. 99, pp. 1190–1202). Elsevier Ltd. https://doi.org/10.1016/j.proeng.2014.12.648

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free