Journal of Physics: Conference Series | 2021
The Electromagnetic Signal Track Correlation Algorithm based on Attribute Clustering and Space-time Constraints
Abstract
In view of the problem that start-end correlation of ship electromagnetic signal, the electromagnetic signal track correlation algorithm, based on attribute clustering and space-time constraints, is proposed. The algorithm consists of K-means dynamic clustering stage, space-time constraints of the track association stage. By extracting the relevant properties of electromagnetic signals, through the clustering stage, obtaining an indefinite number of clusters, and through the space-time correlation stage, obtaining a number of tracks per cluster. The main contributions of this paper are: 1. the K-means dynamic clustering method is proposed; 2. the track correlation method of space-time constraints is proposed. In this paper, the results of the comparison experiment show that the accuracy of this algorithm is better than that of the comparison algorithm, and the practical feasibility is verified.