A radar signal sorting algorithm based on improved k-means dynamic clustering and sub linear time algorithm

In order to meet the need of real time, electromagnetic environment became more and more complex and increasing signal flow density, a radar signal sorting algorithm based on improved k-means dynamic clustering and sub linear time algorithm is presented. The amount of calculation with computing time...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:2017 First International Conference on Electronics Instrumentation and Information Systems (EIIS) S. 1 - 5
Hauptverfasser: Li-peng, Gao, Hui-yu, Shan, Feng-you, Ji
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 01.06.2017
Schlagworte:
Big Data
Classification algorithms
Clustering algorithms
dynamic clustering
Hardware
Heuristic algorithms
k-means
Radar
signal sorting
Sorting
sublinear time algorithm
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:In order to meet the need of real time, electromagnetic environment became more and more complex and increasing signal flow density, a radar signal sorting algorithm based on improved k-means dynamic clustering and sub linear time algorithm is presented. The amount of calculation with computing time and the number of clustering iterations is decreasing sharply via setting clustering termination conditions. Aiming at resolving the disadvantages, which clustering can hardly show all the parameters of radar especially pulse repetition interval sub linear time algorithm, one of the common algorithm of big data is referenced. Firstly, each coming pulse description word is classified by clustering and stop when it reaches the termination conditions. Then the classified sequences are analyzed to find out the pulse repetition interval by sub linear time algorithm after the clustering per 100 ms. Finally, comprehensive radar parameters are searched and the signal is sorted. Experimental results show how the proposed algorithm is applicable and effective to sort signal and satisfy the real time with less calculation and higher accuracy.
DOI:10.1109/EIIS.2017.8298750