IBM Journal of Research and Development
IBM Skip to main content
  Home     Products & services     Support & downloads     My account  

  Select a country  
Journals Home  
  Systems Journal  
Journal of Research
and Development
    Current Issue  
    Recent Issues  
    Papers in Progress  
    Recent publications  
    Author's Guide  
  Contact Us  
  Related links:  
     IBM Research  

IBM Journal of Research and Development  
Volume 29, Number 1, Page 68 (1985)
Nontopical Issue
  Full article: arrowPDF   arrowCopyright info


Adaptive clustering algorithm

by L. V. O'Malley
Output data from many types of sensor systems (radar, radar warning, sonar, electro-optical, etc.) must be associated with one or more possible sources based on multiple observations of the data. This paper presents an algorithm that associates data with their source by simultaneous n-dimensional clustering of multiple data observations. The algorithm first orders the observations by successive nearest neighbor, in the n-dimensional Euclidean sense, from a defined starting point. Clusters are then isolated using a method derived from statistical decision theory. The algorithm's primary feature is its ability to perform clustering adaptively without any assumptions about the size, number, or statistical characteristics of the clusters. Since the algorithm was developed for radar warning system processing, a performance comparison with a well-known algorithm used in that field is included.
Related Subjects: Algorithms; Signal processing