A Novel Neural Network Architecture for Radar Clutter Classification

Eraslan B., GÜVENSEN G. M., TANIK Y.

18th IEEE World Symposium on Applied Machine Intelligence and Informatics (SAMI), Herlany, Slovakia, 23 - 25 January 2020, pp.263-268 identifier identifier

  • Publication Type: Conference Paper / Full Text
  • Volume:
  • Doi Number: 10.1109/sami48414.2020.9108755
  • City: Herlany
  • Country: Slovakia
  • Page Numbers: pp.263-268
  • Middle East Technical University Affiliated: Yes

Abstract

In recent years, potential capabilities of modern radars have become tremendous, with rapid advances in electronics and software technologies. If the radar system processes received echoes based on clutter characteristics, detection performance significantly improves. To classify clutters, neural network structures are studied. A problem specific architecture, specialized in clutter classification, is designed. The design procedure of the neural network is explained with necessary theoretical background information. The performance of the network is illustrated with experimental results.