Chinese Journal of Magnetic Resonance ›› 1997, Vol. 14 ›› Issue (3): 223-228.

Previous Articles     Next Articles

RECOGNITION OF KNOWN SIGNAL IN STRONG NOISE BASED ON FUZZY NEURAL NETWORKS

Pan Tao1, Li Gengying2   

  1. 1 Department of Physcis, Suzhou Railway Teachers College, Suzhou 215009;
    2 Analytical Center, East China Normal University, Shanghai 200062
  • Received:1996-11-26 Revised:1997-01-20 Published:1997-06-05 Online:2018-01-22

Abstract: In this paper, problems associated to recognition of known signal submerged in noise is studied with computer simulations by Fuzzy Neural Networks. The research results show that, under very low signal-to-noise ratio, a very high recognition rate was still kept by the networkss with combination of fuzzy membership function and BP algorithm. In addition, the practicability of this recognition method was investigated.The approach opens a new way to recognition of signal embeded in strong noise in NMR.

Key words: Signal recognition, Fuzzy neural networks, NMR