Generalized neural method for shaping the exponential chip weighting waveforms in direct sequence CDMA systems


Develi I., KALINLI A., Ciftlikli C.

FREQUENZ, cilt.60, ss.60-64, 2006 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 60
  • Basım Tarihi: 2006
  • Doi Numarası: 10.1515/freq.2006.60.3-4.60
  • Dergi Adı: FREQUENZ
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Sayfa Sayıları: ss.60-64
  • Orta Doğu Teknik Üniversitesi Adresli: Hayır

Özet

In recent past, artificial neural networks have been successfully introduced as a useful method for shaping the exponential chip weighting waveforms (ECWW). However, the results available in the literature are applicable to direct sequence-code division multiple access systems using a particular number of users considered, but cannot be directly generalized for systems include arbitrary number of users. In order to deal with this problem, a general and advanced methodology based on multilayer perceptron neural networks is presented in this paper. The proposed network has been tested with new set of data to demonstrate the performance and applicability of the methodology. Numerical results have shown this methodology to be robust and general use for the computation of the optimal value of the parameter belonging to ECWW.