An alternative mixing parameter update rule for complex-valued collaborative adaptive filter structures

Çolak Güvenç B., Mengüç E. C.

3rd International Turkic World Congress on Science and Engineering , Niğde, Türkiye, 14 - 15 Haziran 2021, ss.1-11

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Basıldığı Şehir: Niğde
  • Basıldığı Ülke: Türkiye
  • Sayfa Sayıları: ss.1-11
  • Kayseri Üniversitesi Adresli: Evet

Özet

This paper introduces an alternative mixing parameter update rule for complex-valued collaborative adaptive filter structures, which is based on the minimization of the kurtosis cost function defined by the instantaneous error signal. With the help of this alternative mixing parameter, the advantages of the CLMS and ACLMS algorithms over circular and noncircular complex-valued signals are integrated into the proposed hybrid adaptive filtering algorithm. The simulation results on system identification settings demonstrate that the proposed hybrid algorithm with an alternative mixing parameter update rule is outperformed in a collaborative manner of the CLMS and ACLMS algorithms when compared to the individual algorithms.