The robust algorithms for optimization of nonrecursive filter coefficients by the combined criterion of minimum mean-square error

Authors

  • S. N. Kirillov Ryazan State Radio-Engineering Academy, Russian Federation
  • S. M. Shubin Ryazan State Radio-Engineering Academy, Russian Federation

DOI:

https://doi.org/10.3103/S0735272703010102

Abstract

The authors report of development and tests of new robust algorithms for optimization of nonrecursive digital filters based on a combined criterion of minimum of the mean-square error. The latter contains information concerning variation of the synthesis inaccuracy. The algorithms suggested are shown to provide a decrease by 1.7 dB in the dynamic range of the error signal at the output of the filter under synthesis, and exhibit low sensitivity to inaccurately set initial parameters.

References

  1. CAPPELLINI, V.; CONSTANTINIDES, A.G.; EMILIANI, P.L. Digital Filters and Their Application. Academic Press, 1978.
  2. WIDROW, B.; STEARNS, S.D. Adaptive Signal Processing. Prentice-Hall, 1985.
  3. KIRILLOV, S.N.; STEPANOV, M.V. "Kombinirovannyi kriterii optimizatcii koeffitcientov adaptivnykh filtrov s konechnoi impulsnoi harakteristikoi," Radiotekhnika, n.2, p.39-41, 1999.
  4. KIRILLOV, S.N.; STEPANOV, M.V. "Optimizatciia ustroistv tcifrovoi obrabotki signalov po kombinirovannomu kriteriiu srednego kvadrata oshibki," Tsifrovaya Obrabotka Signalov, n.1, p.27-32, 2000.
  5. KIRILLOV, S.N.; STEPANOV, M.V. "Optimizatciia parametrov ARSS-modelei po kombinirovannomu kriteriiu minimuma srednego kvadrata oshibki," Izv. VUZ. Radioelektronika, v.43, n.3, p.21-26, 2000.

Downloads

Published

2003-01-10

Issue

Vol. 46 No. 1 (2003)

Section

Research Articles