Title :
Comparing Relaxed and Non-Relaxed Fast Affine Projection Adaptation Algorithms
Author_Institution :
Inst. for Microstruct. Sci., Nat. Res. Council
Abstract :
In order to provide a guideline to designers of practical adaptation systems that are based on the family of affine projection algorithms, this paper derives and verifies certain simple formulation for learning curves of relaxed and non-relaxed affine projection algorithms (APAs), to which the fast affine projection (FAP) algorithms approximate. It is observed that the formulation accurately predicts the behavior of the algorithms when the input is independent and identically distributed (i.i.d.), and gives a reasonable estimate when the input is speech-not satisfying the i.i.d. condition
Keywords :
adaptive filters; filtering theory; APA; affine projection adaptation algorithm; learning curve; practical adaptation system; Adaptive filters; Echo cancellers; Equations; Filtering algorithms; Financial advantage program; Linear systems; Noise cancellation; Projection algorithms; Sampling methods; Speech;
Conference_Titel :
Digital Signal Processing Workshop, 12th - Signal Processing Education Workshop, 4th
Conference_Location :
Teton National Park, WY
Print_ISBN :
1-4244-3534-3
Electronic_ISBN :
1-4244-0535-1
DOI :
10.1109/DSPWS.2006.265403
