IEEE Transactions on Acoustics Speech and Signal Processing · 1989 · 28 citations · 5 references
EngineeringComputer EngineeringElectrical EngineeringAnalog ElectronicsDigital CommunicationProcess ControlAnalog FiltersDigital Signal ProcessingFrequency ResponseDigital FilteringFilter DesignFilter Frequency ResponseFiltersFinite Impulse Response FiltersLow-pass FilteringLow-pass FiltersQuantization
A generalization of the statistical approach of D.S.K. Chan and L.R. Rabiner (IEEE Trans. Audio Electroacoust., vol.AU-21; p.354-66, Aug. 1973) for predicting the effects of quantized coefficients in direct-form finite-impulse response (FIR) filters is presented. The analysis is used on more sophisticated quantizing procedures involving linear feedback of rounding errors. It is shown that the expected disturbance filter frequency response due to quantization can be controlled. The effect is concentrated in tolerant frequency bands and significantly reduced in critical ones. The price paid is an increase of overall expected disturbance. The method is demonstrated in a practical filter design. A low-pass filter for factor 2 decimation that has greater margins for quantization effects in the passband than in the stopband is specified.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">></ETX>
5