Design simplifications for bi notch filters

Authors:
Rohini Deshpande;B. Kumar;S. B. Jain
Affiliations:
Department of Electronics and Communication Engineering, RITM, Bangalore, India 560064;Department of Electronics and Communication Engineering, BPIT, New Delhi, India 110089;Department of Electronics and Communication Engineering, IGIT, Delhi, India 110006
Venue:
Analog Integrated Circuits and Signal Processing
Year:
2011

Citing 5
Cited 0

Discrete-time signal processing (2nd ed.)

Discrete-time signal processing (2nd ed.)
An approach to model high-frequency distortion in negative-feedback amplifiers: Research Articles

International Journal of Circuit Theory and Applications
Design of square-root domain filters by substituting the passive elements of the prototype filter by their equivalents

International Journal of Circuit Theory and Applications
Fast communication: Design of maximally flat linear phase FIR notch filter with controlled null width

Signal Processing
Zolotarev polynomials and optimal FIR filters

IEEE Transactions on Signal Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

A methodology for designing FIR bi notch filters (NF) derived from second order prototype IIR NFs is suggested. Rejection bandwidth for the designed filter can be controlled by suitable choice of `r', the pole radius of the IIR prototype NFs. The suggested bi-NF can also be adapted to eliminate second, third and fourth order harmonics of periodic noise besides the fundamental noise frequency component. Such filters find application in switched mode power regulators meant to supply ripple free dc for sensitive electronic gadgets and also in host of communication circuits. A special case when two notch frequencies 驴1 and 驴2 are such that [(cos 驴1)(cos 驴2) = 驴1/2] has also been discussed. The IIR bi-NF design for this special case results reduction in the number of multipliers without affecting the response of the desired NF. For the afore, referred condition, the required design weights of FIR bi-NF reduce to almost half in number resulting in reduced computations. The number of zero weights further reduces with increase in `r' value. Also, the frequency response becomes better, with reduced ripples in the pass bands, when `r' is increased and the length `L' of the FIR NF is chosen appropriately.