Showing papers on "Stopband published in 1970"

Design of nonrecursive digital filters with minimum phase

Abstract: A method is described of transforming nonrecursive filters with equal-ripple attenuation in the passband, stopband and linear phase into those with minimum phase and half the degree, but again with equal-ripple attenuation in the passband and stopband.

Design of nonrecursive digital filters with linear phase

Abstract: A new class of selective nonrecursive digital filters with independently prescribed equiripple passband and stopband attenuation and linear phase is obtained by numerical solution of a set of nonlinear equations. Some examples are given, and a comparison is made of the new solutions and those previously known.

Modulated arm width spiral antenna

Abstract: A multi-arm spiral antenna which allows unlimited broadband operation with dual senses of circular polarization The antenna comprises spiral arms having width variations which are logperiodically scaled to produce local reflection (stopband) regions along the arms The position of the stop-band regions is a function of the period and amplitude of width variations Arm currents are produced by excitation of the antenna These currents are reflected by the stopband regions The relative phase of the reflected currents is a function of the relative scaling of the arms

Maximally flat low-pass filters with steeper slopes at cutoff

Abstract: Maximally flat rational functions with one pair of imaginary axis zeros are compared with the all pole Butterworth function for low-pass filter applications. It is shown that by using the location of the zeros as a parameter an advantageous trade between attenuation in the stopband and sharpness of the cutoff characteristic can be made. Graphs which serve as guides in the placement of these zeros and a practical design example are included.

Equivalent Circuit of Orthogonal-Loop-Coupled Magnetic Resonance Filters and Bandwidth Narrowing Due to Coupling Inductance

Abstract: The equivalent circuit of an orthogonal-loop-coupled magnetic resonance filter is shown to consist of a gyrator, two ferrite-induced inductances, and two coupling loop inductances. The effects of the coupling inductances on the passband and stopband responses are shown to be significant by means of calculations based on this equivalent circuit. It is proved that the maximum passband bandwidth /spl Delta//spl conint//sub -3dB/ = /spl conint//sub 0/ (L/sub f/ / L/sub c/), where /spl conint//sub 0/, is the center frequency, and L/sub f/ and L/sub c/ the ferrite-induced and the coupling-loop inductance, respectively. Other unusual insertion-loss characteristics of this filter which differ from those of a conventional reciprocal-element bandpass filter are shown. Finally, a test circuit for determining experimentally the coupling inductance ratio L/sub c/ /L/sub f/ and the external Q, Q/sub f/ of a ferrite resonator is presented.

Composite crystal filter circuit

Abstract: A single composite crystal filter circuit provides an attenuation versus frequency characteristic which herebefore required a separate band-pass crystal filter, an LC filter, and a band reject crystal filter. The attenuation versus frequency characteristic provides stopband attenuation over a wide frequency range, a passband over a narrow frequency range, and high attenuation at one or more specific frequencies very near an edge of the passband.

Computer optimisation of cascaded noncommensurable-line lowpass filters

Abstract: The performance of lowpass filters composed of cascaded noncommensurable-line elements is improved by a computer; among the different characteristics of the response of the filter, a remarkable width of the stopband may be obtained.

Linear-Phase Transmission-Line Filters With

Abstract: &straci-A new class of transfer functions approximating a constant group delay that can be realized as transmission factors of doubly loaded transmission-line filters of commensurate line lengths is introduced. The magnitude response obtained is more general than the analogous response of filters having a maximally flat type of approximation to a constant delay. This is due to an additional parameter that may be used to obtain a magnitude response having either better passband performance or sharper cutoff than the maximally flat type of delay approximation. Explicit formulas for the evaluation of the coefficients of transmission factors of any order n are derived. HE SYNTHESIS of doubly loaded transmissionline filters approximating a constant delay has been the subject of a number of recent papers [l]-[4]. Scanlan and Rhodes [I] have presented a method of synthesis of a class of microwave network functions, which is known to yield a constant group delay at all frequencies but with rather poor magnitude response outside the useful band. For this reason their method is more suitable for the design of broad-band matching transformers than for the bandpass filters. The procedure proposed by Carlin and Zysman [2], in which the Hilbert transformation is used, is based on the approximation of the magnitude and not the delay characteristic around 0 = 0 so that the accuracy of delay approximation cannot be readily determined beforehand. An exact solution to the problem of determining transmission factors of a class of transmission-line filters that approximate a constant delay in a maximally flat manner has been derived by Abele [3] and explicit formulas and design curves have also been provided in order to facilitate the practical design. By this method the stopband performance can be considerably improved, especially if higher values of a parameter, which, in fact, represents the zero frequency delay in the transformed frequency plane, are chosen. However, the problem is compounded because the design technique that improves the stopband performance tends to increase the passband attenuation of the filter. It may be concluded that, from the point of view of the synthesis of linear phase bandpass filters, the maximally flat approximation of Abele offers considerable advantage over the constant delay function used by SC&Alan and Rhodes [l] but it still leads to better than desired delay response at the expense of the quality of the magnitude response.