Low-pass Transfer Function Research Articles

An elliptic continuous-time CMOS filter with on-chip automatic tuning

A voice-band continuous-time filter is described which was designed based on the technique of fully balanced networks and was fabrication in a 3.5-/spl mu/ CMOS technology. The filter implements a fifth-order elliptic low-pass transfer function with 0.05-dB passband ripple and 3.4 kHz cutoff frequency. A phase-locked loop control system fabricated on the same chip automatically references the frequency response of the filter to an external fixed clock frequency. The cutoff frequency was found to vary by less than 0.1% for an operating temperature range of 0-85/spl deg/C. The absolute value accuracy of the cutoff frequency was 0.5% (standard deviation). With /spl plusmn/5-V power supplies the measured dynamic range of the filter was approximately 100 dB.

IIR digital filters with equiripple stopband transmission

Recently, a simple technique of introducing transmission zeros into a low-pass all-pole filter transfer function to make its stop-band transmission equiripple has been described by Hazra [1]. It is shown that this technique represents a special case of a more general procedure which Provides not only Hazra's solution, but also additional filters which are advantageous with regard to the required number of multiplications per sample. The results are demonstrated with the aid of an example.

A novel two-amplifier universal active switched-capacitor filter

A switched-capacitor (SC) filter circuit realizing low-pass (LP), high-pass (HP), and bandpass (BP) transfer functions is described. This circuit uses only two operational amplifiers (OAs) and is bottom-plate stray-insensitive.

Universal multiple-output second-order digital filter

A universal, second-order, digital-filter structure is proposed, which realises simultaneously a lowpass, a highpass, a bandpass, a notch and allpass transfer function. The structure uses two unit-delays and four multipliers, and is quite attractive for VLSI implementation since a versatile universal chip can be fabricated, which can be used for the implementation of Butterworth, Chebyshev and elliptic digital filters as well as digital equalisers.

A novel circuit for supercapacitor simulation

A new and simple circuit configuration with only one operational amplifier (OA) and one capacitor is proposed for the simulation of a supercapacitor which is also known as a frequency dependent negative resistance. The simulation procedure is based on the one roll-off characteristic of the OA. A second order lowpass transfer function having low active and passive component sensitivities for its Q and ω0 is achieved with the addition of only one resistance to the simulated network structure and is tunable over a wide range of frequencies.

Regulation of Stochastic Systems with Wide-Band Transfer Functions

The design of a regulator for a system with random discontinuities in its linear perturbation model is considered. The perturbation model is assumed to have a wide-band transfer function, rather than a more conventional low-pass transfer function. The resulting design equations are more complicated than are their counterparts derived under the low-pass hypothesis. Approximations are discussed which can be used to reduce the computational burden in some cases. The analytical procedures are illustrated in an example which considers the synthesis of a steam temperature regulator for a solar-powered central receiver.

New grounded-capacitor realizations of a third-order low-pass butterworth characteristic using the DVCCS/DVCVS

Two new configurations employing the differential voltage-controlled current/voltage source (DVCCS/DVCVS) are proposed for the realization of either an inverting or a noninverting third-order low-pass Butterworth transfer function. Both structures use a minimum of passive components which are of equal value with the capacitors being grounded. The sensitivities of the network parameters are low.

Design of low-pass maximally flat RC-active filters with multiple real pole: The MURROMAF polynomials

A new class of low-pass all-pole transfer functions is defined, with maximally flat behavior at the origin. The main feature of this class is the introduction of a multiple real pole, in order to realize RC-active filters by cascading third-order sections and, if necessary, second-order sections. This solution yields a reduction in the number of OA's and, consequently, in power consumption and in SNR.

Active realization of nth-order low-pass transfer functions

A new active RC circuit realizing the general <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">n</tex> th-order low-pass voltage transfer functions using <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">n</tex> Norton (current differencing) amplifiers is presented. In this realization minimum number of capacitors and <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">n</tex> active elements are used, the element values being given in terms of the transfer function coefficients.

Maximally flat low-pass transfer function synthesis using continuous and discrete filters

Discrete linear uniformly sampled digital filters are often employed in situations where there exists some preliminary linear analog continuous filtering operation. The latter continuous filter transfer function <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">P(s)</tex> is usually a low-pass function that is required for such purposes as reducing the effects of aliasing, limiting the dynamic range of the signal prior to digitization, etc. The realization of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">P(s)</tex> is usually achieved quite independently of the realization of the subsequent discrete transfer function <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">H(z)</tex> , where <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">z = e^{sT}</tex> . In this contribution, it is shown that maximally flat solutions exist for the composite transfer function <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">P(s) \cdot H(z)</tex> , so that it is possible to employ a simple baseband prefiliter <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">P(s)</tex> which, combined with <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">H(z)</tex> , results in an overall maximally flat low-pass response.

Equal-ripple rational functions with multiple critical pole pair

A new class of equal-ripple rational functions is considered here for the design of lowpass transfer functions of RC active filters. Using a critical pole pair with a multiplicity greater than unity (as in the MUCROMAF and MUCROER polynomials), we can obtain lower pole Q-factors with this new class than with the classical Cauer functions.

Synthesis of RC active filters having zeros on the real frequency axis

The implementation, using negative impedance convertors, of filters having zeros on the real frequency axis presents certain difficulties, particularly with regard to Sensitivity. It is shown how these problems can be overcome by the use of an arrangement in which the zeros are realized by means of cascaded passive twin-T networks. The restrictions which are imposed by the use of this arrangement are considered. Low-pass transfer functions, restricted to be realizable by this arrangement, are optimized using a least-squares criterion over the filter pass-band. The optimum characteristics are shown to compare favourably with those of elliptic filters. The restricted filter has the further merit, of flexibility in tho choice of number and location of zeros and is suitable for use in all methods of RC active network synthesis.

Economical line filtering for business communication systems

Economical active realizations of low-Q, all-pole, low-pass transfer functions are presented. The realizations allow a unidirectional sequence of functional tuning by trimming only resistors. Also, they have low sensitivities to the passive element variations. Experimental results verified the ease of tuning, the low sensitivity, and the low dependence on temperature and power supply variations.

Time-multiplexed active filters

The inductorless realization of high-order voltage transfer functions is achieved by subdividing a conventional realization into identical memoryless active subnetworks; the realization is then achieved by time multiplexing one of the memoryless active subnetworks such that it effectively replaces the total number of active subnetworks. The result of this time-multiplexing operation is to create a new type of RC-active filter that exhibits many practical advantages, such as lower power dissipation and a considerable saving in circuit complexity. The synthesis technique is illustrated by means of the design of a sixteenth-order low-pass transfer function. Nonideal performance is analyzed.

Low-frequency compensation of FDNR low-pass filters

A solution to the problem of realizing an inductorless low-pass elliptic transfer function has been proposed in which grounded FDNR elements are realized. The practical bias requirements for a resistive branch to ground have resulted in nonideal low-frequency rolloff in the passband. A simple technique is proposed to prevent low-frequency rolloff, to ensure direct-coupled operation, and at the same time to satisfy the practical constraints imposed by the bias requirements of the operational amplifiers.

Active realization of nth-order low-pass transfer functions

Using n differential-input operational amplifiers, two active RC circuits realizing the general nth-order low-pass transfer functions are presented. These realizations are achieved to within a constant multiplier greater or smaller than unity, respectively.

Constraint removal for Huelsman's equal-valued-capacitor active RC circuit

Making use of the well known Hurwitz conditions, it is shown that Huelsman&apos;s equal-valued-capacitor active RC circuit may realise not merely the Butterworth maximally flat characteristic, but any kind of 3rd-order lowpass transfer function, regardless of the precise location of the three left-halfplane poles.

Erratum: Design of an active distributed RC lowpass filter using exponential lines

A new active distributed RC structure, utilising exponential lines, is proposed for realising lowpass transfer functions. Explicit design equations are given for realising a pair of complex poles using this structure. This is based on a dominantpole approximation of the structure. It is shown that the pole Q sensitivity decreases with increasing taper and that it may be made considerably lower than that of Kerwin's structure which uses uniform lines.

Design of an active distributed RC lowpass filter using exponential lines

A new active distributed RC structure, utilising exponential lines, is proposed for realising lowpass transfer functions. Explicit design equations are given for realising a pair of complex poles using this structure. This is based on a dominantpole approximation of the structure. It is shown that the pole Q sensitivity decreases with increasing taper and that it may be made considerably lower than that of Kerwin&apos;s structure which uses uniform lines.

Transient response equalization†

Hitherto, phase response equalization and transversal filters have been used to improve transient performance. This correspondence describes attempts at improving the step response of low-pass transfer functions using all-pass transfer functions.