Search

In this paper, a method is proposed to reduce harmonic fold back HFB problem of N-path filters, without increasing the input reference clock fCLK frequency. The HFB at the N-path filter is analyzed, and simple expressions are extracted to model this problem. Using the results of the analysis, an M-of-N-path filter has been proposed that behaves like an M×N-path filter in terms of HFB problem; however, the fCLK frequency of this structure is the same as an N-path filter. To demonstrate the feasibility of the proposed idea, a 3-of-4-path filter is designed, and its characteristics are compared with 4-path and 12-path filters by simulation. Impacts of different non-idealities like clock-phase error, mismatch, and parasitic capacitance are investigated. The transistor-level implementation of this filter is performed in 0.18µm Complementary Metal Oxide Semiconductor CMOS technology. The simulation results show that the filter has the pass-band gain of 17dB, tuning range of 0.2-1.2GHz, -3dB bandwidth of 25MHz, quality factor of 8-48, 18dB out-of-band rejection, 16dB rejection of the third harmonic of switching frequency fs, and the noise figure of 4.35dB using ideal Gm cells and 6.95dB for practical Gm cells. The strongest harmonic folding to the filter pass-band occurs around 11fs with the attenuation of 23.8dB. Each Gm cell draws about 12.4mA from 1.8V supply, and the out-of-band IIP3 and P1dB,CP are 17 and 4dBm, respectively. Copyright © 2016 John Wiley & Sons, Ltd.

This article presents the design and realization of a new bandpass filter (BPF) fabricated on multilayered substrates. The proposed BPF schematic is based on the traditional elliptic-function filter schematic. The novel schematic features a specific parallel section, which grants the new schematic the ability in reducing the undesirable value of inductors in serial arms of the conventional elliptic BPF circuit schematic. As an illustrative example, a prototype BPF with 800 MHz center frequency is designed and fabricated. The measured stop-band rejection is better than 45 dB up to 2 GHz, and a relatively narrow bandwidth is obtained. The measured result agrees very well with the electromagnetic (EM) designed response. © 2016 Wiley Periodicals, Inc. Microwave Opt Technol Lett 58:1891–1894, 2016

This article presents a tunable bandpass filter (BPF) which is comprised of a chebyshev stepped-impedance lowpass filter (LPF), three shorted stubs, three varactors, and some blocking capacitors. Detailed design of the LPF, the effects of the shorted stubs, the schematic of the tunable BPF, and the tuning strategies are investigated. By tuning the reversed-bias voltages of the varactors, which are shunted with the low impedance of the stepped-impedance LPF, the operating frequency of the filter is changed. After the tunable BPF is designed and simulated, one experiment filter has been fabricated and tested. The measurement results of the experiment filter show that the center frequency of the tunable filter can be reconfigured from 1.09 to 2.44 GHz if the reversed-bias voltages are altered within 0–30 V. © 2014 Wiley Periodicals, Inc. Microwave Opt Technol Lett 56:879–883, 2014

This paper presents a double-sharpened decimation filter based on the application of a Kaiser and Hamming sharpening technique for multistandard wireless systems. The proposed double-sharpened decimation filter uses a pre-droop compensator which improves the passband response of a conventional cascaded integrator-comb filter so that it provides an efficient sharpening performance at half-speed with comparison to conventional sharpened filters. In this paper, the passband droop characteristics with compensation provides –1.6 dB for 1.25 MHz, –1.4 dB for 2.5 MHz, –1.3 dB for 5 MHz, and –1.0 dB for 10 MHz bandwidths, respectively. These results demonstrate that the proposed double-sharpened decimation filter is suitable for multistandard wireless applications.

A bandpass filter using a single-slotted butterfly radial stub connected in shunt with a short-circuited section of a microstrip transmission line is proposed. A two pole filter based on the proposed structure has been designed, analyzed, and fabricated. A good agreement between the measured and the simulated results is observed. © 2007 Wiley Periodicals, Inc. Microwave Opt Technol Lett 49: 1872–1875, 2007; Published online in Wiley Inter-Science (www.interscience.wiley.com). DOI 10.1002/mop.22600

This paper describes a method of configuration of a complex RiCR filter with coupled Chebyshev characteristics. The frequency shift method is a well-known method of obtaining a complex filter from a real filter, in which the transfer characteristic of the real filter is shifted along the frequency axis. The complex coupled Chebyshev filter obtained by the frequency shifting of a coupled Chebyshev filter has an excellent attenuation characteristic, but the number of constituent elements for active realization is large and the circuit is complex. In the proposed method, the complex coupled Chebyshev filter is first transformed to a complex RiCR filter so that the number of constituent elements for active realization is reduced and the circuit is simplified. In this paper, a method is presented for transforming the complex coupled Chebyshev filter to a complex RiCR filter. The attenuation characteristics and the numbers of constituent elements of the complex filter designed by the proposed method, the complex coupled Chebyshev filter, and the conventional complex RiCR filter, are compared for confirmation of the effectiveness of the proposed method. © 2006 Wiley Periodicals, Inc. Electron Comm Jpn Pt 3, 89(5): 12–20, 2006; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/ecjc.20212

This paper proposes a transformerless shunt hybrid filter for harmonic compensation of a three-phase six-pulse diode rectifier, where the AC line-to-line voltage is 3.3 or 6.6 kV. The hybrid filter consists of a single tuned LC filter per phase and an active filter with a DC capacitor voltage as low as 300 or 600 V. The two filters are directly connected in series with each other without a transformer. The passive filter absorbs harmonic currents produced by the rectifier, whereas the active filter improves the filtering characteristics of the passive filter. The required rating of the active filter is much smaller than that of a conventional shunt active filter used alone. Another advantage is that no additional switching-ripple filter is required for the active filter because the LC filter acts not only as a tuned LC filter around the seventh-harmonic frequency but also as a switching-ripple filter around 10 kHz. A feedforward control scheme is also proposed to improve the active filter performance. Experimental results obtained from a 200-V, 5-kW laboratory system and simulation results of a 3.3-kV, 300-kW system confirm the validity and effectiveness of the system. The hybrid filter gives satisfactory compensation performance, thus allowing us to put it into practical use. © 2003 Wiley Periodicals, Inc. Electr Eng Jpn, 146(2): 54–65, 2004; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/eej.10236

In many applications of filter banks, a low-delay filter bank in which the system delay is small is desirable. In this paper, two new design methods are proposed for low-delay perfect-reconstruction filter banks. First, a design method is described for a two-channel perfect-reconstruction filter bank suitable for subband coding. In this method, a high-pass filter for perfect reconstruction is designed for an arbitrary low-pass filter. Next, a design method is described for an oversampling perfect-reconstruction DFT filter bank suitable for subband adaptive filter. In this method, a prototype filter for a perfect-reconstruction synthesizer is designed for the prototype filter for an arbitrary analyzer. The effectiveness of the present method is demonstrated by design examples. © 2001 Scripta Technica, Electron Comm Jpn Pt 2, 84(6): 30–39, 2001

The tracking filter in a radar estimates the true values of target motion parameters such as position, speed, etc., from the observed target location. A typical example is the tracking filter that uses the Kalman filter. For calculating load reduction, tracking filters that use an α-β filter or an α-β-γ filter, which are simplified forms of the Kalman filter, are widely used. Here, the α-β filter is used to calculate the position and speed of a moving target and the α-β-γ filter is used when the acceleration of the target is required. Stability is a requirement for useful tracking filters, and it is known that the α-β filter is always stable when its gains α and β lie in the range from 0 to 1. However, there are no reports about the α-β-γ filter. In this paper, the stability of the α-β-γ filter is clarified when it is used as a tracking filter. It is shown that the α-β-γ filter, in contrast to the α-β filter, is not necessarily stable when its gains α, β, and γ lie in the range from 0 to 1. However, it is shown that three types of α-β-γ filters derived from the Kalman filters etc. are stable. © 2000 Scripta Technica, Electron Comm Jpn Pt 3, 83(5): 102–116, 2000

An ultra-wideband microstrip filter with good out-of-band performance is designed and demonstrated. The filter design is based on a circuit model for an optimum short-circuited stub transmission-line filter whose unit elements or connecting lines are nonredundant. Substituting the connecting lines by the lowpass filter using stepped impedance hairpin resonator, the undesired spurious bands can be efficiently suppressed, and the designed filter length is reduced. According to the different requirements, the broadband bandpass and lowpass filter can be independently designed and modified, which makes the filter more convenient and easily implemented. © 2008 Wiley Periodicals, Inc. Microwave Opt Technol Lett 50: 1735–1737, 2008; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/mop.23517

In the design of nonuniform-band filter banks (NFB), each filter must be designed to satisfy the perfect reconstruction (PR) condition corresponding to division ratio for the system. A simple design method for nonuniform-band maximally decimated filter banks using two-dimensional (2-D) filter to realize two-channel system is presented here. A 2-D filter is designed using a McCl-ellan transform and one axis is fixed to arbitrary frequency in a 2-D plane. Then the frequency response that turns out to be a cross sectional response of 2-D filter, 1-D variable filter that has fixed frequency as a parameter can be realized. Using 2-D filters with complementary power, the lowpass filter and high-pass filter always satisfies power complementarily. the 2-D filter has been designed so that the filter pair satisfies the perfect reconstruction condition and it is adopted as the group of the necessary filters for NFB. In this method, the PR condition is not satisfied strictly but system simplification is achieved. Design examples, amplitude responses of them, and simulation results of the reconstruction of the signal are shown.

This paper proposed an m-derived ladder for combining surface acoustic wave (SAW) filters in a contiguous filter bank and described the fabricated and analytical results for this filter bank. In the conventional filter bank with a constant-k ladder, spurious responses caused by the SAW filter in the stopband are not sufficiently suppressed. In the proposed SAW contiguous filter bank with an m-derived ladder, a parallel resonant circuit is used. It consists of an inductor and a capacitor as the series element instead of an inductor, which is used as the series element of a constant-k ladder. By setting the attenuation pole caused by the resonance of the series element to the spurious frequency of the SAW filter, the spurious response of the SAW filter with the m-derived ladder are suppressed more than with the constant-k ladder. Then a 10-channel SAW filter bank was constructed and the effectiveness of the proposed SAW contiguous filter bank with an m-derived ladder was verified.

Rapid-rate filter with sand or anthracite as filter media was studied at filtration rates of 4.5 and 9.0 m3/m2-h. Different parameters were evaluated by adopting a proposed simplified model. A suitable combination of anthracite-sand as filter media was given by using corresponding filter parameters of single-medium filters and break-even optimization technique.

Two-dimensional (2D) filter banks are used in subband coding of images, and are systems which split a signal into the narrow bands and/or resynthesize it. This paper presents a method of constructing 2D filter banks using 2D prefilter and 2D polyphase filter banks. First, the input signal is divided into four kinds of band on the 2D plane by 2D prefilter with IFIR (Interpolated FIR) transfer function. Then the 2D prefilter is designed so that the four kinds of band become 1. These resulting output signals are band-split separately and synthesized by 2D DFT polyphase filter banks. Hence, its design method becomes very simple and both the required order and the number of multiplications can be reduced.

It is shown that the two-stage cascade construction is very effective in the realization of a digital filter with a sufficiently large attenuation in the stop-band. A design method for the two-stage cascade FIR filter with limited word-length is described. Compared with the direct construction, the word-length of the filter coefficient in the two-stage cascade FIR filter is reduced to less than 60 percent. In general, compared with the direct construction, the word-length of the filter coefficient is reduced in the cascade construction, while the order is increased. However, in the proposed method, the order of the two-stage cascade construction need not be very large, compared with the case of direct construction, being 1.5 times in the worst case. By imposing less strict limitation on the word-length of the filter coefficient, the order of the two-stage cascade construction FIR filter never exceeds 4/3 times that of the direct-type FIR filter. In the realization of a linear-phase, two-stage cascade FIR filter, lower-order minimum- and maximum-phase filters than the linear-phase filter can be applied as subfilters. This is effective in further reducing the order of the two-stage cascade FIR filter, thereby reducing the order to 0.8 to 0.9 times the case of the construction based on the linear-phase subfilter.

The second-order recursive fundamental section up to now, has larger coefficient sensitivities of the amplitude at the filter central frequency and the central angular frequency itself. This makes it difficult to realize high-Q bandpass characteristics. This paper proposes a new structure with a feedback loop containing sampled-data filter with medium-Q bandpass filter and a longtime delay element. When a second-order bandpass filter is used as the feedback filter, the transfer function of the new structure is a pseudo-second-order function. Then the approximate relation is derived between Q of the feedback filter and Q of the new structure filter. the validity of the expression is verified by numerical calculations. It is shown also that when realizing a bandpass filter of the same Q, to what extent the new structure is superior to the traditional second-order structure from the viewpoints of coefficient sensitivity and round-off noise. the design procedure for the new structure is shown also. In the digital filter, the long-time delay can be realized by using a register or a memory. In the analog sampled-data filter, it is realized by a charge-transfer device. Thus, the new structure can be realized by a digital filter, or by combining switched capacitor circuits and a charge-transfer device. Finally, an experimental example of a high-Q filter by the latter method is shown.

This paper proposes an ϵ-separating nonlinear digital filter (called an ϵ-filter). This filter is intended for effective filtering of low-amplitude noise superposed on the signal with sharp discontinuities and can be realized by combining a simple nonlinear element with a conventional linear filter. In this paper, the basic ϵ-filter and its modifications to a trend-adaptive filter and a two-dimensional filter are described. The effectiveness of the new filter is demonstrated by computer simulation. Some of its application to EEG analysis, image processing and coding are also presented.

A recursive ϵ-nonlinear filter that removes random noise from signals with sharp discontinuities is proposed. It consists of a linear recursive low-pass filter with the addition of simple nonlinear fuctions. the proposed filter with a smaller number of taps is as effective as the nonrecursive ϵ-filter repored previously. Methodologies, realizations, and stability problems of the proposed filter are discussed. It is shown that the filter must be realized in cascade form for stable operation. Realizations of the multistage recursive ϵ-filter should be used from consideration of stability.

This paper presents a construction method of a uniform filter bank using an N- path digital network. This filter bank is constructed by connecting an N-path digital network to DFT polyphase filter banks. First, the input signal is divided into two parts by an N-path digital network and its complement. These resulting output signals are separately band-split and synthesized by DFT polyphase filter banks. Hence, its design method becomes very simple and both the required order and the number of multi plications can be reduced. Also, there quired structure is not so complicated. Further, a design method of the low- pass digital filter using this digital filter bank is presented. the low-pass filter with cutoff near π/2 which cannot be obtained easily in the usual multirate digital filter, can be realized by using these filter banks.

A consideration of recent progress in the field of rapid sand filtration discloses a trend towards improvements in structural arrangement and functional design of the component parts of the plant, particularly through the introduction of numerous auxiliary units and equipment to strengthen the process as a whole and make it more flexible and reliable. It appears that more progress has been achieved in perfecting the functions of the pre-treatment and auxiliary treatment stages of filtration than in adding to the usefulness of the filter itself. Current sanitary engineering practice is unmistakably in favor of producing the best possible water before it reaches the filters. Generally speaking, the filter itself has remained about as it was ten or more years ago; and the fact that filtered water is now much improved, or is at least of more uniformly acceptable quality (especially from the taste point of view), is not so much because the filters are better or more efficient per se, as that the process in its entirety, preceding the filters, has been materially improved and brought under the operator's more nearly complete control. Fundamentally it has always been the filter's main task to polish up the water delivered to it by the pre-treatment works. This operation of refining relates primarily to elimination of residual turbidity and bacteria. In many cases even this latter function is somewhat incidental, as in water softening plants where excess lime treatment is employed, or in straight filtration plants where preand super-chlorination, followed by postor de-chlorination is practised in operation. During the past decade the filter has been relieved of doing very much work. Logically, then, a promising avenue of further advance in rapid sand filtration plant design would be to improve the filter unit itself and thus permit a more nearly balanced design of the purification works as a whole, from the economic and operating standpoints. In making the filter do more of the work than at present, the designer returns to one of the first principles in the cycle of development of sand filtration. Such plants as now dispense with coagulation and

attempting to offer a complete hydraulic analysis of filtration or filter washing, or to duplicate the many detailed studies of filtration theory and operation which have been published, this paper concerns itself with filtration media, underdrain systems, rates of filtration and surface wash systems. Only the design and construction of the filters themselves are discussed, and no consideration has been given to coagulation, sedimentation or other phases of water treatment.

Two-dimensional band-pass filters can be constructed by a simple extension of the theory of one-dimensional band-pass filters. Similarly to the one-dimensional analogue the shape of the two-dimensional filter is important in determining its effectiveness. The band-pass filter formulation can be further refined so that the filter will concentrate its rejection energies in certain areas of the ω, k plane. Such band-pass, band-reject filters are found by solving a set of simultaneous equations.

A frequency-locked RC active ladder filter is described, which is frequency-locked to a sinusoidal control tone. The tunable ladder filter is used as a frequency discriminator by using two of the filter branch voltages to generate a control voltage. The control voltage is then used to tune or frequency-scale the ladder filter transfer function to lock the filter to the frequency of the control tone. A simple linear frequency-signal control system is developed to describe the frequency-locked performance of the adaptive filter. The frequency-locked filter can be designed so that the sinusoidal control tone is rejected at the filter output, making it possible to adapt the filter transfer function during transmission of a message. A practical frequency-locked filter is given in which a frequency-locked third order elliptic lowpass filter is designed. The performance of the practical filter is analysed using a simple linear control system model.