scispace - formally typeset
Search or ask a question

Showing papers on "Stopband published in 1994"


Journal ArticleDOI
TL;DR: A novel approach to the design of M-channel pseudo-quadrature mirror filter (QMF) banks is presented, where it is possible to design a pseudo-QMF bank where the stopband attenuation of the analysis (and thus synthesis) filters is on the order of /spl minus/100 dB.
Abstract: A novel approach to the design of M-channel pseudo-quadrature mirror filter (QMF) banks is presented. In this approach, the prototype filter is constrained to be a linear-phase spectral-factor of a 2Mth band filter. As a result, the overall transfer function of the analysis/synthesis system is a delay. Moreover, the aliasing cancellation (AC) constraint is derived such that all the significant aliasing terms are canceled. Consequently, the aliasing level at the output is comparable to the stopband attenuation of the prototype filter. In other words, the only error at the output of the analysis/synthesis system is the aliasing error which is at the level of stopband attenuation. Using this approach, it is possible to design a pseudo-QMF bank where the stopband attenuation of the analysis (and thus synthesis) filters is on the order of /spl minus/100 dB. Moreover, the resulting reconstruction error is also on the order of /spl minus/100 dB. Several examples are included. >

233 citations


Proceedings ArticleDOI
28 Nov 1994
TL;DR: It is shown that without any pre-receiver processing, the level of interchannel interference is lower in the cosine-modulated family of filters, because of their higher stopband attenuation.
Abstract: This paper examines the comparative performance of DFT and cosine modulated filter banks used in multicarrier signaling over a linearly distorted channel, affected by both white Gaussian and NEXT noise. The interference caused by the overlap in the frequency responses of the signaling filters is calculated analytically and simulation results are given to support the theoretical calculations. It is shown that without any pre-receiver processing, the level of interchannel interference is lower in the cosine-modulated family of filters, because of their higher stopband attenuation. We also examine how pre-receiver processing (short equalizer plus cyclic prefix) can lower the level of interference, especially in the DFT system. Finally the capacity of a sample channel, similar to the twisted-pair channel of a subscriber loop, is calculated with and without the effect of interference taken into consideration.

65 citations


Journal ArticleDOI
01 Dec 1994
TL;DR: In this article, a stereo sigma delta A/D-converter for audio applications is presented, where two identical cascaded fourth-order sigma-delta modulators and a sophisticated multistage linear-phase FIR decimation filter with oversampling ratio of 64 are implemented on the same die.
Abstract: A stereo sigma delta A/D-converter for audio applications is presented. In this converter, two identical cascaded fourth-order sigma-delta modulators and a sophisticated multistage linear-phase FIR decimation filter with oversampling ratio of 64 are implemented on the same die. The analog part is designed to operate at a low voltage with a low power consumption. Techniques to achieve simultaneously a high performance and a low power consumption are discussed in details. The minimum stopband attenuation of the decimator is more than 120 dB and the passband ripple of the overall converter is less than 0.0003 dB. The first decimation stage is a special tapped comb filter, whereas the remaining stages are realized without general multipliers by simultaneously implementing all the filter coefficients by using special bit-serial networks. For the integrated overall stereo converter, the power consumption and the signal-to-noise ratio are 180 mW and 97 dB (85 mW and 95 dB) for a 5 V (3 V) power supply. The circuit die area is only 4.7 mm/spl times/5.5 mm using a 1.2 /spl mu/m double-poly BiCMOS process. >

43 citations


Patent
20 Sep 1994
TL;DR: In this article, a filter with an adjustable bandwidth was proposed, where a variable reactance element was used to adjust the bandwidth of the passband, coupled with the resonator structures.
Abstract: A filter (102) with an adjustable bandwidth. The filter (102) has a predetermined passband and stopband, and an input (104), an output (106) and at least two resonator structures (116, 124), and a variable reactance element (108) for adjusting the bandwidth of the passband, coupled between the resonator structures (116, 124), whereby the frequency response is adjustable.

37 citations


Proceedings ArticleDOI
01 Jan 1994
TL;DR: In this article, the authors proposed simple analytical expressions for the three above mentioned effects, as terms to be added to the COM model to improve the agreement between theory and the measured curves.
Abstract: Coupling-of-modes (COM) model is known to be very accurate and easy to use for the design of narrow band SAW filters and resonators. The use of the model is so successful that there is a temptation to expand the application of the model to other types of SAW,like leaky SAW, and to use it for the design of relatively wide band devices. Thus it is desirable to modify the COM model so that it gives acceptable results for wide band devices based on the use of leaky waves, while keeping the skeleton of the COM model unchanged. When applied directly to leaky wave, 3 to 5% bandwidth resonator filters, standard COM model has the following drawbacks: The dispersion, (i.e. the frequency dependence of the reflectivity, velocity changes and other COM parameters) must be taken into account; It doesn't take into account the increase of attenuation at frequencies higher than the stopband due to the scattering into the bulk waves; The generation of bulk waves, which occurs also at frequencies slightly higher than the stopband, results in an additional loss mechanism. In the present work we propose simple analytical expressions for the three above mentioned effects, as terms to be added to the COM model to improve the agreement between theory and the measured curves. Comparisons between this lpatchedr COM model and wideband devices on LiNbO3 show improved agreement

34 citations


Journal ArticleDOI
TL;DR: In this article, face-centred-cubic photonic crystals with overlapping stop bands are stacked in tandem to obtain an ultrawideband photonic crystal with a stop band that extends from 16 to at least 25 GHz and a minimum rejection of approximately 7 dB at normal incidence.
Abstract: Face-centred-cubic photonic crystals with overlapping stop bands are stacked in tandem to obtain an ultrawideband photonic crystal. This crystal has a stop band that extends from 16 to at least 25 GHz and has a minimum rejection of approximately 7 dB at normal incidence. Different stacking configurations have been studied, and cavity modes between nonadjacent crystals with overlapping stop bands have been observed. >

31 citations


Proceedings ArticleDOI
01 Jan 1994
TL;DR: In this paper, a new class of SAW filters, balanced bridge impedance element filters (BEMF filters), is proposed, which have numerous advantages: low insertion loss, extremely small size and weight, a wide variety of center frequencies and passbands, excellent suppression in the stopband and no external matching circuits.
Abstract: Over the last few years a new generation of Surface Acoustic Wave (SAW) devices with low loss and operating at high frequency have been invented to meet the demand of mobile communication systems. In this work we discuss a new class of SAW filters-balanced bridge impedance element filters,-which have numerous advantages: low insertion loss, extremely small size and weight, a wide variety of center frequencies and passbands, excellent suppression in the stopband and no external matching circuits

27 citations


Journal ArticleDOI
TL;DR: In this article, the poles of an elliptic filter are defined as a circle whose radius is equal to the geometric mean value of the passband and stopband edge frequencies, and the transfer functions of such filters are directly and easily defined by the assigned specifications.
Abstract: Elliptic filters have minimal Q-factors for a specified ripple factor determined by the value of the characteristic function at the stopband edge frequency. The poles of such a filter are on a circle whose radius is equal to the geometric mean value of the passband and stopband edge frequencies. The Q-factors are determined by means of a simple relation as a function of the stopband edge and zero loss frequencies. As a consequence, the transfer functions of such filters are directly and easily defined by the assigned specifications.

20 citations


Journal ArticleDOI
TL;DR: In this paper, the design and measured results of a prototype high-frequency fifth order switched-current bilinear elliptic low-pass filter were presented, which was implemented on a standard 1.2 /spl mu/m double-metal single-polysilicon CMOS process.
Abstract: Design considerations for and measured results of a prototype high-frequency fifth order switched-current bilinear elliptic lowpass filter are presented in this paper. The prototype filter was implemented on a standard 1.2 /spl mu/m double-metal single-polysilicon CMOS process and occupied a total die area of 1.5 mm/sup 2/. When clocked at 2 MHz, the filter achieved a passband edge of 350 kHz, a stopband edge of 420 kHz, a passband ripple of 0.6 dB and a minimum stopband attenuation of 26 dB. With a single 5 V power supply the filter consumed 28 mW. >

20 citations


Proceedings ArticleDOI
P.N. Heller1
25 Oct 1994
TL;DR: In this article, a new explicit formula for the construction of M-th band maximally flat lowpass filters based on Lagrange interpolation was developed, and the minimum length and arbitrary length solutions were obtained.
Abstract: This paper develops a new explicit formula for the construction of M-th band maximally flat lowpass filters based on Lagrange interpolation. Both the minimum length and arbitrary length solutions are obtained. This leads to the construction of M-band wavelet filters with N vanishing wavelet moments as spectral factors of the new M-th band filters. One can then vary the free parameters of the general (non-minimal length) M-th band filter to obtain filters with better stopband attenuation than the minimal length filters, and wavelet bases with greater smoothness than the corresponding minimal length solutions. Remarkably, in the M-band case (M>2), one can simultaneously improve the stopband attenuation of the lowpass filters and obtain smoother wavelet bases. >

15 citations


Proceedings ArticleDOI
01 Jan 1994
TL;DR: In this paper, a low-loss SAW filter in the range of 300 to 400 MHz for communication systems has been developed, which have ladder circuit structure using 36d rotated Y-cut LiTaO3.
Abstract: Low loss SAW filters in the range of 300 to 400 MHz for communication systems have been developed, which have ladder circuit structure using 36d rotated Y-cut LiTaO3. This type of filter has low loss and high stop band rejection, however, a spurious mode appears, which depends on electrode thickness and a reflector's electrode pitch of a resonator. Low frequency range filters need thick Al electrodes more than 7% of wave length to shift the resonant frequency of spurious mode to a frequency range higher than the pass-band frequency range. To reduce the spurious response, the design of an electrode pattern and electrode material having more mass effect than Al were investigated by simulation and experiment. The spurious response can be suppressed by optimizing a reflector's electrode pitch. By using Au film, thin film of about 1.5% of wave length can be applied to a filter. SAW filters in the range of 300 to 400 MHz with fractional band width from 2 to 4% were manufactured and mounted in SMD type 5t7t1.4 mm ceramic packages, which have 2 dB minimum insertion loss and attenuation characteristics of about 40 dB. Input and output impedance of filter are 50 O

01 Jan 1994
TL;DR: In this article, the coefficient bounds for half-band fixed-point polyphase filters with all zeros on the unit circle were established, and the algebraic details of the established coefficient interactions and the corresponding polo-zero-patterns (pzp) for the 5 th and 7 th order cases were shown.
Abstract: This paper reports on the establishment of the coefficient bounds for half-band fixed-point polyphase filters, having all their zeroes on the unit circle. The paper also reports on the algebraic details of the established bounds, exposing pictorially the coefficient interactions and the corresponding polo-zero-patterns (pzp) for the 5 th and 7 th order cases. Potential applications as well as sample filters, designed through the use of these bounds are also reported. Half-band filters having extremely stringently specified frequency domain magnitude response attributes, such as μdB passband ripple, with stopband attenuation in excess of 100dB can be effectively composed of parallel combinations of digital allpass recursive filter constructs in a polyphase environment. Furthermore, the coefficient accuracy requirements for such filters is

Proceedings ArticleDOI
02 Oct 1994
TL;DR: In this paper, a new frequency weighted stopband energy function is introduced, which leads to considerable flexibility in the design process and reduces the design equations to an eigenvector problem.
Abstract: The paper presents a method for the design of quadrature mirror filters. A new frequency weighted stopband energy function is introduced, which leads to considerable flexibility in the design process. Unlike other techniques which involve searches and nonlinear optimization, the formulation reduces the design equations to an eigenvector problem. The resulting filters are regular and have additional desirable properties as discussed in the paper. Also considered is application to pyramidal coding of images, which together with DPCM and PCM, leads to high compression ratios. >

Proceedings ArticleDOI
01 Jan 1994
TL;DR: In this paper, the relative importance of different parasitic effects in multi-transducer SAW filters at UHF frequencies was discussed by using a lumped-element circuit model which links the circuit elements directly to the filter-geometry and enables one to identify the sources of unwanted features in the filter frequency characteristics.
Abstract: This paper discusses the relative importance of the different parasitic effects in multi-transducer SAW filters at UHF frequencies. This is accomplished by using a lumped-element circuit model which links the circuit elements directly to the filter-geometry and enables one to identify the sources of unwanted features in the filter frequency characteristics. Further, the model complexity may be reduced by retaining only those circuit elements that lead to the most severe effects. It is shown how Ohmic losses in connection pads and bond wires, coupling capacitances between pads, and inductive effects due to the bond wires are modeled and how they contribute to the filter characteristics. An experimental interdigitated IDT (IIDT) filter consisting of 15 transducers and operational at 1 GHz serves as an example. Its passband distortions could easily be traced back to stray capacitances between pads or IDTs while the stopband characteristics are mainly determined by inductive coupling effects between the 17 bond wires. The latter result indicates that in multi-transducer devices like IIDT filters or image-impedance connected twin-track filters parasitic effects negligible in simpler structures are of paramount importance

Proceedings ArticleDOI
23 May 1994
TL;DR: In this article, an easy-to-implement and fast lumped-element circuit model for the parasitics of complex low-loss surface acoustic wave (SAW) devices such as interdigitated interdigital transducer (IIDT) and image impedance connected (IIC) filters is presented.
Abstract: This work presents an easy-to-implement and fast lumped-element circuit model for the parasitics of complex low-loss surface acoustic wave (SAW) devices such as interdigitated interdigital transducer (IIDT) and image impedance connected (IIC) filters. The model parameters can be derived from knowledge of the device geometry without recourse to optimization methods. We investigated the influence of the measurement setup, the package and the layout itself. It is shown how the model parameters are determined by either experiment or analysis. The greatly enhanced predictive power of the extended simulation model is demonstrated by applying it to a 15-transducer IIDT filter and a 22-transducer IIC filter at 900 MHz. It was found, for instance, that parasitic capacitances strongly influence the input impedance in the passband while the stopband level is easily explained by the inductive coupling between bond wires. >

Journal ArticleDOI
TL;DR: Simple formulae are presented for designing filters based on repeatedly applied moving average operations with unit coefficients to synthesise the filter in a way that satisfies specified passband and stopband specifications.
Abstract: Simple formulae are presented for designing filters based on repeatedly applied moving average operations with unit coefficients. Design formulae are derived to synthesise the filter in a way that satisfies specified passband and stopband specifications. These filters are attractive because of the reasonable frequency characteristics, the computational efficiency of the design and filter algorithms, and the uncomplicated implementation in software.

Proceedings ArticleDOI
01 Oct 1994
TL;DR: In this paper, the authors describe the design and performance of a new BP filter using λg/2 slotline and ρ/4 CPW short-end resonators.
Abstract: This paper describes the design and performance of a new BP filter using λg/2 slotline and λg/4 CPW short-end resonators. A compact circuit, with an optimized geometry of the slotline elements, to reduce radiation loss, has been fabricated and tested. The experimental results with a center frequency of 8 GHz and a bandwidth of about 50% are shown, and compared to the simulation of a simple transmission line electrical model; a very good agreement is obtained between them over a very large bandwidth, as well in the passband as in the stopband.

Journal ArticleDOI
TL;DR: In this paper, a new type of rectangular waveguide ring cavity has been developed for applications to resonators, filters, and multiplexers, which have advantages of high Q and high power handling capability.
Abstract: A new type of rectangular waveguide ring cavity has been developed for applications to resonators, filters, and multiplexers. The circuits have advantages of high Q and high power handling capability. H-plane and E-plane waveguide ring cavities have been investigated thoroughly in single-mode and dual-mode operations. For single-mode operation, regular resonant modes, split resonant modes, and forced resonant modes were explained by resonant mode charts of E-field points. Mechanically tuned and electronically tuned resonators built by adjusting the resonant modes between regular resonant modes and forced resonant modes have been demonstrated. For dual-mode operation, a dual-mode filter using a single H-plane ring cavity has been built with a bandwidth of 0.77%, a stopband attenuation of more than 40 dB, and a sharp gain slope transition. Another dual-mode filter which was formed by cascading two E-plane ring cavities has also been fabricated with a bandwidth of 1.12%, a stopband attenuation of 70 dB, a mode purity of 2 GHz at the center frequency of 26.82 GHz, and a sharp gain slope transition. >

Journal ArticleDOI
TL;DR: In this article, the effect of disordered GaAs/(Al,Ga)As distributed Bragg reflectors (DBRs) on thermal annealing of semiconductor DBRs is studied by simulating interdiffusion of Al and Ga at the interfaces within the DBR.
Abstract: The variation of the reflectivity and stopband width of disordered GaAs/(Al,Ga)As distributed Bragg reflectors (DBRs) is studied by thermal annealing of semiconductor DBRs and modeled by simulating interdiffusion of Al and Ga at the interfaces within the DBR. Undoped mirrors show stability in their reflectivity and stopband width for annealing durations of up to 24 h. Heavily Si‐doped mirrors show strong degradation for annealing durations as short as 1 h. Using the model, excellent agreement with experimental results is found and interdiffusion coefficients can be estimated. The results show that important optical parameters are unchanged for short diffusion lengths, but drastically degrade for larger diffusion lengths. The results indicate the limits of thermal processing for device structures containing GaAs/(Al,Ga)As DBRs. The ability to accurately predict the effects of disordering on DBRs permits the proper design of vertical cavity structures containing DBRs, by specifically accounting for the effe...

Proceedings ArticleDOI
01 Jan 1994
TL;DR: In this paper, two types of network coupled SAW resonator filter are presented, one is a low loss, small transition width (shape factor of 1.4 from -3 dB to -28 dB) IF filter for DECT system and the other is a single chip duplexer filter for the CT1+ system.
Abstract: The dynamic market for compact mobile communication systems has greatly stimulated new SAW design techniques, especially SAW coupled resonator filters. Purely acoustical coupled SAW resonator filter can offer very low loss in the passband but the performance is not good enough for stopband rejection and transition width. Usually, electrical coupling is helpful in this respect and the required performance can be achieved by properly designed coupling networks. In this paper two types of network coupled SAW resonator filter are presented, One is a low loss, small transition width (shape factor of 1.4 from -3 dB to -28 dB) IF filter for DECT system and the other is a single chip duplexer filter for the CT1+ system. Both devices have losses of less than 4 dB over moderate bandwidths and high stopband rejections (g45 dB). The duplexer also requires large separation between RX and TX channels without any matching elements. It is well known that the coupling-of-mode theory is a good tool for analysis and design of these kinds of filters. However, since the substrate employed is 36dYX-LiTaO3, which has the required strong coupling to leaky surface modes, and the velocity of leaky surface mode is very close to that of fast shear bulk wave, there exists strong coupling between two modes within the metal strip periodic structure. This implies modifications to the conventional COM theory in order to accurately model the device operation. In this paper theoretical analysis results based on a modified coupling-of-modes model for both filters are compared with the measured results. Reasonable agreement has been obtained

Journal ArticleDOI
TL;DR: In this paper, the variation of reflectivity, stopband width, and penetration depth of disordered GaAs/(Al,Ga)As distributed Bragg reflectors (DBRs) was analyzed by simulating interdiffusion of Al and Ga at the interfaces within the DBR.
Abstract: The variation of the reflectivity, stopband width, and penetration depth of disordered GaAs/(Al,Ga)As distributed Bragg reflectors (DBRs) is analyzed by simulating interdiffusion of Al and Ga at the interfaces within the DBR The results show that important optical parameters are unchanged for short diffusion lengths, but drastically degrade for larger diffusion lengths The results indicate the limits of thermal processing for device structures containing GaAs/AlGaAs DBRs

ReportDOI
16 Mar 1994
Abstract: The harmonic components of magnetic field imperfections in the AGS Booster has been determined through careful measurements of the field corrections required to compensate imperfections which drive various transverse resonances. An analysis of the required correction yielded phase and amplitude information which points to possible locations and strength of field imperfections, and the dependence of these corrections on the bending field (B), dB/dt, and the mean closed orbit radius (dR) suggested possible sources of imperfections. In particular the dependence of the required correction on dB/dt strongly suggested an error of the windings on the dipole vacuum chambers which compensate quadrupole and sextupole fields produced by the eddy-currents. The dependence of the required quadrupole and the sextupole correction on dR indicated the presence of sextupole and octupole imperfections of the same harmonics. The observations also suggested the presence of a strong harmonic remnant field. I

Proceedings ArticleDOI
30 May 1994
TL;DR: In this paper, the authors generalized the classical recursive digital filters such that the resulting filters are implementable using low-sensitivity and low-noise structures originally developed for classical filters.
Abstract: The class of classical recursive digital filters is generalized such that the resulting filters are implementable using low-sensitivity and low-noise structures originally developed for classical filters. These filters can be synthesized to have several passband and stopband regions and arbitrary weightings can be used in these regions. In addition, some transmission and attenuation zeros can be fixed. An efficient algorithm is constructed for optimizing the proposed filters based on iteratively adjusting the passband and stopband responses of the filter with the aid of a Remez-type algorithm. Several examples are included illustrating the usefulness of the proposed generalized filters as well as the efficiency and flexibility of the algorithm. >

Proceedings ArticleDOI
01 Oct 1994
TL;DR: In this paper, a 4-port short-line meshes (SLM) stacked in a multilayer MMIC structure is proposed to act as a point connection at low frequencies and as a growing frequency dependent connection in higher frequency regions.
Abstract: Miniature low-pass and hand-pass filters that use novel 4-port short-line meshes (SLM) stacked in a multilayer MMIC structure are proposed. The SLM is a periodic structure of very short TFMS lines. It acts as a point connection at low frequencies and as a growing frequency-dependent connection in higher frequency regions. This feature is effectively combined with inductors and capacitors to build a 4-GHz low-pass filter and a 8-GHz band-pass filter with various stop band characteristics on a 1.4 mm × 1.2mm chip.

Proceedings ArticleDOI
01 Jan 1994
TL;DR: In this paper, the first reported laser probe measurements of leaky SAW propagation on both 41d and 64d rotated-Y X-propagate lithium niobate substrates are presented.
Abstract: This paper presents the first reported laser probe measurements of leaky SAW propagation on both 41d and 64d rotated-Y X-propagate lithium niobate substrates. A conventional specular reflection based Rayleigh wave laser probe is used in conjunction with standard signal processing techniques. The laser probe detects the net vertical surface motion on the SAW substrate where both Rayleigh and leaky SAW components are present. This motion can also be contributed by bulk waves that are reflected from the bottom of the substrate. These components are resolvable in the time domain owing to their different time-of-arrival when a stationary probe spot is placed at a distance away from the launching transducer. Their identification can also be verified by examining the corresponding time-gated frequency domain responses. The on-axis propagation velocity of the Rayleigh and leaky wave components as well as additional evolutionary behavior are obtained by probing along the principal SAW propagation path. One example is the direct confirmation of a secondary propagation path in both substrates. This path results from radiation into the bulk that is in turn reflected from the bottom surface of the substrate and, upon reaching the top surface, regenerates both leaky and Rayleigh waves. A second example is the quantitative determination of the leaky wave stopband as well as the leaky-Rayleigh scattering stopband in a reflection grating structure. Also, longitudinal profiles measured from leaky wave based 1-port SAW resonator allowing possible deduction of propagation attenuation inside the stopband and the dispersion curve are presented

Journal ArticleDOI
TL;DR: A modified low-pass filter functions are proposed, which are realizable in the doubly-terminated ladder structures for the ordern even or odd, thus lending themselves amenable to high-quality activeRC or switched capacitor filters through the simulation techniques.
Abstract: A modified low-pass filter functions are proposed. A modified inverse Chebyshev function possesses progressively diminishing ripples in the stopband whereas the modified Chebyshev function exhibits ripples diminishing towardω = 0 in the passband. Both are realizable in the doubly-terminated ladder structures for the ordern even or odd, thus lending themselves amenable to high-quality activeRC or switched capacitor filters through the simulation techniques.

Proceedings ArticleDOI
01 Jun 1994
TL;DR: In this paper, the authors presented low loss ring SAW filters on 49/spl deg/YX, 64/spl yX, 128/spl YX LiNbO/sub 3/ realized on the basis of the reflecting multistrip couplers (RMSCs).
Abstract: This paper presents low loss ring SAW filters on 49/spl deg/YX, 64/spl deg/YX, 128/spl deg/YX LiNbO/sub 3/ realized on the basis of the reflecting multistrip couplers (RMSCs). Using the RMSCs with 3 electrodes per /spl lambda/ (/spl lambda/-is the SAW wavelength at the center frequency) and self-resonance approach when filter specified real input/output impedances are achieved by IDT static capacitance compensation with radiation susceptance, ring filters have shown very low insertion losses of 0.8-1 dB, 3 dB fractional bandwidth of 2-5% with very low ripple of 0.1 dB stopband attenuation over 50 dB at 10-33% offset from the center frequency of 45 MHz. At 50 Ohm 148, 164, 172 MHz ring filters on 128/spl deg/YX for the low power transceivers have provided insertion loss of 1 dB, 1 dB bandwidth of 2-2.3 MHz, the stopband attenuation over 55 dB at 25 MHz offset from the center frequency. Two cascaded filters at 164.5 MHz have shown insertion losses below 3 dB and stopband attenuation over 90 dB. The chip size is 5/spl times/4/spl times/0.7 mm. >

Proceedings ArticleDOI
01 Jun 1994
TL;DR: In this paper, a new class of surface acoustic wave (SAW) filters called balanced bridge impedance element filters (BEEF) is introduced. But the BEEF filters require no external matching circuits and cannot handle power levels up to a few Watts.
Abstract: Over the last few years a new generation of surface acoustic wave (SAW) devices with low loss and operating at high frequency have been invented to meet the demand of mobile communication systems. Besides outstanding electrical characteristics, these filters need to have small size and weight, must be highly reproducible, need no tuning, must be very cheap and if necessary, able to handle power levels up to a few Watts. In this work we introduce a new class of SAW filters-balanced bridge impedance element filters-which have all these advantages: low insertion loss, extremely small size and weight, a wide variety of centre frequencies and passbands, excellent suppression in the stopband and require no external matching circuits. >

Journal ArticleDOI
TL;DR: In this article, a block-like distribution of transducers within the IIDT structure is studied in detail, which leads to almost identical impedances at the filter input and output.
Abstract: Wide band low-loss SAW filters employing interdigitated interdigital transducers (IIDT's) and suitable for cellular radio applications are presented. After an investigation of the many design parameters that have to be taken into account, a new block-like distribution of transducers within the IIDT structure is studied in detail. This distribution leads to almost identical impedances at the filter input and output. An automatic optimization algorithm is also described. In particular, the finger numbers and the withdrawal weighting functions employed to increase the stopband rejection have been optimized by this algorithm. Filters fabricated on 36/spl deg/ rot Y, X-LiTaO/sub 3/ exhibit fractional bandwidths of up to 5.2% with passband ripples smaller than 1 dB at a center frequency of 500 MHz. The minimum insertion loss and the stopband rejection obtained were 1.8 and 23 dB, respectively. >

Proceedings ArticleDOI
M. Bolle1
19 Apr 1994
TL;DR: An analytical solution is presented which solves simultaneously the approximation problems for attenuation and phase of the ideal migration filter in a way which is equivalent to the approximation of the wave equation by the McClellan transform.
Abstract: A closed form design method for migration wave digital filters (MWDF) is presented. Migration filters are 2-D fan shaped filters with a prescribed nonlinear phase in the passband. An analytical solution is presented which solves simultaneously the approximation problems for attenuation and phase. The attenuation is approximated in a maximally flat sense in pass- and stopband. The phase of the ideal migration filter is approximated in a way which is equivalent to the approximation of the wave equation by the McClellan transform. Efficient WDF structures, including a multiplier-free scheme, for the realization of MWDF are derived. >