Showing papers on "Filter design published in 1987"

A theory of multirate filter banks

Abstract: Multirate filter banks produce multiple output signals by filtering and subsampling a single input signal, or conversely, generate a single output by upsampling and interpolating multiple inputs. Two of their main applications are subband coders for speech processing and transmultiplexers for telecommunications. Below, we derive a theoretical framework for the analysis, synthesis, and computational complexity of multirate filter banks. The use of matrix notation leads to basic results derived from properties of linear algebra. Using rank and determinant of filter matrices, it is shown how to obtain aliasing/ crosstalk-free reconstruction, and when perfect reconstruction is possible. The synthesis of filters for filter banks is also explored, three design methods are presented, and finally, the computational complexity is considered.

Subband/Transform coding using filter bank designs based on time domain aliasing cancellation

Abstract: A new, oddly stacked, critically sampled, single side-band (SSB) [7] analysis/synthesis system based on Time Domain Aliasing Cancellation (TDAC) [1],[2] is described in this paper. The specifications for the analysis and synthesis filter responses are developed and a number of designs which satisfy the reconstruction requirements are described. The application of TDAC systems to Subband/Transform coding is also discussed and the objective performance of a 32 band coder using several different window designs is presented and compared with a coder based on Frequency Domain Aliasing Cancellation (FDAC) filter banks [3]-[5].

FIR-median hybrid filters

Abstract: This paper introduces a new generalized class of median-type filters which we call FIR-median hybrid (FMH) filters. The input signal x(n) is filtered with M linear phase FIR filters, and the output of the FMH filter is the median of the outputs of the FIR filters. The statistical properties of the FMH filters are analyzed for input signals with Gaussian, double exponential, and uniform density functions. It is shown that FMH filters have essentially the same statistical properties and same type of root signals as the standard median (SM) filters. FMH filters preserve corner points and ramps and attenuate impulsive-type noise components effectively. An interesting subclass of the FMH filters requiring only one scaling multiplier, two additions, and three compare/swap operations irrespective of the filter length is introduced. In the examples, speed improvement by a factor of 450 is obtained over an otherwise approximately equivalent SM filter.

Design of Filters to Reduce Harmonic Distortion in Industrial Power Systems

Abstract: Shunt filters are effective in minimizing voltage distortion caused by nonlinear loads in industrial power systems. Different alternatives of filter design should be considered before making the final decision on filter configuration. Among the criteria used for performance evaluation are losses, current, and voltage ratings of each of the filter components, and the effect of filter and system contingency conditions. The design and performance of single-tuned and high-pass filters and the methodology used for the analysis will be discussed.

Theoretical analysis of the max/Median filter

Abstract: Median filtering has been used successfully for extracting features from noisy one-dimensional signals; however, the extension of the one-dimensional case to higher dimensions has not always yielded satisfactory results. Although noise suppression is obtained, too much signal distortion is introduced and many features of interest are lost. In this paper, we introduce a multidimensional filter based on a combination of one-dimensional median estimates. It is shown that threshold decomposition holds for this class of filters, making the deterministic analysis simpler. Invariant signals to the filter, called root signals, consist of very low resolution features making this filter much more attractive than conventional median filters.

Adaptive nonlinear filters for simultaneous removal of different kinds of noise in images

Abstract: A novel adaptive median filter is proposed. It allows the simultaneous removal of a combination of signal-dependent and additive random noise in addition to mixed impulse noise in images, processed in a single filtering pass. The adaptation algorithm is based on the local signal-to-noise ratio. An extension of the class of nonlinear mean filters to adaptive filters is considered. The performance of the adaptive median filter is compared to the commonly used median filter and the nonlinear mean filter.

A 'trick' for the design of FIR half-band filters

Abstract: Based on a well-known property of FIR half-band filters, this correspondence shows how the design time for equiripple half-band filters can be reduced by a considerable amount. The observation which leads up to this improved procedure also places in evidence new implementation schemes, which simultaneously ensure low passband and stopband sensitivities. Extension of the method to M th-band filter design is also outlined.

Recursive Nth-band digital filters- Part I: Design and properties

Abstract: We investigate the properties of a special class of recursive polyphase filters whose subfilters are distinct allpass filters. On the basis of the discovered properties, a Remez-type algorithm is constructed for designing these filters. The proposed algorithm is faster and gives higher filter selectivities than other existing methods. Better understanding of the filter properties allows us to select the branch filter orders such that the filter complexity is minimized. In addition to nonlinear-phase filters, the algorithm is applicable to the design of approximately linear-phase filters. Several examples are included illustrating the efficiency of the proposed design scheme. The characteristics of the resulting filters are exposed by means of experimental results, and the implementation of the allpass branch filters is discussed. In a companion paper [33], we use these polyphase filters as basic building blocks in constructing efficient filters for sampling rate alteration. In addition, it is shown how the don't care bands, which cannot be avoided in filters of this type, can be suppressed by using an additional correction filter stage.

First Principles of Discrete Systems and Digital Signal Processing

Abstract: 1. Introduction. Preview. Processing of Speech Signals. Processing of Seismic Signals. Radar Signal Processing. Image Processing. Kalman Filtering and Estimators. Review. References and Other Sources of Information 2. Signals and Systems. Preview. Types of Signals. Sequences Some Basic Sequences. Shifted and Special Sequences. Exponential and Sinusoidal Sequences. General Periodic Sequences. Sampling Continuous-Time Sinusoids and the Sampling Theorem. Systems and Their Properties. Linearity. Time-Invariance. Linear-Time Invariant (LTI) Systems. Stability. Causality. Approximation of Continuous-Time Processes by Discrete Models. Discrete Approximation of Integration. Discrete Approximation of Differentiation. Review. Vocabulary and Important Relations. Problems. References and Other Sources of Information. 3. Linear Time-Invariant Systems. Preview. Linear Constant-Coefficient Difference Equations. The Geometric Series--An Important Relationship. Difference Equations for Nth-Order Systems. Computer Solution of Difference Equations. System Diagrams or Realizations. Unit Sample Response. Convolution. A General Way to Find System Response. Computer Evaluation of Convolution. Analytical Evaluation of Convolution. An Application: Stability and the Unit Sample Response. Interconnected Systems. Cascade Connection. Parallel Connection. Initial Condition Response and Stability of LTI Systems. Forced and Total Response of LTI Systems. Review. Vocabulary and Important Relations Problems. References and Other Sources of Information. 4. Frequency Response and Filters. Preview. Sinusoidal Steady-State Response of LTI Systems. Frequency Response. Sinusoidal Steady--State Response--General Statement. The Nature of H(Ejq). Computer Evaluation of Frequency Response. Frequency Response from the System Difference Equations. Filters. A Typical Filtering Problem. Comparison of Two Filters. Ideal Filters. Interconnected Systems. Review. Vocabulary and Important Relations Problems. References and Other Sources of Information. 5. Frequency Response--A Graphical Method. Preview. Graphical Concepts. Geometric Algorithms for Sketching the Frequency Response. Graphical Design of Filters. Stability. Effects of Poles and Zeros on the Frequency Response. Correspondence Between Analog and Digital Frequencies. Some Design Problems. Review. Vocabulary and Important Relations Problems. References and Other Sources of Information. 6. Z-Transforms. Preview. Definitions. Right-Sided Sequences--Some Transform Pairs. Sample (Impulse) Sequence. Step Sequence. Real Exponential Sequence. Complex Exponential Sequence. General Oscillatory Sequence. Cosine Sequence. Properties and Relations. Linearity. Shifting Property. Multiplication by n and Derivatives In z. Convolution. Transfer Functions. Stability. Frequency Response Revisited. The Evaluation of Inverse Transforms. Inverse Transforms from the Definition. Inverse Transforms from Long Division. Inverse Transforms from Partial Fraction Expansions and Table Look-Up. Partial Fraction Expansion--General Statement. Checking Partial Fraction Expansions and Inverse Transforms. Solution of Difference Equations. Connections Between the Time Domain and the z-Domain. Poles and Zeros and Time Response. System Response to Some Special Inputs. General Results and Miscellany. Noncausal Systems. Convergence and Stability. The Inversion Formula. Review. Vocabulary and Important Relations. Problems. References and Other Sources of Information. 7. Discrete Fourier Transform. Preview. Periodic Sequences. Complex Exponentials. Discrete Fourier Series. Finite Duration Sequences and the Discrete Fourier Transform. Some Important Relationships. DFTs and the Fourier Transform. Relationships Among Record Length, Frequency Resolution, and Sampling Frequency. Properties of the DFT. Linearity. Circular Shift of a Sequence. Symmetry Properties. Alternative Inversion Formula. Duality and the DFT. Computer Evaluation of DFTs and Inverse DFTs. Another Look at Convolution. Periodic Convolution. Circular Convolution. Frequency Convolution. Correlation. Some Properties of Correlation Sequences. Circular Correlation. Computer Evaluation of Correlation. Block Filtering or Sectioned Convolution. Spectrum Analysis. Periodogram Methods for Spectrum Estimation. Use of Windows In Spectrum Analysis. Review. Vocabulary and Important Relations Problems. References and Other Sources of Information. 8. The Fast Fourier Transform. Preview. Decomposition in Time. Development of the Basic Algorithm. Computer Evaluation of the Algorithm. Decomposition in Frequency.Variations of the Basic Algorithms. Fast Convolution. Review. Vocabulary and Important Relations Problems. References and Other Sources of Information. 9. Nonrecursive Filter Design. Preview. Design by Fourier Series. Fourier Coefficients. Lowpass Design. Highpass, Bandpass and Bandstop Design. Gibbs' Phenomenon. Windows in the Fourier design. Design of a Differentiator. Linear Phase Characteristics. Comb Filters. Design by Frequency Sampling. Design Using the Inverse Discrete Fourier Transform. Frequency Sampling Filters. Computer-Aided Design (CAD) of Linear Phase Filter. Review. Vocabulary and Important Relations Problems. References and Other Sources of Information. 10. Recursive Filter Design. Preview. Analog Filter Characteristics Sinusoidal Steady-State. Frequency Response, Graphical Method. Computer Evaluation of Frequency Response. Determination of Filter Transfer Function from Frequency Response. Analog Filter Design. Butterworth Lowpass Prototype Design. Chebyshev Lowpass Prototype Design. Elliptic Lowpass Prototype Design. Analog Frequency Transformations. Design of Lowpass, Highpass, Bandpass, and Bandstop Filters. Digital Filter Design. Matched z-Transform Design. Impulse and Step-Invariant Design. Bilinear Transform Design. Digital Frequency Transformations. Direct Design of Digital Lowpass, Highpass, Bandpass, and Bandstop Filters. Optimization. Some Comments on Recursive and Nonrecursive Filters. Review. Vocabulary and Important Relations Problems. References and Other Sources of Information. 11. Structures, State Equations, and Applications. Preview. System Implementations. Direct Structure. Second-Order Substructures. Cascade Realization. Parallel Realization (Partial Fraction Expansion). Lattice Filters. Mason's Gain Rule. State Difference Equations. Writing State Equations. Solution of State Equations. Computer Solution of State Equations. Two Different Systems. Digital Control of a Continuous-Time System. Deconvolution Review. Vocabulary and Important Relations Problems. References and Other Sources of Information. Appendix A: Complex Numbers. Appendix B: Fourier Series. Appendix C: Laplace Transform. Appendix D: Frequency Response of Continuous-Time (Analog) Systems. Appendix E: .A Summary of Fourier Paris. Appendix F: Matrices and Determinants. Appendix G: Continuous-Time Systems with a Piecewise Constant Input. Answers to Selected Problems. Index.

Design of FIR filters as a tapped cascaded interconnection of identical subfilters

Abstract: A general theory is presented for the design of linear-phase FIR digital filters as a tapped cascaded interconnection of identical FIR subfilters. The approach is an extension of the Kaiser-Hamming procedure [1] proposed for sharpening the response of an FIR filter. The new approach allows the subfilter and the tap coefficients to be simultaneously optimized to minimize either the number of subfilters for the given order of the subfilter or the subfilter order for the given number of subfilters. The optimization is based on the use of standard FIR filter design algorithms. Several examples demonstrate how the new approach leads to implementations requiring significantly fewer distinct multipliers than equivalent direct-form minimax FIR designs at the expense of a slight increase in the overall filter order. The number of distinct multipliers can be reduced to approximately \sqrt{2.6L} , where L is the order of the direct-form minimax design. Alternatively, the design of the subfilter and tap coefficients can be separated. This makes it possible to construct the subfilter so that it roughly meets the overall specifications with a highly reduced number of arithmetic operations. In this case, the tap coefficients are optimized to minimize the required number of subfilters to meet the given criteria. Even multiplier-free designs can be obtained by carefully constructing the subfilter and determining the tap coefficients. Several structures are discussed for implementing the overall filter. These structures are compared with each other and with equivalent directform minimax designs in terms of the number of distinct multipliers, overall filter order, overall multiplication rate, number of delay elements, coefficient sensitivity, and output noise variance.

Reduced-order estimation Part 1. Filtering

Abstract: This paper presents a method for designing an ‘optimum’ unbiased reduced-order filter. For the proposed approach to work, the order of the filter must be greater than a certain minimum determined by the number of independent observations of the system available. The filler is much like a Luenberger observer for the state to be estimated, but with parameters optimized with respect to the noises in the system. A reduced-order innovation process is proposed that has properties similar to those of the full-order innovation process when the reduced filter is optimized. The approach offers the possibility of significant reduction in real-time computational requirements compared with the full-order filter, though at the cost of some loss of performance. The algorithm for the reduced-order filter is simple to implement— quite similar to that of the Kalman filter. An example is presented to compare the performance of the proposed method with the full-order Kalman filter.

Truncated subband coding of images

Abstract: Image data is analyzed in a number of iterated analysis procedures, using two-dimensional quadrature mirror filters to separate low-pass spatial filter response component and three differently oriented high-pass spatial filter response components, which filter response components are decimated in both dimensions. The high-pass filter response components are coded as is. The low-pass filter response component is coded as is only in the last iteration; in the earlier analysis procedures the low-pass filter response component provides the input data for the secceeding analysis procedure.

On root structures of median and median-type filters

Abstract: In this correspondence, we analyze the root structures of the standard median (SM) filters, the recursive median (RM) filters, and the FIR median hybrid (FMH) filters. It is shown that these filters have many oscillatory infinitely long root signals. When a section of an oscillatory root is present in a signal, the filter's noise attenuation of the filter is not as good as predicted by statistical measures. We also suggest methods to prevent the possible harmful effect caused by the existence of these roots.

Polyphase filter matrix for rational sampling rate conversions

Abstract: The polyphase filter array has been used for efficient implementations of filters with integer sampling rate conversions. [1] The filter in the high sampling rate side is decomposed into its polyphase filters which can be moved to the lower sampling rate side without changing their functions. For FIR filters the computational complexity is reduced by a factor equal to the sampling rate ratio. A rational (L/M) sampling rate conversion system realized with a 1-to-L interpolator followed by an M-to-1 decimator has three sampling rates F, LF and (L/M)F involved. By using the polyphase filter array a filter operating at the sampling rate of LF can be implemented in either the input side or the output side with lower sampling rates. The polyphase filter matrix structure will operate at the sampling rate of F/M, which does not show in the above model and is lower than any one of those three rates. For FIR filters the computational complexity is reduced by a factor of LM compared to the direct realization of the integral filter or by a factor of M (or L) compared to the polyphase filter array realization while the system input-output relation is maintained.

Efficient IIR and FIR fan filters

Abstract: A procedure for designing computationally efficient FIR and IIR fan filters is developed. The fan filters are derived from one-dimensional prototypes which are used in changing sampling rates by a factor of two. The transfer function B(z) of the prototype is expressed as B(z) = T_1(z^2)+zT_2(z^2) , which leads to a simple expression for the fan filter transfer function. The relation between the passband and stopband deviation requirements of the prototype and the fan filter is derived and employed in the design for meeting two commonly given sets of specifications. A special class of elliptic filters with zeros on the imaginary axis in the z plane is shown to yield computationally efficient IIR fan filters. FIR half-band filters are shown to be suitable one-dimensional prototypes for FIR fan filter design. In this case, symmetry constraints on FIR fan filter impulse response are automatically satisfied.

Simplified analysis of the LMS adaptive filter using a transfer function approximation

Abstract: A simple approximation to the behavior of the LMS adaptive filter as a discrete transfer function is developed. This representation is a valid description for both deterministic inputs and for the expected results with random inputs (including correlated inputs). The results are shown to be exact for some classes of input including periodic signals. One result of this analysis is the demonstration that the LMS filter can produce results which are biased from the least-squares solution under the combined conditions of a nonzero mean primary and a correlated reference input.

Multiplierless decimating low-pass filter for a noise-shaping A/D converter

Abstract: A method and means for filtering the quantization noise from the output of a 1-bit analog-to-digital converter (noise-shaping coder) is disclosed. The A/D utilizes oversampling of the analog input signal, and decimation of the digital output signal. The multiplierless low-pass filter is comprised of a coefficient ROM for storing the filter coefficients, a counter for addressing the memory, and a true/complement gate for selectively complementing the output of the memory in response to the 1-bit data stream from the noise-shaping coder. An accumulator sums the selectively complemented output words for all samples, and the accumulated output is then applied to the decimator. A second embodiment is also disclosed which utilizes an overlapping digital filter approach, wherein a plurality of digital multiplierless filters are overlapped to provide an arbitrary length filter capable of producing an arbitrary filter response.

Implementation of real coefficient digital filters using complex arithmetic

Abstract: Complex coefficient digital filters, with applications for processing real sequences, are examined. The method is quite general, and allows any real rational transfer function to be expressed in terms of a complex rational transfer function of reduced order. When implemented in complex hardware form, the reduction of filter order can provide an increase in computational efficiency and speed. Conventional filter structures, such as parallel, cascade, lattice, and state-space forms, are extended to the complex domain. Illustrative examples of complex coefficient filter synthesis are included, along with coefficient sensitivity comparisons between the complex coefficient filters and their real coefficient counterparts.

Tree-structured complementary filter banks using all-pass sections

Abstract: Tree-structured complementary filter banks are developed with transfer functions that are simultaneously all-pass complementary and power complementary. Using a formulation based on unitary transforms and all-pass functions, we obtain analysis and synthesis filter banks which are related through a transposition operation, such that the cascade of analysis and synthesis filter banks achieves an all-pass function. The simplest structure is obtained using a Hadamard transform, which is shown to correspond to a binary tree structure. Tree structures can be generated for a variety of other unitary transforms as well. In addition, given a tree-structured filter bank where the number of bands is a power of two, simple methods are developed to generate complementary filter banks with an arbitrary number of channels, which retain the transpose relationship between analysis and synthesis banks, and allow for any combination of bandwidths. The structural properties of the filter banks are illustrated with design examples, and multirate applications are outlined.

System modeling using an adaptive delay filter

Abstract: In this paper, we present a special technique for modeling an unknown system. This technique requires a type of adaptive filter called an Adaptive Delay Filter [1]-[3]. This filter structure includes variable delay taps in addition to variable gains. The Adaptive Delay Filter is especially applicable to system modeling problems in which the system to be modeled has a sparse impulse response [4]-[7]. Using the standard adaptive filter to model a sparse system could require a very large filter [8], while an Adaptive Delay Filter could model the sparse impulse response with very few elements in the filter since the delay taps spread out to adapt to the unknown system. An analysis of the mean-squared error surface using this technique is included, along with the computer simulation results of the performance in modeling both the delay taps and the gains of an unknown system. A comparison of this technique with the conventional approach using Widrow's LMS algorithm will be addressed.

Digital demodulator for frequency-division-multiplexed signals

Abstract: An improved digital demodulator capable of demodulating a frequency multiplexed input signal is disclosed. The frequency division multiplexed input signal is sampled in an analog-to-digital converter. The samples are translated by mixing with base band frequency signals to yield real and imaginary values corresponding to phase information in the original modulation signals. After translation, the translated samples are filtered in real and imaginary digital filters. After filtering, the original modulation information is recovered by analysis of the position of vectors in the complex plane represented by the real and imaginary values. In the preferred embodiment, a successive approximation technique is used to determine the angle of the vectors to the real axis. Low pass output filtering may be employed after recovery of the modulation signal for improvements in signal to noise ratio. In the preferred embodiment, the translation is performed by multiplying the input samples by digital values corresponding to sine and cosine values of local oscillator signals at base band frequencies. Preselect filtering may be employed prior to translation to decimate the input samples where possible to reduce subsequent processing requirements. In the preferred embodiment, the unit is operated in a first initialization mode in which a microprocessor supplies appropriate filter coefficients corresponding to the frequencies of the channels to be demodulated. Thereafter, the microprocessor does not directly control supply of samples or coefficients to the filters for evaluation. Very high data rates are thus made possible.

Iterative technique for the synthesis of distortion-invariant optical correlation filters.

Abstract: A procedure is presented for designing distortion-invariant correlation filters. Optical correlation filters designed using this technique retain full position invariance. The filter design begins by finding the distortion-invariant modes (eigenfunctions) for a particular image. The input image, filter, and correlation response are all spectrally expanded in terms of these orthogonal eigenfunctions. An iterative technique between the spatial domain and the spectral domain is used to rephase the invariant modes so that a filter composed of a linear combination of modes has the proper overall invariance. The iterative technique also controls the information content of the filter by maintaining specified amplitudes for each invariant mode in the filter. Targets are detected by spanning the filter to determine points of constant amplitude.

Computer-aided design of VLSI FIR filters

Abstract: A CAD tool is presented for producing very high-throughput FIR filters. Because the CAD tool is application-specific, it is a very high-level tool. An engineer only needs to specify 1) the filter order, N; 2) the input word size; and 3) the output word size. Using this information, the CAD tool generates CIF files for a filter system that can process 10N million samples per second. The purpose of the paper is to illustrate the benefits of applying both bit-level systolic array architecture and application-specific CAD to the problem of FIR filtering. The resulting CAD system reduces the costs of very high-throughput FIR filters with respect to design, fabrication, and operation.

Signal source distortion compensator.

Abstract: A signal processing system in a transmitter/receiver system (Fig. 1) for reducing undesired amplitude, frequency and/or phase distortions arising due to variations from transmitter pulse to transmitter pulse. A plurality (24) of sequential samples of transmitted pulses are averaged (33) and a plurality of filter coefficients (35) are determined therefrom. The coefficients are used in a plurality of filters (36) which respond to a plurality of sequential samples of received pulses, the filters providing output signals (37) in which such distortions are reduced.

Multi-rate digital voice coder apparatus

Abstract: An analog to digital converter for a speech signal is implemented in modules to allow for changes in bit rate and changes in bit stream length according to requirements of the digital transmission system. A pre-emphasis circuit provides an array of pre-emphasized speech samples which are stored in memory. A linear predictive coder provides an array of reflection coefficients and an array of filter coefficients. A pulse processor receives the speech samples and filter coefficients and generates speech amplitude and location signals. These signals are multiplied to generate quantized speech samples. The quantized speech samples and reflection coefficients are provided to a buffer which provides an output signal of a proper bit stream length and bit rate for the digital transmission system.

Tone signal processing device using a digital filter

Abstract: At least two sets of filter coefficients corresponding to different filter characteristics are interpolated by using a control signal for controlling tone color as a parameter of interpolation. Filter coefficients obtained by the interpolation are supplied to a digital filter to determine its filter characteristics and an input tone signal is modified in accordance with the filter characteristics thus determined. Filter characteristics of diverse variation as compared with the number of prepared filter coefficients can thereby be realized. Further, timewise change of filter characteristics can be realized by changing a parameter of interpolation with lapse of time or changing two sets of filter coefficients to be interpolated with lapse of time. Designation of filter coefficients can be made by designating coordinate data of coordinates having at least two axes. In this case, filter coefficients can be changed by changing coordinate data of at least one axis in accordance with tone color control information whereby filter characteristics can be variably controlled.

Spacecraft attitude determination using a second-order nonlinear filter

Abstract: The stringent attitude determination accuracy and faster slew maneuver requirements demanded by present-day spacecraft control systems motivate the development of recursive nonlinear filters for attitude estimation. This paper presents the second-order filter development for the estimation of attitude quaternion using three-axis gyro and star tracker measurement data. Performance comparisons have been made by computer simulation of system models and filter mechanization. It is shown that the second-order filter consistently performs better than the extended Kalman filter when the performance index of the root sum square estimation error of the quaternion vector is compared. The second-order filter identifies the gyro drift rates faster than the extended Kalman filter. The uniqueness of this algorithm is the online generation of the time-varying process and measurement noise covariance matrices, derived as a function or the process and measurement nonlinearity, respectively.

A High-Speed Wave Digital Filter using Carry-Save Arithmetic

Abstract: To test the performance of recursive digital filters for high sampling rate applications, such as pulse former filters or filters for intermediate frequencies of TV sets and radios, a bit-parallel third order lowpass filter has been designed and fabricated in a 2 ?m CMOS technology. This filter has a bireciprocal transfer function and is fully functional at a sampling frequency of 35 MHz. In order to obtain such high frequencies with a recursive filter, a carry-save arithmetic has been used rather than a conventional carry-propagate arithmetic. This gives rise to several realization problems which will be described in this contribution. The prototype chip contains about 9000 transistors and occupies an area of 14.7 mm2.