Showing papers on "Linear phase published in 1979"

Invertibility of a room impulse response

Abstract: When a conversation takes place inside a room, the acoustic speech signal is distorted by wall reflections. The room’s effect on this signal can be characterized by a room impulse response. If the impulse response happens to be minimum phase, it can easily be inverted. Synthetic room impulse responses were generated using a point image method to solve for wall reflections. A Nyquist plot was used to determine whether a given impulse response was minimum phase. Certain synthetic room impulse responses were found to be minimum phase when the initial delay was removed. A minimum phase inverse filter was successfully used to remove the effect of a room impulse response on a speech signal.

Design of stable two‐dimensional analogue and digital filters with applications in image processing

Abstract: A method is proposed for the frequency domain design of linear two-dimensional analogue and digital filters with guaranteed stability. The technique used is based on the result that the numerator and the denominator of the input immittance of a two-variable network (which is passive and lossy) are strictly Hurwitz polynomials. One of these strictly Hurwitz polynomials is assigned to the denominator of a two-variable analogue transfer function and the network elements are then used as the variables of optimization thereby guaranteeing the stability of the analogue transfer function. The transfer function of the corresponding two-dimensional discrete (digital) filter is obtained from the analogue transfer function by the bilinear transformation. Examples illustrating the versatility of the technique in designing 2D digital filters of arbitrary order approximating a given magnitude and group delay response are presented. These filters are used to process a simple binary image. The results obtained demonstrate the importance of linear phase in image processing applications. The method presented here can be extended to the design of stable m-dimensional analogue and digital filters.

Frequency transformations for linear-phase variable-cutoff digital filters

Abstract: The first- and second-order forms of a transformation for implementing a variable-cutoff linear-phase FIR filter are investigated. In the first-order case, correct constraints have been placed on the variable parameter A_0 ; it is pointed out that the cutoff frequency increases, while the cutoff slope decreases, monotonically with increasing A_v; A_0 = 0 corresponds to the prototype filter. The second-order transformation is shown to overcome these drawbacks without increasing the computational load or the number of variable parameters.

Video signal time base error correction circuit

Abstract: A system for compensating and correcting errors in the time base of video signals read from a video disc in which the errors are corrected first coarsely in response to phase differences between horizontal sync signals of the video signals following which a fine correction is made in response to measured phase differences of burst signals of the video signal. A first phase detector produces an output signal in response to phase differences between the burst signals and an output of a reference frequency generator operating at the burst frequency. A second phase detector produces an output signal in response to phase differences between the horizontal signals and the reference signal divided down by a predetermined factor. The output signals of the two phase detectors are summed and then applied to the control input of a variable delay line which adjusts the delay time and hence phase errors in the video signal accordingly.

Characterization and spectral equalization for hight-density disk recording

Abstract: For a 3350 type head/disk interface, the amplitude spectrum of an experimental characteristic pulse deviates slightly from the idealized Lorentzian distribution in the signal bandwidth; but the phase spectrum of the experimental pulse is quite distorted from the desired linear phase, causing the widely observed asymmetrical time-domain waveform. A quantitative characterization and comparison are presented and individual transfer functions for ideal linear equalization are proposed. To remove the intersymbol interference, i.e., the linear portion of the distortion, from the playback signal of a high-density system, several cosine-power-law spectral candidates, with minimized noise enhancement and the capability of restoring amplitude loss and timing error, are explored. A unified relationship for the entire family of cosine-power-law spectral shapes and their bandwidth tradeoffs are identified, allowing for optimum read equalization at a tolerable signal-to-noise penalty. This is a useful tool for exploring practical density limitations of any given head/disk interface.

Linear phase variable digital bandpass filters

Abstract: A transformation for obtaining a linear phase variable cut-off low-pass digital filter has been recently described in the literature [1]. In this letter, this transformation is worked out for the case of linear phase bandpass digital filters with variable bandwidth and center frequency.

Delay line signal equalizer for magnetic recording signal detection circuits

Abstract: A signal amplitude equalizer circuit, having linear phase, for reforming high density magnetically recorded read-back signals to remove peak shift and the resulting errors caused thereby. The equalizer circuit employs series balanced differential amplifier circuits in connection with properly terminated delay lines across the amplifier output terminals to provide an adjustable transfer function that transforms the readhead signal, that follows an exponential frequency spectra curve, into a cosine power frequency spectra curve which, in the time domain, possesses a sufficiently narrow base to eliminate peak shift and its resulting amplitude decay.

A novel approach for implementing pitch prediction in sub-band coding

Abstract: In this paper a novel method is proposed for implementing pitch prediction inside of the sub-bands of a sub-band coder. The method uses digital linear phase shifters based on a bandpass interpolation scheme to achieve the non-integer delays necessary in the feedback loops of the pitch predictors.

A frequency-independent phase shifter

Abstract: The circuit described gives the frequency-independent phase shift for a band-limited signal, while keeping its amplitude invariant. The key component is a field-effect transistor which, driven by a frequency discriminator, provides a resistance which is inversely proportional to the frequency. A prototype circuit produces a phase shift flat to within +or-2 degrees over the frequency range from 20 to 40 kHz.

Even-Degree Low-Pass Prototype Networks for the Design of Microwave Linear-Phase Filters

Abstract: AIbstrszct-The element vafues of a selection of even-degree nooroirrimum-phase low-pass prototype networks with eqrriripple passbmrd arnplitrrde and constant group delay in the least squares sense over a large percentage of the passband are tabulated. At] the prototypes have passband insertion loss ripple R =0.01 dB and cutoff frequency tiC = 1.0 rad/s at the O.01-dB puint. Five tables contain the element values of networks up to degree N =20. The tables are classified according to tbe snssnber of transmissinrs zeros at infinite frequency NZm and the passband frequency to which the group delay is constant in ttse least squares seose Wd. The following combinations of NZm arsd ad are tabrstatak NZW = 2 and c,rd==0.9; ,VZm =4 and tid =0.8; NZO = 6 and od=0.7; NZm =8 and ~J = 0.6; i~d NZ ~ =10 and Wd= 0.5, Tbe maximum phase and delay errors for each network are tabulated. Plots of the psmsband group delay and stopbausd insertion lass versus frcqnency, for each networkj accompany the tables to facilitate selection of a prototype. The prototypes are suitable for the design of narrow-baud generafii interdigital, generafk?ed direetcorrpled cavity wavegrside, and generalii combline linear-phase filters A simple algmitfrm for the anafysis of the prototypes is given.

Tables for Nonminimum-Phase Even-Degree Low-Pass Prototype Networks for the Design of Microwave Linear-Phase Filters

Abstract: The element values of a selection of even-degree nonminimum phase low-pass prototype networks with equiripple passband amplitude and constant group delay in the least squares sense over a large percentage of the passband are tabulated. All the prototypes have passband insertion loss ripple R=0.01 dB and cutoff frequency omega/sub c/ = 1.0 rad/s at the 0.01-dB point. Five tables contain the element values of networks up to degree N=20. The tables are classified according to the number of transmission zeros at infinite frequency NZ/sub infin/ and the passband frequency to which the group delay is constant in ttse least squares sense omega/sub d/. The following combinations of NZ/sub infin/ and omega/sub d/ are tabulated: NZ/sub infin/ = 2 and omega/sub d/=0.9; NZ/sub infin/ =4 and omega/sub d/ =0.8; NZ/sub infin/ = 6 and omega/sub d/=0.7; NZ/sub infin/ =8 and omega/sub d/ = 0.6; and NZ/sub infin/ =10 and omega/sub d/= 0.5. The maximum phase and delay errors for each network are tabulated. Plots of the passband group delay and stopband insertion loss versus frequency, for each network, accompany the tables to facilitate selection of a prototype. The prototypes are suitable for the design of narrow-band generalized interdigital, generalized direct-coupled cavity waveguide, and generalized combline linear-phase filters A simple algorithm for the analysis of the prototypes is given.

Frequency Content in Earthquake Simulation

Abstract: Random processes with frequency content similar to that of recorded earthquakes are generated by using linear filters to remove the very low and very high frequency components of white noise The mean-squared responses of single-degree-of-freedom structures to these filtered processes are compared with the corresponding responses to white noise excitation For linear structures exact analytical solutions for mean-squared displacement and velocity response are presented Using an equivalent white noise to approximate the nonwhite noise excitation generally extends the frequency range of adequate approximation, particularly for small structural damping For yielding structures Monte Carlo simulation is used to obtain response levels for some representative situations The effect of a second-order filter on yielding system response is somewhat different than for a linear system; in fact, it is more like the effect of a first-order filter on linear system response

Phase synchronism circuit

Abstract: PURPOSE:To reduce the synchronism pull-in period at re-input of input signal, by performing phase comparison between the output of the voltage controlled oscillator and the input signal, and controlling the oscillation frequency of the voltage controlled oscillator with the low frequency component of the phase error signal. CONSTITUTION:In the phase synchronism circuit which can reduce the synchronism pull-m in period, the input signal from the terminal 1 is fed to the discrimination circuit 2 and the selection circuit 3. This circuit 3 feeds input signal to the phase comparator 4 if the input signal is present and the output of the phase shifter 6 is fed to the phase comparator 4 if absent. Phase comparison is made with the output of the voltage controlled oscillator 7, and harmonics are rejected at LPF 5 from the phase error signal and the result is fed to the oscillator 7 to control the oscillation frequency. Thus, even with long time interruption of input signal, the frequency variation of the oscillator 7 is made very less to reduce the synchronism pull-m in period when the input signal is re-inputted.

Magnetostatic Wave Reflective Array Filter

Abstract: Results from experimental and theoretical studies, carried out in the 3 GHz region in epitaxial YIG, on tunable magnetostatic forward volume wave (MSFVW) oblique Incidence reflective array filters are reported. Uniform reflecting gratings consisting of either metal bars or metal dot arrays have been studied as MSFVW reflectors and yield an octave tunable bandpass filter. Width weighting of the bar widths has yielded significant reduction in side lobe levels of these bandpsss filters. Initial experiments on double reflection from a pair of 10 element metal bar reflecting arrays has been carried out. A graded 14 element metal bar reflecting array has yielded a constant 100 nsec delay 400 MHz bandwidth linear phase filter with a phase ripple of /spl plusmn/15/spl deg/ from linear. This is the first reported controlled dispersion MSFVW filter. Initial studies of quadratic phase MSFVW filters are reported.

Phase-lock control considerations for coherently combined lasers

Abstract: Fundamental performance limitations of a phase-lock control loop used to coherently combine the output of two lasers are presented. The phase-lock loop is designed to lock the differential phase (frequency and phase) between the two lasers to a specified reference phase. An optical heterodyne configuration is used to determine the differential phase of the laser pair, which in turn is compared with the reference phase to create an error voltage. The error voltage is filtered and used to frequency modulate one of the lasers in an attempt to null the error. An integro-differential loop equation, valid for the linear operating range, is derived in terms of the reference phase, the heterodyne measurement noise, and the various laser phase instabilities. The solution of the equation results in an expression for the phase error variance in terms of the closed-loop noise equivalent bandwidth W(H). An expression for the value of W(H) which minimizes the phase error variance is developed. In addition to the noise effects, the steady-state and dynamic performance of the loop is examined for different loop filters and modulation formats. A design example pairing a CO(2) waveguide and conventional laser is presented. Implications for coherent laser arrays are discussed.

A Fast Switching, Low Loss, Low Drive, 12 GHz Pin Phase Shifter

Abstract: A 4-bit PIN microstrip phase shifter switching in 1 ns/bit with only 9 mW/bit of driving power is described. The phase shifter is combined with its TTL-compatible driver in a single compact package. RF insertion loss is 1.6 dB /spl plusmn/ 0.2 dB for the 16 phase states over the 11.7 - 12.2 GHz band. Forty-two phase shifter modules have been built. A linear phase array employing these modules has been constructed.

Phase control device which compensates for input variations

Abstract: A phase control device for use in a phase control circuit, which forms a comparison signal synchronous with an a.c. power source, then compares this comparison signal directly with a preset phase control angle signal, and controls a phase control element provided in the a.c. phase control circuit by the compared output.

Optimum weights in delay equalization

Abstract: In network design practice it is quite typical to design a network having a suitable amplitude response and to follow this with a network having unit amplitude response whose purpose is to modify the phase response so that the total network meets requirements on the phase response. Such phase-correcting networks are called all-pass delay equalizers. Their design amounts to function approximation of a special form. It is common practice to design equalizers so that the residual error in the envelope delay (d\phi/d\omega) has an equal ripple characteristic (constant weight function). The resulting residual distortion has a form which can be interpreted in a television system as two "ghosts" displaced in time from the desired signal by an amount inversely proportional to the ripple frequency. We determine a weight function, dependent upon the spectrum of the signal being processed, which minmmizes the power in these "ghost" distortions. These results are illustrated by application to example networks and signals.

Variable-centre-frequency constant-bandwidth linear-phase digital bandpass filters

Abstract: A third-order cosine transformation is used to obtain a variable-centre-frequency constant-bandwidth linear-phase digital bandpass filter, the bandwidth being the same as that of the prototype bandpass filter.

Microstrip delay line compensated for thermal phase variations

Abstract: A delay line for microwaves, whose transfer function is subject to phase variations with changes in ambient temperature, is connected in cascade with a transmission line whose phase varies in the opposite sense. The transmission line, which may be integrated with the associated delay line in a common microstrip structure, is of such length as to maintain the overall phase shift substantially constant within a certain temperature range.

Detection of sinusoids by linear prediction filter frequency response

Abstract: Four processors for the detection of a sinusoid of known frequency but unknown phase in Gaussian white noise of known spectrum level are described: the likelihood ratio detector; a processor employing incoherently averaged discrete Fourier transforms (DFTs); and two systems utilizing the frequency response function of an adaptive linear prediction filter. Power curves for these receivers are compared, showing that the adaptive detectors are similar in performance to the incoherent DFT processor. An expression for the optimal adaptive filter feedback constant is derived.

Linear phase approximation for two-dimensional recursive digital filters

Abstract: A new technique is presented for the design of two-dimensional IIR digital filters with symmetrical amplitude response and linear phase specifications. The design method is described and an example illustrating the use of the technique is also included.

An adaptive nonrecursive linear phase MTI filter based on nonuniform spectral resolution of input

Abstract: This correspondence describes a method of adaptively filtering signals with time-varying spectra. In this method the short term spectrum of the signal is estimated by using a modified periodogram scheme. Also, the spectral resolution is made unequal so as to get more spectral components in the frequency region of interest. This spectrum estimate is used in designing an MTI filter which is desired to give good clutter rejection.

Pseudo-Elliptical Phase-Equalized Filter

Abstract: The paper considers a filter with double transmission zeros at both real and imaginary frequencies. The new approach allows satisfactory phase equalization for the stated out-of-band attenuation without increasing the filter degree as would be required for a linear phase filter.

An alternative 90° phase-shifting technique

Abstract: A circuit is proposed that, assures constant 90° phase shifting (and, eventually, any value between 0° and 180°) of a sinusoidal signal over a 1 to 100 frequency range. A brood input amplitude rang...

Characterization and spectral equalization for high-density disk recording

Abstract: For a 3350 type head/disk interface, the amplitude spectrum of an experimental characteristic pulse deviates slightly from the idealized Lorentzian distribution in the signal bandwidth; but the phase spectrum of the experimental pulse is quite distorted from the desired linear phase, causing the widely observed asymmetrical time-domain waveform. A quantitative characterization and comparison are presented and individual transfer functions for ideal linear equalization are proposed. To remove the intersymbol interference, i.e., the linear portion of the distortion, from the playback signal of a high-density system, several cosine-powerlaw spectral candidates, with minimized noise enhancement and the capability of restoring amplitude loss and timing error, are explored. A unified relationship for and their bandwidth tradeoffs are identified, allowing the entire family of cosine-power-law spectral shapes for optimum read equalization at a tolerable signal-tonoise penalty. This is a useful tool for exploring practical density limitations of any given head/disk interface.

A method for the design of phase equalizers

Abstract: A new method for the design of all-pass filters with arbitrary group delay characteristics is presented. Given the desired group delay, the magnitude of the corresponding minimum phase filter is found. The problem is then reduced to that of approximating a magnitude response by a stable all-pole filter or by a Finite-duration Impulse Response (FIR) filter with all its zeros outside the unit circle. It is shown with several examples that the resulting errors can present an almost equal ripple behavior. Comparisons with optimum L p designs are made.

Selective, constant delay wave digital filters

Abstract: A wave digital filter design method is discussed which produces good selectivity and flat delay characteristics. The filter is designed in the microwave domain with monotonic stopband response using Carlin and Wu's technique for minimum-phase linear phase commensurate line transducers, then realized in the digital domain with Fettweis relations. A comparison is made of frequency response sensitivity to multiplier coefficient truncation between this wave filter and some FIR's, all with approximately the same number of multipliers. The cascade realizations of the FIR's show considerable passband amplitude response deterioration due to coefficient truncation, whereas the direct realizations for the FIR's show significant deterioration of stopband amplitude response owing to coefficient truncation. Coefficient truncation for the wave digital filter caused relatively small deterioration of both passband and stopband amplitude responses as well as small distortion of the flat delay characteristic in the passband.