Abstract: A continuous ADC for generating a digital representation, corresponding to a given quantized voltage level, of an analog input signal and for updating the digital representation at times when the approximate amplitude of the analog signal changes from one quantized level to another. A preferred embodiment includes a set of serially connected output stages with each stage for generating a bit output signal included in the digital representation and an analog residual output signal having an amplitude equal to a linear combination of the amplitude of the input analog signal and the prior bit output signals.

Abstract: A method and apparatus for modeling and synthesizing the response of a network to selected input signals where the input signals may be multi-channel in analog or digital form, and where the network's response may be linear or non-linear and depend on the present input signals as well as its past responses. A relationship is established between the input signals and the network's response to the input signals. Basis vectors are selected capable of representing the relationship. Weighting values for each basis vector are calculated. The calculated weighting values are applied to the corresponding basis vectors to form weighted basis vectors. The synthesis and modeling are then performed by forming a linear combination of the weighted basis vectors. Provision is made for feeding back the linear combination to account for memory and other history-dependent responses of the network.

Abstract: A randomization unit for use in a vibrational control system for generating a series of randomized drive terms where the magnitude of each randomized drive term is the convolution of a psuedo-random sequence of bistable signals and a drive data sequence characterized by a drive power spectral density (PSD). The randomization unit does not required multipliers to perform the convolution operation.

Abstract: A randomization unit for use in a vibrational control system for generating a series of randomized drive terms where the magnitude of each randomized drive term is the convolution of a psuedo-random sequence of bistable signals and a drive data sequence characterized by a drive power spectral density (PSD). The randomization unit does not require multipliers to perform the convolution operation.

Abstract: A network of filter elements for transforming an analog signal of infinite bandwidth into a discrete filter space representation having a bandwidth equal to .pi.. In one embodiment the filter network includes Laguerre filter elements. Other embodiments include methods for estimating the values of the amplitudes, frequencies, and phases of the frequency components of the analog signal, for reconstructing an input signal, and correcting for non-ideal filter elements.

Abstract: A system for determining the s-plane parameters, s.sub.i and r.sub.i, of the transient response, ##EQU1## of a network-under-test (NUT). A filter bank, including N serially connected filter elements, has an input port and N output ports being the output ports of the filter elements. The system includes circuitry for sampling the response signal, at the N filter bank output ports. T seconds after a transient signal is received at the input port. A processor utilizes the values of the sampled response signal to determine the values of s.sub.i and r.sub.i.

Abstract: A method for determining the transfer function of a data acquisition system involves inputting a set of bilevel input test patterns to the system and measuring the system's response. The input test patterns from a complete orthogonal set. The input test patterns or the system's response can be weighted to correct for aliasing error and/or to represent any arbitrary input waveform. From the ratio of the Fourier transform of the sum of the responses to the Fourier transform of the sum of the input test patterns the transfer function is derived. In operation, the transfer function is used to accurately determine the characteristics of any input waveform to the system.

Abstract: A high precision analog to digital converter comprises the combination of an imperfect or low resolution digital to analog converter having an error function known in terms of orthonormal components and an error compensating device capable of generating correction terms which do not interact with one another. The correction terms are based on orthonormal components namely, the Walsh function components, of each signal level to be compensated. At most only one weighting value per bit is required, the combination of which will compensate for errors of any bit combination. In a specific embodiment employing feedback compensation, the output of the low resolution converter and of the compensating device may be summed to produce a high performance, high precision converter with increased accuracy and resolution.

Abstract: A high precision digital to analog converter comprises the combination of an imperfect or low resolution digital to analog converter having an error function known in terms of orthonormal components and an error compensating device capable of generating correction terms which do not interact with one another. The correction terms are based on orthonormal components namely, the Walsh function components, of each signal level to be compensated. At most only one weighting value per bit is required, the combination of which will compenate for errors of any bit combination. In a specific embodiment employing feedforward compensation, the output of the low resolution converter and of the compensating device may be summed to produce a high performance, high precision converter with increased accuracy and resolution.

Abstract: A digital test system for generating a test signal in the form of an amplitude and/or phase modulated sinusoidal signal at a given carrier frequency is disclosed. Arbitrary amplitude and phase modulation functions may be selected and are provided to the system in the form of digitally encoded data streams. The system does not require a digital multiplier. A signal in the form of an unmodulated carrier signal may also be generated.

Abstract: A method of calibrating a feed forward, DAC post-correction system. The method includes fitting the Walsh co-efficiencts of a DAC transfer function to a straight line in the log domain. The deviation of these terms in the log domain is utilized to compute Walsh correction terms for use in the post correction system.

Abstract: A method for statistically calibrating an analog-to-digital converter with an electronic test system. A digital-to-analog converter which has been calibrated by premeasured weighting coefficients with respect to two-state orthogonal signals is excited with two state signals at each input bit which together represent a single signal with uniform amplitude probability with respect to time, and wherein each excitation signal is orthogonal with respect to all other excitation signals. The output of the digital-to-analog converter is detected by the analog-to-digital converter under test. The digital time domain output signals are then mapped into a transform domain to obtain weighting coefficients of each bit of the output response. Finally the transform domain weighting coefficients are weighted by the reciprocal of the premeasured weighting coefficients to obtain the unbiased weight of each bit of the analog-to-digital converter under test.

Abstract: A method for statistically calibrating an analog-to-digital converter with an electronic test system. A digital-to-analog converter which has been calibrated by premeasured weighting coefficients with respect to two-state orthogonal signals is excited with two state signals at each input bit which together represent a single signal with uniform amplitude probability with respect to time, and wherein each excitation signal is orthogonal with respect to all other excitation signals. The output of the digital-to-analog converter is detected by the analog-to-digital converter under test. The digital time domain output signals are then mapped into a transform domain to obtain weighting coefficients of each bit of the output response. Finally the transform domain weighting coefficients are weighted by the reciprocal of the premeasured weighting coefficients to obtain the unbiased weight of each bit of the analog-to-digital converter under test.

Abstract: A method and apparatus is described for dynamically testing the overall performance characteristics of digital-to-analog converts and analog-to digital converters which involve excitation of the converters by an orthogonal function signal. Specifically the method comprises dynamically exercising a converter with an analog or digital signal pattern characterized by the sum of a set of mutually orthogonal functions, the sum having substantially uniform amplitude distribution among allowable states (maximum entropy), and simultaneously examining the output response of the converter for a plurality of basic performance parameters. The basic performance parameters typically include distortion, linearity and optimum gain. The simultaneous examination involves sorting out expected responses to simultaneously applied orthogonal signals. The method yields a relatively complete statistical description of the performance characteristics. The preferred excitation is based on the Walsh functions.

Abstract: A method and apparatus for exciting the analog input of an analog-to-digital converter and examining the output by means of analysis with histograms to determine the amplitude probability distribution of each digital output value. The amplitude probability density functions corresponding to accurately known excitation signals may then be processed to obtain a description of the transfer characteristic for the analog-to-digital converter to be calibrated. In a particular embodiment, the excitation signals applied to the analog input of the analog-to-digital converter are pairs of signals representing accurately known rising and decaying exponential signals, and particularly exponential signals which are complementary to one another. Histograms of the digital response to the analog excitation signals are then constructed which correspond to amplitude density functions.

Abstract: A method for statistically calibrating a digital-to-analog converter with an electronic test system. The digital-to-analog converter is excited with two state signals at each input bit which together represent a single signal with uniform amplitude probability with respect to time, and wherein each excitation signal is orthogonal with respect to all other excitation signals. The output of the digital-to-analogconverter is detected by an analog-to-digital converter which has been calibrated by premeasured weighting coefficients with respect to two-state orthogonal signals. The digital time domain output signals are then mapped into a transform domain to obtain weighting coefficients of each bit of the output response. Finally the transform domain weighting coefficients are weighted by the reciprocal of the premeasured weighting coefficients to obtain the unbiased weight of each bit of the digital-to-analog converter under test.

Abstract: A system for determining the modal characteristics of a structure in which a plurality of driving signals which have a common preselected frequency and a preselected set of amplitudes and phases corresponding to the estimated parameters of a mode of interest are applied to the structure. The common preselected frequency is in general a complex frequency. The response of the structure to the driving signals are then sensed. Means are provided for receiving the driving signals and the sensed responses, for generating a set of transfer functions characteristic of the structure corresponding to the mode of interest and for obtaining from such set of transfer functions improved estimates of the complex frequencies and complex residues of the mode of interest, that is the damping, frequency, magnitudes and phases of the mode of interest.

Abstract: A system for determining the location of poles in the complex s-plane and their residues from empirical data is disclosed. The system utilizes the impulse response or similar data of a specimen and through repeated differentiations of a frequency domain representation of this response locates the frequency of the poles. The results of any two consecutive differentiations are used to determine the location of the poles in the s-plane and their residues.

Abstract: The Laplace transform (s-plane) is obtained for contiguous or overlapping frames of speech (or other signals) and polepair parameters (frequency, damping, magnitude and phase) are selected for transmission so as to preserve maximum energy. Speech is reconstructed from the transmitted parameters, using, for example, a damped sine wave as the equivalent of a pole pair. No separate pitch determination is made, nor is a voiced/unvoiced decision required.

Abstract: A system for determining the transfer function of a specimen from an input or driving signal and an output or response signal. The driving signal and response signal are each divided into segments or frames of data, each frame of equal duration and a sum is formed for each frame of data. A time-domain to frequency-domain transform is taken of the two sums; the transfer function is then determined from the transformed data.