Patents by Inventor Kenneth D. Poulton

Kenneth D. Poulton has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • Patent number: 11329640
    Abstract: An analog delay line includes a clock generator, an analog sampling circuit, a bank of analog memory cells, a memory controller, an analog readout circuit, and an analog multiplexer. The clock generator is configured to output plural reception clock signals of different frequencies and plural transmission clock signals of different frequencies, the transmission clock signals offset in accumulated phase relative to the reception clock signals. The analog sampling circuit is controlled by at least one of the reception clock signals, and is configured to output a sequence of sampled voltages of an analog input signal. The memory controller is configured to control a write operation at a write frequency of at least one of the reception clock signals and a read operation at a read frequency of at least one of the transmission clock signals.
    Type: Grant
    Filed: February 18, 2021
    Date of Patent: May 10, 2022
    Assignee: Keysight Technologies, Inc.
    Inventors: Charles Wu, Ken A. Nishimura, Kenneth D. Poulton
  • Patent number: 10069505
    Abstract: A circuit for digital-to-analog conversion includes a first digital-to-analog converter (DAC), a second DAC, and an output node. The first DAC provides charges from multiple first charge sources segmented into a first group for most significant bits of a digital input to the first DAC and a second group for least significant bits of the digital input. Dither is both added to the digital input to the first DAC and used as sole digital input to the second DAC. Analog output from the second DAC is subtracted from analog output of the first DAC at the output node so as to cancel the dither added to the first DAC.
    Type: Grant
    Filed: September 13, 2017
    Date of Patent: September 4, 2018
    Assignee: Keysight Technologies, Inc.
    Inventors: Kenneth D. Poulton, Robert Edward Jewett
  • Patent number: 7492302
    Abstract: In an analog-to-digital converter, an analog-to-digital conversion stage comprising a comparator and an analog residual signal generator. The comparator is operable to compare an analog input signal or a sample of the analog input signal with a threshold to generate a bit signal. The analog residual signal generator is operable to generate an analog residual signal from signals comprising the sample of the analog input signal and the bit signal such that, at a level of the analog input signal equal to the threshold of the comparator, the analog residual signal has a level independent of the state of the bit signal. The analog residual signal generator comprises a summing element, a selective inverter and an amplifier in series. The summing element is operable to sum a signal input to it with a reference signal.
    Type: Grant
    Filed: April 30, 2007
    Date of Patent: February 17, 2009
    Assignee: Agilent Technologies, Inc.
    Inventor: Kenneth D. Poulton
  • Patent number: 7450044
    Abstract: A digital-to-analog conversion system comprises a digital input, a digital-to-analog converter and a modified digital signal generator. The digital-to-analog converter has a conversion frequency and is subject to a periodic error having a periodicity equal to that of an N-th sub-harmonic of the conversion frequency, where N is an integer. The digital input is operable to receive a digital input signal. The modified digital signal generator is interposed between the digital input and the digital-to-analog converter and is operable in response to the digital input signal to generate a modified digital signal. The modified digital signal comprises a dynamic digital mitigation component that mitigates the periodic error of the digital-to-analog converter.
    Type: Grant
    Filed: March 27, 2007
    Date of Patent: November 11, 2008
    Assignee: Agilent Technologies, Inc.
    Inventors: George S. Moore, Nico Lugil, Kenneth D. Poulton
  • Publication number: 20080266163
    Abstract: In an analog-to-digital converter, an analog-to-digital conversion stage comprising a comparator and an analog residual signal generator. The comparator is operable to compare an analog input signal or a sample of the analog input signal with a threshold to generate a bit signal. The analog residual signal generator is operable to generate an analog residual signal from signals comprising the sample of the analog input signal and the bit signal. The analog residual signal generator comprises a summing element, a selective inverter and an amplifier in series. The summing element is operable to sum a signal input to it with a reference signal. The selective inverter precedes the summing element, and is operable in response to a first state of the bit signal to pass a signal input to it, and is operable in response to a second state of the bit signal to invert the signal input to it.
    Type: Application
    Filed: April 30, 2007
    Publication date: October 30, 2008
    Inventor: Kenneth D. Poulton
  • Publication number: 20080266157
    Abstract: Staggered interleaved Nyquist regions associated with differing ADC clock rates (FCLK) avoids spectrum lost through disjoint guard bands at the end of or between adjacent Nyquist regions. The staggered interleaved Nyquist regions overlap by an amount at least as much as is consumed by the guard bands. Selectable anti-aliasing filters associated with each Nyquist region and its ADC clock rate are used to enforce the staggered Nyquist regions and their various guard bands. For example, and neglecting guard bands, an initial raw band of operation RB1 may be the First Nyquist region for a basic sampling frequency Fs. An adjacent raw band of operation RB2 that overlaps RB1 may be the Second Nyquist region for an alternate sampling frequency 2Fs/3. An adjacent raw band of operation RB3 that overlaps RB2 may be the Second Nyquist region for the basic sampling frequency Fs. These raw bands interleave and overlap: RB1: DC to Fs/2 1st Nyq. for FCLK = Fs RB2: (?)Fs ? (½)(?Fs) = Fs/3 to 2nd Nyq.
    Type: Application
    Filed: April 26, 2007
    Publication date: October 30, 2008
    Applicant: AGILENT TECHNOLOGIES, INC.
    Inventors: Joseph M. Gorin, Kenneth D. Poulton
  • Patent number: 7439897
    Abstract: Staggered interleaved Nyquist regions associated with differing ADC clock rates (FCLK) avoids spectrum lost through disjoint guard bands at the end of or between adjacent Nyquist regions. The staggered interleaved Nyquist regions overlap by an amount at least as much as is consumed by the guard bands. Selectable anti-aliasing filters associated with each Nyquist region and its ADC clock rate are used to enforce the staggered Nyquist regions and their various guard bands. For example, and neglecting guard bands, an initial raw band of operation RB1 may be the First Nyquist region for a basic sampling frequency Fs. An adjacent raw band of operation RB2 that overlaps RB1 may be the Second Nyquist region for an alternate sampling frequency 2Fs/3. An adjacent raw band of operation RB3 that overlaps RB2 may be the Second Nyquist region for the basic sampling frequency Fs. These raw bands interleave and overlap: RB1: DC to Fs/2 1st Nyq. for FCLK = Fs RB2: (2/3)Fs ? to 2(2Fs/3)/2 = 2nd Nyq.
    Type: Grant
    Filed: April 26, 2007
    Date of Patent: October 21, 2008
    Assignee: Agilent Technologies, Inc.
    Inventors: Joseph M. Gorin, Kenneth D. Poulton
  • Publication number: 20080238747
    Abstract: A digital-to-analog conversion system comprises a digital input, a digital-to-analog converter and a modified digital signal generator. The digital-to-analog converter has a conversion frequency and is subject to a periodic error having a periodicity equal to that of an N-th sub-harmonic of the conversion frequency, where N is an integer. The digital input is operable to receive a digital input signal. The modified digital signal generator is interposed between the digital input and the digital-to-analog converter and is operable in response to the digital input signal to generate a modified digital signal. The modified digital signal comprises a dynamic digital mitigation component that mitigates the periodic error of the digital-to-analog converter.
    Type: Application
    Filed: March 27, 2007
    Publication date: October 2, 2008
    Inventors: George S. Moore, Nico Lugil, Kenneth D. Poulton
  • Patent number: 7375667
    Abstract: Staggered consecutive Nyquist regions associated with differing DAC synthesizer clock rates (FCLK) avoids spectrum lost through disjoint guard bands at the end of or between adjacent Nyquist regions. The staggered consecutive Nyquist regions overlap by an amount at least as much as is consumed by the guard bands. Selectable reconstruction filters associated with each Nyquist region and its DAC clock rate are used to enforce the staggered Nyquist regions and their various guard bands.
    Type: Grant
    Filed: September 19, 2006
    Date of Patent: May 20, 2008
    Assignee: Agilent Technologies, Inc.
    Inventors: Kenneth D. Poulton, Stephen T Sparks
  • Publication number: 20080068244
    Abstract: Staggered consecutive Nyquist regions associated with differing DAC synthesizer clock rates (FCLK) avoids spectrum lost through disjoint guard bands at the end of or between adjacent Nyquist regions. The staggered consecutive Nyquist regions overlap by an amount at least as much as is consumed by the guard bands. Selectable reconstruction filters associated with each Nyquist region and its DAC clock rate are used to enforce the staggered Nyquist regions and their various guard bands. For example, and neglecting guard bands, an initial raw band of operation RB1 may be the First Nyquist region for a basic sampling frequency Fs. An adjacent raw band of operation RB2 that overlaps RB1 may be the Second Nyquist region for an alternate sampling frequency 2Fs/3. An adjacent raw band of operation RB3 that overlaps RB2 may be the Second Nyquist region for the basic sampling frequency Fs. These raw bands overlap: In this example the smallest overlap is Fs/6.
    Type: Application
    Filed: September 19, 2006
    Publication date: March 20, 2008
    Inventors: Kenneth D. Poulton, Stephen T. Sparks
  • Publication number: 20080043991
    Abstract: An instrument, analyzer or item of test equipment for some technical discipline is equipped with a numeric keypad that, in addition to the digits zero through nine and the decimal point, and which instead of carrying all the keys ordinarily needed, has been augmented to carry additional sequences of ordered symbols or legends. The multi-tap and T9Word techniques may be used to enter text strings that may include technical legends and abbreviations. In a system where another row or column of keys has been added, a CMD (‘COMMAND’) key can institute a Symbol Mode where selected keys, including the digit keys exhibit multi-tap or T9Word operation, whereas they ordinarily do not, so as to be able to support rapid numeric entry of arbitrary sequences of digits, including consecutively repeated digits. The Symbol Mode may be automatically entered and exited during the execution of commands or operations that need Symbol Mode input.
    Type: Application
    Filed: July 28, 2006
    Publication date: February 21, 2008
    Inventor: Kenneth D. Poulton
  • Patent number: 7206907
    Abstract: The present method reduces variations in noise and temperature in a mixed-signal circuit including memory. Memory electrically proximate an analog circuit is provided and a digital data word received at the memory. When the data word is not a desired data word, a dummy write to the memory is performed. When the data word is a desired data word, the data word is written to the memory. The mixed-signal circuit includes an analog circuit, memory electrically proximate to the analog circuit and connected to receive digital data words, and a memory controller connected to the memory. The memory controller is operable to cause the memory to write to the memory each of the data words that is a desired data word and additionally to perform a dummy write to memory for each of the data words that is not a desired data word.
    Type: Grant
    Filed: December 31, 2003
    Date of Patent: April 17, 2007
    Assignee: Agilent Technologies, Inc
    Inventors: Kenneth D. Poulton, Thomas E. Kopley
  • Patent number: 7148828
    Abstract: A method for calibrating time interleaved samplers comprising applying a calibration signal to a time-interleaved sampling device, wherein the signal is coherent with at least one sample clock on the device and is periodic and has a predetermined spectral content and frequency, sampling, by said time-interleaved sampling device, the calibration signal at a plurality of phases to form samples, averaging the formed samples, and calculating the phase error of each sample based on the average calibration signal sample.
    Type: Grant
    Filed: May 3, 2005
    Date of Patent: December 12, 2006
    Assignee: Agilent Technologies, Inc.
    Inventors: Andrew D. Fernandez, Vamsi K. Srikantam, Robert M. R. Neff, Kenneth D. Poulton
  • Patent number: 6909310
    Abstract: A line driver fabricated from CMOS devices that provides a substantially constant output impedance over a significant range of a time-varying input voltage includes a time-varying current source, a pair of CMOS output loads, and a pair of biasing circuits. Each CMOS output load includes a NMOS transistor and a PMOS transistor connected in parallel and each biased into a linear range of operation. In response to a time-varying input voltage, the time-varying current source draws current from the pair of CMOS output loads in a manner that operates each CMOS output load to collectively establish a time-varying output voltage component at an associated output terminal.
    Type: Grant
    Filed: January 30, 2003
    Date of Patent: June 21, 2005
    Assignee: Agilent Technologies, Inc.
    Inventors: Kenneth D. Poulton, Robert M. R. Neff, Jorge A. Pernillo, Mehrdad Heshami
  • Publication number: 20040150432
    Abstract: A line driver fabricated from CMOS devices that provides a substantially constant output impedance over a significant range of a time-varying input voltage includes a time-varying current source, a pair of CMOS output loads, and a pair of biasing circuits. Each CMOS output load includes a NMOS transistor and a PMOS transistor connected in parallel and each biased into a linear range of operation. In response to a time-varying input voltage, the time-varying current source draws current from the pair of CMOS output loads in a manner that operates each CMOS output load to collectively establish a time-varying output voltage component at an associated output terminal.
    Type: Application
    Filed: January 30, 2003
    Publication date: August 5, 2004
    Inventors: Kenneth D. Poulton, Robert M. R. Neff, Jorge A. Pernillo, Mehrdad Heshami
  • Patent number: 6720895
    Abstract: A method of calibrating a high-speed analog to digital converter and an ADC that implements the method. Multiple linear regression analysis is used to calibrate the stages of a pipeline ADC to compensate for variations in gain from stage to stage and optionally to compensate for harmonic distortion. Current amplifiers each having gain of about 1.6 are used for low power consumption, minimal surface area requirements, and rapid sampling speed. Weighting factors are stored in lookup tables to minimize the number of adders required to generate the output digital word.
    Type: Grant
    Filed: February 1, 2002
    Date of Patent: April 13, 2004
    Assignee: Agilent Technologies, Inc.
    Inventors: Kenneth D. Poulton, Robert M. R. Neff, Matthew S. Holcomb, James Kang
  • Patent number: 6707411
    Abstract: The analog-to-digital conversion system comprises an analog-to-digital converter that includes a digital output, memory having a data input and a data output, an output port, an input data bus that extends from the digital output of the analog-to-digital converter to the data input of the memory and an output data bus that extends from the data output of the memory to the output port. The analog-to-digital converter is structured to generate digital samples at a sampling rate. The input data bus is structured to operate at the sampling rate of the ADC. At least one of the data output of the memory, the output data bus and the output port is structured to operate at a maximum rate less than the sampling rate.
    Type: Grant
    Filed: October 30, 2002
    Date of Patent: March 16, 2004
    Assignee: Agilent Technologies, Inc.
    Inventors: Kenneth D. Poulton, Thomas E. Kopley, Robert M. R. Neff
  • Publication number: 20030146861
    Abstract: A method of calibrating a high-speed analog to digital converter and an ADC that implements the method. Multiple linear regression analysis is used to calibrate the stages of a pipeline ADC to compensate for variations in gain from stage to stage and optionally to compensate for harmonic distortion. Current amplifiers each having gain of about 1.6 are used for low power consumption, minimal surface area requirements, and rapid sampling speed. Weighting factors are stored in lookup tables to minimize the number of adders required to generate the output digital word.
    Type: Application
    Filed: February 1, 2002
    Publication date: August 7, 2003
    Inventors: Kenneth D. Poulton, Robert M. R. Neff, Matthew S. Holcomb, James Kang
  • Patent number: 5155388
    Abstract: Apparatus for introduction of a controllable time delay in the transition of a logic device output signal from a first logical level to a second logical level, in response to change of an input signal, or the difference of two input signals from a third logical level to a fourth logical level, delivered to the logic device. In a first embodiment, a limiting amplifier receives an input signal S.sub.in and a threshold signal V.sub.T at its two terminals and forms an amplified difference output signal S.sub.out. A selected voltage level signal V.sub.L and S.sub.out are received at a signum function module and the output signal sgn(S.sub.out -V.sub.L) is formed and issued. The signum output signal is multiplied by a controllable real number m and is added to V.sub.L to form the threshold signal V.sub.T. The desired output signal S.sub.out makes a transition between the first and second logical levels, with a time delay that is controllable by choice of the number m.
    Type: Grant
    Filed: December 20, 1990
    Date of Patent: October 13, 1992
    Assignee: Hewlett-Packard Company
    Inventors: John J. Corcoran, Kenneth D. Poulton