Patents by Inventor Sugbong Kang

Sugbong Kang 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).

  • Publication number: 20160373182
    Abstract: A system may comprise a plurality of signal processing paths, a bin-wise combiner, an inverse transformation block, and a DAC. Each signal processing path may comprise a transformation block that is operable to transform a first time-domain digital signal to an associated frequency-domain signal having a plurality of subband signals. The bin-wise combiner may be operable to combine corresponding subband signals of the plurality of signal processing paths. The inverse transformation block may be operable to transform output of the bin-wise combiner to an second time-domain signal. The DAC may be operable to converts the second time-domain signal to a corresponding analog signal.
    Type: Application
    Filed: August 30, 2016
    Publication date: December 22, 2016
    Inventors: Anand K. Anandakumar, Curtis Ling, Sugbong Kang
  • Patent number: 9432698
    Abstract: Received data packets are groomed to improve performance of MPEG-2 transport stream packet in a digital video broadcasting system. Multitude of crosschecking techniques are applied to ensure that crucial pieces of information such as the packet identifier (PID) field, the continuity counter (CC) field, table ID, section length, IP header checksum, table and frame boundaries, application data table size are corrected if necessary.
    Type: Grant
    Filed: July 15, 2014
    Date of Patent: August 30, 2016
    Assignee: MaxLinear, Inc.
    Inventors: Sridhar Ramesh, Sugbong Kang, Brenndon Lee
  • Patent number: 9432104
    Abstract: A system may comprise a plurality of signal processing paths, a bin-wise combiner, an inverse transformation block, and a DAC. Each signal processing path may comprise a transformation block that is operable to transform a first time-domain digital signal to an associated frequency-domain signal having a plurality of subband signals. The bin-wise combiner may be operable to combine corresponding subband signals of the plurality of signal processing paths. The inverse transformation block may be operable to transform output of the bin-wise combiner to an second time-domain signal. The DAC may be operable to converts the second time-domain signal to a corresponding analog signal.
    Type: Grant
    Filed: April 20, 2015
    Date of Patent: August 30, 2016
    Assignee: Maxlinear, Inc.
    Inventors: Anand K. Anandakumar, Curtis Ling, Sugbong Kang
  • Patent number: 9419661
    Abstract: An impulse noise mitigation circuit (INMC) may set a cut-off frequency of each of two high pass filters to bound a frequency bandwidth of a desired signal, wherein a first of the two filters allows frequencies higher than the frequency bandwidth of the desired signal, and a second of the two filters allows frequencies lower than the frequency bandwidth of the desired signal. The INMC may compute and store a mean magnitude separately for a first signal response of the first filter and a second signal response of the second filter. The INMC may select the first filter for impulse noise mitigation when the mean magnitude of the second filter is greater than the mean magnitude of the first filter. The INMC may select the second filter for impulse noise mitigation when the mean magnitude of the first filter is greater than the second filter.
    Type: Grant
    Filed: October 19, 2015
    Date of Patent: August 16, 2016
    Assignee: Maxlinear, Inc
    Inventors: Andy Lo, Sugbong Kang
  • Publication number: 20160049971
    Abstract: An impulse noise mitigation circuit (INMC) may set a cut-off frequency of each of two high pass filters to bound a frequency bandwidth of a desired signal, wherein a first of the two filters allows frequencies higher than the frequency bandwidth of the desired signal, and a second of the two filters allows frequencies lower than the frequency bandwidth of the desired signal. The INMC may compute and store a mean magnitude separately for a first signal response of the first filter and a second signal response of the second filter. The INMC may select the first filter for impulse noise mitigation when the mean magnitude of the second filter is greater than the mean magnitude of the first filter. The INMC may select the second filter for impulse noise mitigation when the mean magnitude of the first filter is greater than the second filter.
    Type: Application
    Filed: October 19, 2015
    Publication date: February 18, 2016
    Inventors: Andy Lo, Sugbong Kang
  • Publication number: 20150326301
    Abstract: A system may comprise a plurality of signal processing paths, a bin-wise combiner, an inverse transformation block, and a DAC. Each signal processing path may comprise a transformation block that is operable to transform a first time-domain digital signal to an associated frequency-domain signal having a plurality of subband signals. The bin-wise combiner may be operable to combine corresponding subband signals of the plurality of signal processing paths. The inverse transformation block may be operable to transform output of the bin-wise combiner to an second time-domain signal. The DAC may be operable to converts the second time-domain signal to a corresponding analog signal.
    Type: Application
    Filed: April 20, 2015
    Publication date: November 12, 2015
    Inventors: Anand K. Anandakumar, Curtis Ling, Sugbong Kang
  • Patent number: 9166637
    Abstract: An impulse noise mitigation circuit (INMC) may set a cut-off frequency of each of two high pass filters to bound a frequency bandwidth of a desired signal, wherein a first of the two filters allows frequencies higher than the frequency bandwidth of the desired signal, and a second of the two filters allows frequencies lower than the frequency bandwidth of the desired signal. The INMC may compute and store a mean magnitude separately for a first signal response of the first filter and a second signal response of the second filter. The INMC may select the first filter for impulse noise mitigation when the mean magnitude of the second filter is greater than the mean magnitude of the first filter. The INMC may select the second filter for impulse noise mitigation when the mean magnitude of the first filter is greater than the second filter.
    Type: Grant
    Filed: July 28, 2014
    Date of Patent: October 20, 2015
    Assignee: MAXLINEAR, INC.
    Inventors: Andy Lo, Sugbong Kang
  • Patent number: 9014649
    Abstract: A system may comprise a plurality of signal processing paths, a bin-wise combiner, an inverse transformation block, and a DAC. Each signal processing path may comprise a transformation block that is operable to transform a first time-domain digital signal to an associated frequency-domain signal having a plurality of subband signals. The bin-wise combiner may be operable to combine corresponding subband signals of the plurality of signal processing paths. The inverse transformation block may be operable to transform output of the bin-wise combiner to an second time-domain signal. The DAC may be operable to converts the second time-domain signal to a corresponding analog signal.
    Type: Grant
    Filed: March 31, 2014
    Date of Patent: April 21, 2015
    Assignee: MaxLinear, Inc.
    Inventors: Anand K. Anandakumar, Curtis Ling, Sugbong Kang
  • Publication number: 20150078242
    Abstract: Received data packets are groomed to improve performance of MPEG-2 transport stream packet in a digital video broadcasting system. Multitude of crosschecking techniques are applied to ensure that crucial pieces of information such as the packet identifier (PID) field, the continuity counter (CC) field, table ID, section length, IP header checksum, table and frame boundaries, application data table size are corrected if necessary.
    Type: Application
    Filed: July 15, 2014
    Publication date: March 19, 2015
    Inventors: Sridhar RAMESH, Sugbong KANG, Brenndon LEE
  • Publication number: 20150030110
    Abstract: An impulse noise mitigation circuit (INMC) may set a cut-off frequency of each of two high pass filters to bound a frequency bandwidth of a desired signal, wherein a first of the two filters allows frequencies higher than the frequency bandwidth of the desired signal, and a second of the two filters allows frequencies lower than the frequency bandwidth of the desired signal. The INMC may compute and store a mean magnitude separately for a first signal response of the first filter and a second signal response of the second filter. The INMC may select the first filter for impulse noise mitigation when the mean magnitude of the second filter is greater than the mean magnitude of the first filter. The INMC may select the second filter for impulse noise mitigation when the mean magnitude of the first filter is greater than the second filter.
    Type: Application
    Filed: July 28, 2014
    Publication date: January 29, 2015
    Inventors: Andy Lo, Sugbong Kang
  • Publication number: 20140294056
    Abstract: A system may comprise a plurality of signal processing paths, a bin-wise combiner, an inverse transformation block, and a DAC. Each signal processing path may comprise a transformation block that is operable to transform a first time-domain digital signal to an associated frequency-domain signal having a plurality of subband signals. The bin-wise combiner may be operable to combine corresponding subband signals of the plurality of signal processing paths. The inverse transformation block may be operable to transform output of the bin-wise combiner to an second time-domain signal. The DAC may be operable to converts the second time-domain signal to a corresponding analog signal.
    Type: Application
    Filed: March 31, 2014
    Publication date: October 2, 2014
    Applicant: MAXLINEAR, INC.
    Inventors: Anand K. Anandakumar, Curtis Ling, Sugbong Kang
  • Patent number: 8817780
    Abstract: Received data packets are groomed to improve performance of MPEG-2 transport stream packet in a digital video broadcasting system. Multitude of crosschecking techniques are applied to ensure that crucial pieces of information such as the packet identifier (PID) field, the continuity counter (CC) field, table ID, section length, IP header checksum, table and frame boundaries, application data table size are corrected if necessary.
    Type: Grant
    Filed: August 7, 2008
    Date of Patent: August 26, 2014
    Assignee: MaxLinear, Inc.
    Inventors: Sridhar Ramesh, Sugbong Kang, Brenndon Lee
  • Patent number: 8792543
    Abstract: An impulse noise mitigation circuit (INMC) may set a cut-off frequency of each of two high pass filters to bound a frequency bandwidth of a desired signal, wherein a first of the two filters allows frequencies higher than the frequency bandwidth of the desired signal, and a second of the two filters allows frequencies lower than the frequency bandwidth of the desired signal. The INMC may compute and store a mean magnitude separately for a first signal response of the first filter and a second signal response of the second filter. The INMC may select the first filter for impulse noise mitigation when the mean magnitude of the second filter is greater than the mean magnitude of the first filter. The INMC may select the second filter for impulse noise mitigation when the mean magnitude of the first filter is greater than the second filter.
    Type: Grant
    Filed: July 15, 2013
    Date of Patent: July 29, 2014
    Assignee: MaxLinear, Inc.
    Inventors: Andy Lo, Sugbong Kang
  • Patent number: 8688064
    Abstract: A receiver may comprise a plurality of signal processing paths, a bin-wise combiner, and an inverse transformation block. Each signal processing path may comprise a transformation block that is operable to transform a time-domain digital signal to an associated frequency-domain signal having a plurality of subband signals. The bin-wise combiner may be operable to combine corresponding subband signals of the plurality of signal processing paths. The inverse transformation block may be operable to transform output of the bin-wise combiner to an associated time-domain signal. The transformation block in each said signal processing path may be a Fast Fourier Transform (FFT) block. The number of points used by the FFT block of any one of said plurality of signal processing paths may be based on the delay spread of a signal input to the one of the signal processing paths.
    Type: Grant
    Filed: July 24, 2012
    Date of Patent: April 1, 2014
    Assignee: MaxLinear, Inc.
    Inventors: Anand K. Anandakumar, Curtis Ling, Sugbong Kang
  • Publication number: 20130301766
    Abstract: An impulse noise mitigation circuit (INMC) may set a cut-off frequency of each of two high pass filters to bound a frequency bandwidth of a desired signal, wherein a first of the two filters allows frequencies higher than the frequency bandwidth of the desired signal, and a second of the two filters allows frequencies lower than the frequency bandwidth of the desired signal. The INMC may compute and store a mean magnitude separately for a first signal response of the first filter and a second signal response of the second filter. The INMC may select the first filter for impulse noise mitigation when the mean magnitude of the second filter is greater than the mean magnitude of the first filter. The INMC may select the second filter for impulse noise mitigation when the mean magnitude of the first filter is greater than the second filter.
    Type: Application
    Filed: July 15, 2013
    Publication date: November 14, 2013
    Inventors: Andy Lo, Sugbong Kang
  • Patent number: 8548411
    Abstract: A system may comprise a plurality of signal processing paths, a bin-wise combiner, an inverse transformation block, and a DAC. Each signal processing path may comprise a transformation block that is operable to transform a first time-domain digital signal to an associated frequency-domain signal having a plurality of subband signals. The bin-wise combiner may be operable to combine corresponding subband signals of the plurality of signal processing paths. The inverse transformation block may be operable to transform output of the bin-wise combiner to an second time-domain signal. The DAC may be operable to converts the second time-domain signal to a corresponding analog signal.
    Type: Grant
    Filed: August 1, 2012
    Date of Patent: October 1, 2013
    Assignee: MaxLinear, Inc.
    Inventors: Anand K. Anandakumar, Curtis Ling, Sugbong Kang
  • Patent number: 8488663
    Abstract: A noise abatement method and system for impulse noise in an RF receiver where the RF analog signal is converted to a digital signal prior to being connected to a demodulator. Two filters are used to detect impulse noise signals even under out-of-band interferer conditions, and prevent the impulse noise from reaching the input to the demodulator. A first of the two filters detects impulse noise using signals lower than the frequency bandwidth of the desired signal, and a second of the two filters detects impulse noise using signals higher the frequency bandwidth of the desired signal. A mean magnitude of the signal is detected over a predetermined time T and is used to select which filter to use for noise abatement.
    Type: Grant
    Filed: September 22, 2010
    Date of Patent: July 16, 2013
    Assignee: MaxLinear, Inc.
    Inventors: Andy Lo, Sugbong Kang
  • Patent number: 8472912
    Abstract: A diversity receiver includes a first RF front end module for receiving a first RF signal, and frequency converting the first RF signal and outputting a first diversity signal, a second RF front end module for receiving a second RF signal, frequency converting the second RF signal and outputting a second diversity signal, a first converter for converting the first diversity signal to a first time-domain signal, a second converter for converting the second diversity signal to a second time-domain signal, a first transformer for translating the first time-domain signal to a first frequency-domain signal, a second transformer for translating the second time-domain signal to a second frequency-domain signal, a first pre-equalizer for equalizing the first frequency-domain signal, a second pre-equalizer for equalizing the second frequency-domain signal, and a combiner for combining the first and second pre-equalized frequency-domain signals.
    Type: Grant
    Filed: December 13, 2010
    Date of Patent: June 25, 2013
    Assignee: MaxLinear, Inc.
    Inventors: Sridhar Ramesh, James Qiu, Sugbong Kang
  • Publication number: 20120294399
    Abstract: A system may comprise a plurality of signal processing paths, a bin-wise combiner, an inverse transformation block, and a DAC. Each signal processing path may comprise a transformation block that is operable to transform a first time-domain digital signal to an associated frequency-domain signal having a plurality of subband signals. The bin-wise combiner may be operable to combine corresponding subband signals of the plurality of signal processing paths. The inverse transformation block may be operable to transform output of the bin-wise combiner to an second time-domain signal. The DAC may be operable to converts the second time-domain signal to a corresponding analog signal.
    Type: Application
    Filed: August 1, 2012
    Publication date: November 22, 2012
    Inventors: Anand K. Anandakumar, Curtis Ling, Sugbong Kang
  • Publication number: 20120289177
    Abstract: A receiver may comprise a plurality of signal processing paths, a bin-wise combiner, and an inverse transformation block. Each signal processing path may comprise a transformation block that is operable to transform a time-domain digital signal to an associated frequency-domain signal having a plurality of subband signals. The bin-wise combiner may be operable to combine corresponding subband signals of the plurality of signal processing paths. The inverse transformation block may be operable to transform output of the bin-wise combiner to an associated time-domain signal. The transformation block in each said signal processing path may be a Fast Fourier Transform (FFT) block. The number of points used by the FFT block of any one of said plurality of signal processing paths may be based on the delay spread of a signal input to the one of the signal processing paths.
    Type: Application
    Filed: July 24, 2012
    Publication date: November 15, 2012
    Inventors: Anand K. Anandakumar, Curtis Ling, Sugbong Kang