Patents by Inventor Ko-Chung Tseng

Ko-Chung Tseng 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: 20200220558
    Abstract: A method and apparatus is described for assigning columns of an LDPC H matrix to a plurality of decoding logics for efficient decoding of codewords. The rows of the LDPC H matrix are evaluated in a number of different orderings, and for each row in each ordering, a number of columns containing non-zero circulants are determined that cannot be evenly distributed to a plurality of decoding logics. As each row is evaluated, one or more columns of the LDPC H matrix are assigned to temporary bins for storage. After the LDPC H matrix has been evaluated a plurality of times, the arrangement that resulted in the fewest number of “mismatched” columns is selected, and the columns of the LDPC H matrix that were assigned to the temporary storage bins for that particular row arrangement is used to assign the columns in the bins to the plurality of decoding logics.
    Type: Application
    Filed: December 20, 2018
    Publication date: July 9, 2020
    Inventors: Ko-Chung Tseng, Chandra Varanasi, Engling Yeo
  • Patent number: 10476524
    Abstract: A method and apparatus is described for assigning columns of an LDPC H matrix to a plurality of decoding logics for efficient decoding of codewords. The rows of the LDPC H matrix are evaluated in a number of different orderings, and for each row in each ordering, a number of columns containing non-zero circulants are determined that cannot be evenly distributed to a plurality of decoding logics. As each row is evaluated, one or more columns of the LDPC H matrix are assigned to temporary bins for storage. After the LDPC H matrix has been evaluated a plurality of times, the arrangement that resulted in the fewest number of “mismatched” columns is selected, and the columns of the LDPC H matrix that were assigned to the temporary storage bins for that particular row arrangement is used to assign the columns in the bins to the plurality of decoding logics.
    Type: Grant
    Filed: December 20, 2017
    Date of Patent: November 12, 2019
    Assignee: Goke US Research Laboratory
    Inventors: Ko-Chung Tseng, Chandra Varanasi, Engling Yeo
  • Patent number: 10419026
    Abstract: A method and apparatus for efficient data decoding is described. Data is encoded by an LDPC encoder using a G matrix. An LDPC decoder uses a modified H matrix to decode encoded blocks of data, the modified H matrix having at least two columns of its circulants swapped with each other. The encoded blocks of data are stored, decoded and reconstructed in an order that considers the circulants in the columns that have been swapped.
    Type: Grant
    Filed: November 27, 2017
    Date of Patent: September 17, 2019
    Assignee: Goke US Research Laboratory
    Inventors: Ko-Chung Tseng, Chandra Varanasi, Engling Yeo
  • Patent number: 10374631
    Abstract: Look-ahead LDPC decoder. In an exemplary embodiment, a method includes generating a message (QA) to a first check node, the QA message is generated from a result (RA) from the first check node, and generating a message (QB) to a second check node, the QB message is generated from the result (RA) and is transmitted to avoid decoder stall. The method also includes receiving a result (R?A) from the first check node, computing a difference (R?A) between the result (R?A) and the result (RA), and receiving a result (R?B) from the second check node. The method also includes computing a bit value P(B) using the difference (R?A) and the result (R?B).
    Type: Grant
    Filed: August 22, 2017
    Date of Patent: August 6, 2019
    Assignee: Goke US Research Laboratory
    Inventors: Ko-Chung Tseng, Engling Yeo
  • Publication number: 20190165809
    Abstract: A method and apparatus is described for assigning columns of an LDPC H matrix to a plurality of decoding logics for efficient decoding of codewords. The rows of the LDPC H matrix are evaluated in a number of different orderings, and for each row in each ordering, a number of columns containing non-zero circulants are determined that cannot be evenly distributed to a plurality of decoding logics. As each row is evaluated, one or more columns of the LDPC H matrix are assigned to temporary bins for storage. After the LDPC H matrix has been evaluated a plurality of times, the arrangement that resulted in the fewest number of “mismatched” columns is selected, and the columns of the LDPC H matrix that were assigned to the temporary storage bins for that particular row arrangement is used to assign the columns in the bins to the plurality of decoding logics.
    Type: Application
    Filed: December 20, 2017
    Publication date: May 30, 2019
    Inventors: Ko-Chung Tseng, Chandra Varanasi, Engling Yeo
  • Publication number: 20190165810
    Abstract: A method and apparatus for efficient data decoding is described. Data is encoded by an LDPC encoder using a G matrix. An LDPC decoder uses a modified H matrix to decode encoded blocks of data, the modified H matrix having at least two columns of its circulants swapped with each other. The encoded blocks of data are stored, decoded and reconstructed in an order that considers the circulants in the columns that have been swapped.
    Type: Application
    Filed: November 27, 2017
    Publication date: May 30, 2019
    Inventors: Ko-Chung Tseng, Chandra Varanasi, Engling Yeo
  • Publication number: 20190068223
    Abstract: Look-ahead LDPC decoder. In an exemplary embodiment, a method includes generating a message (QA) to a first check node, the QA message is generated from a result (RA) from the first check node, and generating a message (QB) to a second check node, the QB message is generated from the result (RA) and is transmitted to avoid decoder stall. The method also includes receiving a result (R?A) from the first check node, computing a difference (R?A) between the result (R?A) and the result (RA), and receiving a result (R?B) from the second check node. The method also includes computing a bit value P(B) using the difference (R?A) and the result (R?B).
    Type: Application
    Filed: August 22, 2017
    Publication date: February 28, 2019
    Inventors: Ko-Chung Tseng, Engling Yeo