Abstract: A data networking system designed to provide efficient, yet reliable, transportation of data across a time division multiplexed network. This invention allows for redundantly sending data segments wherein if bad data segments are received, they can be resent in a dynamic manner depending on the network condition and segments can be fragmented, resized and sent with low overhead due to dynamic header sizes, segment sizes, forward error correction, and cyclic redundancy checks. The result is a very adaptive network that meets the demanding requirements of networks such as power line and wireless while still being able to transport data for voice, audio, video, computer, control, and the like.
Type:
Grant
Filed:
February 3, 2004
Date of Patent:
November 11, 2008
Assignee:
Phonex Broadband Corporation
Inventors:
Michael J. Miller, Lyman D. Horne, W. Paul Willes, Douglas M. Grover
Abstract: A digital content transmission and reception system that uses feedback and re-transmission of missing content is described. A content transmission system broadcasts a complete set of digital content to a plurality of content reception systems via a communication link. A content reception system receives a corresponding incomplete set of digital content, determines particular content portions that are missing, and provided feedback indicating the missing content portions to the content distribution system. The content distribution system then re-transmits content based on the feedback.
Abstract: A method for determining Cyclic Redundancy Check (CRC) parity of data, such data comprising a plurality of bytes, each one of the bytes having a parity bit, the plurality of bytes of data having a CRC. The method includes generating the parity of the parity bits of the plurality of bytes of the data, such generated parity being the parity of the CRC of such data. The method includes: generating parity of the parity bits of the plurality of data bytes; and comparing such generated parity with the parity bit of the CRC of the data. The method receives data having a plurality of N bytes: [D(0), D(1), . . . , D(N?1]) each byte having a parity bit p and computes the parity of [P(0), P(1), . . . P(N?1)].
Abstract: A method of managing faults in a computer system including the steps of detecting an error, handling the error along a functional hierarchy and, further, handling a fault that caused the error in a management hierarchy which can be operated separately from the functional hierarchy.
Type:
Grant
Filed:
October 26, 1994
Date of Patent:
April 14, 1998
Assignee:
Digital Equipment Corporation
Inventors:
Jeffrey L. Gardiner, Gerhard K. Heider, Larry W. Emlich, Bruce H. Luhrs, Michael C. Li, Michael R. Masters, Russell Lloyd Myers, Harlo A. Peterson, Frank M. Robbins, Mark J. Seger