The patent badge is an abbreviated version of the USPTO patent document. The patent badge does contain a link to the full patent document.
The patent badge is an abbreviated version of the USPTO patent document. The patent badge covers the following: Patent number, Date patent was issued, Date patent was filed, Title of the patent, Applicant, Inventor, Assignee, Attorney firm, Primary examiner, Assistant examiner, CPCs, and Abstract. The patent badge does contain a link to the full patent document (in Adobe Acrobat format, aka pdf). To download or print any patent click here.
Patent No.:
Date of Patent:
Nov. 30, 2004
Filed:
Oct. 29, 1999
Gordon Taylor Davis, Chapel Hill, NC (US);
Jeffrey Haskell Derby, Chapel Hill, NC (US);
Malcolm Scott Ware, Raleigh, NC (US);
Charles Robert Young, Cary, NC (US);
International Business Machines Corporation, Armonk, NY (US);
Abstract
Data rate control systems, methods, and computer program products in which an error counter is maintained that contains an error count. The error counter is periodically sampled to determine a sampling interval error count corresponding to a change in the error count since a previous read of the error counter. The sampling interval error count is provided to a first filter that is characterized by a slow time-constant and a second filter that is characterized by a fast time-constant. The first filter generates a slow-filtered sampling interval error count while the second filter generates a fast-filtered sampling interval error count, which are used as a basis for generating a data rate slowdown request signal. More specifically, the data rate slowdown request signal is generated if either the slow-filtered sampling interval error count or the fast-filtered sampling interval error count exceeds a threshold respectively associated therewith. Advantageously, the first filter, which is characterized by a slow time-constant, may be used to respond to lower numbers of errors that are sustained over an extended time period while the second filter, which is characterized by a fast time-constant, may be used to respond to large error bursts spanning a shorter time period. Moreover, by filtering the errors accumulated during each sampling interval, sensitivity with respect to the position in time of a given error burst relative to boundaries between sampling intervals may be avoided.