Multiplex communications – Wide area network – Packet switching
Patent
1992-09-25
1995-04-25
Chin, Wellington
Multiplex communications
Wide area network
Packet switching
370 941, 371 33, H04L 112, H04L 116
Patent
active
054105366
DESCRIPTION:
BRIEF SUMMARY
TECHNICAL FIELD OF THE INVENTION
This invention relates to the field of data communication and more particularly to a method for recovering from errors occurring during transmission of data between nodes in a digital system.
BACKGROUND OF THE INVENTION
Data communication systems and the elements making up data communication systems involve the electronic transmission of data over a link from one node (e.g computer or terminal) of the system to another. To ensure communication in an orderly fashion over the data link, a uniform method of sending and receiving information is required. This uniformity is achieved by means of a protocol (set of rules) used for the management of a data link in the communication system. Protocols are used to perform such functions as establishing the conversation between two nodes in the communication system, identifying time sender and receiver, acknowledging received information and node initialisation. The exact procedure and function performed depends on the protocol used. Data link protocols may be classified in two categories; bit oriented protocols (BOP) and byte oriented protocols.
Prior bit oriented protocols include the Synchronous Data Link Control (SDLC) protocol which was introduced by IBM in 1973 and the High Level Data Link Control (HDLC) protocol. All communications in a BOP system are in the form of frames of uniform format which comprise a number of fields each having a definite location and precise meaning. In HDLC a frame commonly starts with an eight bit flag sequence which is followed by ADDRESS and CONTROL fields after which may follow an INFORMATION field (depending on the function of the frame). The INFORMATION field is followed by a FRAME CHECK SEQUENCE field and the end of the frame is delimited by another flag sequence. The Address and Control fields in HDLC each comprise a single octet of bits. The information field may contain a variable number of bits, in the form of an integral number of octets, up to a predefined limit. The FCS field commonly comprises a pair of octets.
In HDLC, the CONTROL field defines the function of the frame. There are three basic types: Information, Supervisory and Unnumbered which are referred to as I-frames, S-frames and U-frames. The I-frame is used to provide for information transfer across the link and contains an INFORMATION field. The S-frame is used to perform supervisory functions on the link and may be used to acknowledge I-frames, or to request retransmission of frames. The U-frame is used particularly in error recovery.
HDLC is often employed in systems wherein data communication is over relatively long distances where there will be a number of data frames on the link at any one time. The method of acknowledgement that data has been received has to be capable of detecting the incorrect transmission of any one of these data frames. An implied acknowledgement technique is used which enables frame acknowledgement information to be included within an I-frame. This is accomplished by assigning identification numbers, called sequence numbers, to received and transmitted frames. These numbers contain information pertaining to the number of frames transmitted and received by the individual node. By checking these numbers, the node can compare the number of received frames with the number of transmitted frames and take the appropriate error recovery action if a discrepancy exists. Although the packet sequence numbers used in the described implied acknowledgment technique may be included within an I-frame, if information frames are not being sent by the node receiving the data frame to be acknowledged, then it is necessary to include the sequence number information in a separate S-frame. Details of one subset of HDLC can be found in `X25 explained` by R J Deasington, published by Ellis Horwood Limited.
Errors occurring during data transmission between nodes are corrected in different ways depending on the type of error detected. If the receiver receives a frame which is out of sequence, an S-frame with a reject (REJ)
REFERENCES:
patent: 4422171 (1983-12-01), Wortley et al.
patent: 4439859 (1984-03-01), Donnan
patent: 4601035 (1986-07-01), Marzec et al.
patent: 5010553 (1991-04-01), Scheller et al.
Cockburn Gordon J.
Grainger Bernard J.
Judd Ian D.
Shah Vinay V.
Bussan Matthew J.
Chin Wellington
International Business Machines - Corporation
Truelson Roy W.
LandOfFree
Method of error recovery in a data communication system does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method of error recovery in a data communication system, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method of error recovery in a data communication system will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-1572894