Optical: systems and elements – Deflection using a moving element – Using a periodically moving element
Reexamination Certificate
1998-07-30
2001-05-15
Chan, Jason (Department: 2733)
Optical: systems and elements
Deflection using a moving element
Using a periodically moving element
C370S244000
Reexamination Certificate
active
06233073
ABSTRACT:
TECHNICAL FIELD
This invention relates to high speed fiber optic networks having built in protection against faults, and, more particularly, to the testing of such networks for fault tolerance and error recovery.
BACKGROUND OF THE INVENTION
Fault tolerance presents one of the greatest challenges to the design of modern high speed fiber optic networks. Unlike earlier point-to-point fiber optic subsystems, such as ESCON, for connecting computer systems, the modern fiber optic networks include interfaces which can handle a variety of different classes of transfers, for example, as defined for the “Fibre Channel” fiber optic network: Class
1
—Dedicated connection (point-to-point); Class
2
—Frame multiples (shared but guaranteed); Class
3
—Datagram (ship-and-pray); and Class
4
—Lossy voice and video. The characteristics of the interfaces are implemented in standard protocols, and will support many different transfer mechanisms, such as modern ESCON, FC, SSA, ATM, and FDDI protocols, to name a few. These protocols all employ a similar serial transport layer, comprising a “Fiber Optic Subassembly” (FOSA), fiber, and encoding/decoding modules. They are designed to be fast, dependable, and cover great distances, linking computer systems and components that are in different rooms, buildings and cities.
Demands on such links and networks have required that data availability and integrity be an important part of any network component design. As components have become more reliable, redundant, and fault tolerant, the interfaces linking these components or systems are now typically the most failure-prone aspect.
Hence, in order to provide error detection capabilities, most fiber optic protocols incorporate the IBM 8 to 10 bit encoding scheme with running disparity. This arrangement is described in the IBM Journal of Research and Development, volume 27, number 5, Sep. 1983, pp. 440-450, A. X. Widmer and P. A. Franazek, “A DC-Balanced, Partition-Block, 8B/10B Transmission Code” (herein, the Widmer and Franazek publication). There are also other checks built in to these protocols, providing several layers of protection. Many of the checks are logic circuits to detect that the frames are complete and that the protocols are correct.
Additionally, circuits are provided for assuring a level of error detection. For example, U.S. Pat. No. 5,229,875, Glista, provides a fiber optic coupler-repeater which analyzes the power of the optical signals from upstream terminals and compares the signal power to a preset value range and rejects those out of the range, activating a built-in “test circuit” alarm telling downstream terminals of the failure. As another example, U.S. Pat. No. 5,396,357, Goosen et al., describes a cross-channel data link which includes wavelength division multiplexing in opposite directions and logic circuity that “guarantees” that the interconnected systems have identical data irrespective of any single point failure in the data link.
SUMMARY OF THE INVENTION
It is an object of the present invention to verify that a fault tolerant fiber optic network design has successfully implemented error detection schemes.
Disclosed are a subsystem and a method for diagnosing a fiber optic network, the fiber optic network comprising a fiber having an optic stream adhering to a defined protocol of the fiber optic network, and comprising a target for receiving and responding to the optic stream. The diagnosis comprises receiving the fiber optic stream and converting the fiber optic stream to an electrical input stream with a FOSA (fiber optic subassembly), and recognizing, from the electrical input stream, a selected event related to the optic stream defined protocol. Then, in response to the recognition of the selected event, recognizing a selected pattern of the electrical input stream, preferably employing mask logic, and injecting, in response to the recognition of the selected pattern, a transmission error into the electrical stream and converted into the optic stream, and transmitting the optic stream with the injected transmission error to the target, employing a FOSA.
A FIFO buffer is coupled to the first FOSA for buffering the electrical input stream, and the mask logic is coupled to the FIFO buffer for injecting the transmission error at a selected point in the optic stream.
REFERENCES:
patent: 5228105 (1993-07-01), Glista
patent: 5229875 (1993-07-01), Glista
patent: 5321813 (1994-06-01), McMillen et al.
patent: 5396357 (1995-03-01), Goossen et al.
patent: 5524218 (1996-06-01), Byers et al.
patent: 5559963 (1996-09-01), Gregg et al.
patent: 5602667 (1997-02-01), Patel
patent: 5706278 (1998-01-01), Robillard et al.
patent: 5774242 (1998-06-01), O'Syllivan
Bowers Michael Dana
Denning Donald Eugene
Emberty Robert George
Klein Craig Anthony
Chan Jason
Holcombe John H.
International Business Machines - Corporation
Nguyen Chau M.
Sullivan Robert M.
LandOfFree
Diagnostic injection of transmission errors in fiber optic... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Diagnostic injection of transmission errors in fiber optic..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Diagnostic injection of transmission errors in fiber optic... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2563837