Multiplex communications – Diagnostic testing – Fault detection
Reexamination Certificate
2000-11-13
2004-07-20
Rao, Seema S. (Department: 2666)
Multiplex communications
Diagnostic testing
Fault detection
C714S025000, C714S048000, C714S704000, C714S715000, C370S392000
Reexamination Certificate
active
06765874
ABSTRACT:
TECHNICAL FIELD
The present invention relates to the measurement of a time interval during which asynchronous transfer mode (ATM) transmission of data in a data transmission channel is disrupted.
BACKGROUND ART
A process called “protection switching” is used in linear and ring Synchronous Digital Hierarchy (SDH) networks, sometimes called Synchronous Optical Networks (SONET), in order to re-establish asynchronous transfer mode (ATM) transmission of data when the performance of a data transmission channel is degraded, the channel is cut, or otherwise disrupted. This can happen because of a cut to an optical fibre, or equipment failure at end points or switching points in the network. Such service disruption events can lead to an unrecoverable loss of data, and so it is important to be able to quantify the possible loss up until the time protection switching re-establishes error-free transmission of data. Such testing is often done when an SDH network or part of a network is installed and commissioned.
It is not enough simply to verify that protection switching will be activated when the transmission of data is disrupted. To minimise the disruption to data transmission, the protection switching must be completed within specified time limits recommended by the standards ITU-T Recommendations G.783 and G.841. It is then therefore necessary to measure or estimate the protection switching performance of SDH network equipment to compare this against the ITU-T Recommendations.
ATM data transmission consists of identifiable data cells transmitted in a defined order along an ATM Virtual Channel, i.e. a group of ATM data cells each having the same header. More than one such ATM virtual channel may undergo protection switching in a service disruption event. A service disruption event can be simulated in testing by transmitting ATM data along one or more virtual channels, the data having a data transmission rate comparable to the expected utilisation of the virtual channels, and then by disconnecting or breaking a fibre or by disabling equipment using the virtual channels, such as an ATM switch. Although the so-called “physical layer”, the fibres and connecting equipment, can be reconfigured very quickly to establish new virtual channels, the data cell stream may suffer impairments, such as erroneous detected data in a cell, lost cells, or misinserted cells where the ordering of received cells is incorrect.
One simple way of estimating if the protection switching is within the recommended limits is to count the number of lost or errored cells, and then divide this number by the cell transmission rate. If 100 cells are lost or errored at a transmission rate of 100,000 cells per second, then the duration of the service disruption event can be estimated as 10 ms. It has been found, however, that this simple way of estimating the extent of a service disruption event, overestimates the effectiveness of protection switching in reestablishing data transmission in an ATM network.
It is an object of the invention to provide an improved method and apparatus for measuring the duration of a service disruption event in an asynchronous transfer mode network.
DISCLOSURE OF INVENTION
According to the invention, there is provided a method of measuring the duration of a service disruption event in an asynchronous transfer mode (ATM) data transmission channel, using a data analyzer that has both a disruption timer and a guard timer, the method comprising the steps of:
a) providing an ATM stream of data into the data transmission channel, the data comprising a stream of identifiable data cells in a defined order;
b) detecting with a data analyzer the stream of data after transmission through the data transmission channel in order to determine whether or not any of the data cells have been corrupted in transmission through the data transmission channel;
c) disrupting the transmission of the stream of data through the data transmission channel so that at least some of the cells are corrupted;
d) using the disruption timer to time the service disruption event when a first corrupted data cell is detected by the data analyzer;
e) using the guard timer to time a guard time interval when a subsequent data cell that is not corrupted is detected by the data analyzer whilst continuing to use the disruption timer to time the service disruption event; and then either
f) resetting the guard timer if the data analyzer has detected a further corrupted data cell prior to the guard timer timing a predetermined guard time; or
g) determining the duration of the service disruption event from the disruption timer when the guard timer has timed the predetermined guard time.
A protection switching event will often include a period of increased error rate for data in cells, for example due to impaired signal-to-noise levels, or increased timing jitter, prior to a complete loss of cells. Some cells can be lost completely, and others can be in the wrong order or misinserted when the data cells are received. In all of these cases, the data cells have become corrupted. The data analyzer may therefore characterise corrupted data cells according to whether cells have been errored, for example, comprising data which is in error according to a cyclic redundancy check (CRC), or lost or misinserted.
The service disruption event therefore lasts at least from the last uncorrupted data cells prior to the switching event up until the time when data cells are again consistently uncorrupted. The guard timer therefore accommodates the observed performance of switching events, in which isolated uncorrupted cells or groups of consecutive uncorrupted cells can be transmitted, so that the duration of the service disruption event is measured by the disruption timer as continuing even when some correct data cells are being transmitted. The predetermined guard time can then be selected empirically according to the performance of typical protection switching events.
In step g) the duration of the service disruption event may be a time measured from a start time prior to detection by the data analyzer of the first corrupted data cell until an end time at or following the consistent resumption of uncorrupted data cells. The end time may be at a time prior to the time at which the guard timer has reached a time at least as great as the guard time in step g).
Including a proportion of an appropriate predetermined guard time in the measured disruption time, whether at the beginning and/or the end of the measured period, provides the benefit that the measurement does not underestimate the performance of the switching event in comparison with the performance recommended under international standards.
The start time may be a time no later than the last uncorrupted data cell before the service disruption event, and in a preferred embodiment of the invention, the start time may be in advance of this time, for example being a time which precedes said last uncorrupted data cell by no more than the guard time. The guard time then serves the purpose of providing a conservative measure of both the start and the end of the service disruption event.
In some circumstances, it may be necessary to measure the service disruption event hen there is a prolonged break in the data cell stream. In this case, when the prolonged break is at the start of the protection switching event, the start time may be a time that may precede said last uncorrupted data cell by more than the guard time when there is a gap in the stream of data cells following the last uncorrupted data cell greater than the guard time.
When cells are lost or misinserted, it may be that cells arrive in the right order after a series of erroneously inserted cells. It will not be possible to say with any certainty that cells are again arriving in the correct order until more than one such correct cell has been received. Therefore, in step e) said subsequent data cell may be the second consecutive data cell that is not corrupted as detected by the data analyzer.
Preferably, in step e) the guard time is timed starting f
Carder Norman
Johnstone Colin
Moore, Jr. Michael J.
Rao Seema S.
LandOfFree
ATM service disruption measurement does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with ATM service disruption measurement, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and ATM service disruption measurement will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3204107