Patent Application
20060230180
The invention is based on a priority application EP 05290767.2 which is hereby incorporated by reference.
The invention relates to a method for optimization of dispersion compensation in an optical data network, the network comprising two nodes (A, B), at least two paths connecting the two nodes, and means for switching between the paths.
Such a method is described in U.S. Pat. No. 6,674,935.
Optical data networks, such as 10 or 40 Gbit/s networks, are used to transfer data at high speed over large distances in the order of hundreds of kilometers. Such a network comprises nodes, i.e. devices for accessing the network in order to read out data and/or feed data into the network. These nodes are linked to each other by data transfer lines. The connections between the nodes are called paths. Typically, each node is connected to several neighboring nodes.
However, problems may occur when a path in the network fails. Such a failure may be due to a planned deactivation of the path, e.g. in order to modify the network, or it may be involuntarily due to an unpredicted line break caused by a material defect. The data traffic of the network cannot use the failed path any more, endangering the data traffic as a whole.
One way to protect a network against path failure is the 1+1 protection scheme. In this scheme, a link between two nodes comprises two (or more) complete, independent and parallel paths, including two transponders at the source node and two receivers at the receiver node. Both paths are used continuously and simultaneously for the same data. In case of a failure of one of the paths, the data connection is still working without cessation by means of the other path. The 1+1 protection scheme is disadvantageous in that it requires double equipment and causes lots of redundant data traffic.
Another way to protect a network is protection switching, as mentioned in U.S. Pat. No. 6,674,935. In case of a path failure, the traffic is routed around the failed path. For example, if traffic is regularly sent from a node A to a node B via a direct path AB, and said direct path AB fails, then the traffic is forced via a node C connected to both nodes A and B. The traffic then takes an alternate path a to C to B (=ACB).
The known protection switching works well for 10 Gbit/s systems, since chromatic dispersion variation between a regular path and an alternate path is typically small enough to be below an acceptance window of a receiver at the receiving node. However, in 40 Gbit/s systems, the tolerance to residual dispersion is nearly 16 times narrower than for 10 Gbit/s systems. Thus the chromatic dispersion variation between the regular path and the alternate path is typically too large for a receiver at the receiving node.
U.S. Pat. No. 6,674,935 describes an optical connection arrangement for facilitating network modifications during operation. An optical signal input is connected to an optical switch, and an optical signal output is connected to another optical switch. Between the optical switches, two optical ports are located. In a first switching position, a link between the signal input and the signal output is established through the first optical port and a first optical fiber, and in a second switching position, the link is established through the second optical port and a second optical fiber. More optical ports can be selectively included. At the optical ports, dispersion compensation modules can be mounted. The optical connection arrangement minimizes traffic interruption upon switching between the links.
So in U.S. Pat. No. 6,674,935, separate dispersion compensation modules for a regular and an alternate path are used. This, again, requires expensive equipment multiple times.
It is therefore the object of the invention to provide a method for optimizing dispersion compensation in a 40 Gbit/s (or higher) system upon switching of paths, which requires only a simply designed optical data network, and at the same time allows a fast re-establishing of data transfer upon switching of paths.
This object is achieved, according to the invention, by a method as mentioned in the beginning, characterized in that that the network comprises a common tuneable dispersion compensator (=TDC) for all of the at least two paths, that prior to use of the network, for each path, a set of optimized operating parameters of the network is determined and stored, wherein the operating parameters of the network include operating parameters of the TDC, and that in use of the network, upon switching to a path, the network is automatically adjusted to a set of initial operating parameters based upon the set of predetermined operating parameters of the respective path.
The TDC is capable of compensating dispersion of every path, sufficiently to allow the operation of a receiver at a receiver node in a 40 Gbit/s system. Thus, all paths can be operated with one common TDC, keeping the network design simple.
In principle, the TDC—and the network as a whole—could be adjusted to a path switched to by standard methods such as feedback from bit error rate (=BER) measured by forward error correction (=FEC) after switching. However, the required time for such an adjustment is rather large, well above 10 seconds. This would constitute a minimum of time required for re-establishing the data transfer upon switching (=restoration time), considered too long for many purposes.
In order to make the restoration time shorter, for each path, a set of optimized operating parameters of the network is determined and stored. When switching to a new path, a set of initial operating parameters is derived from this predetermined set of operating parameters for the new path, and the network is adjusted to this initial set of operating parameters unhesitatingly. A set of operating parameters is considered optimised for a path if it allows the derivation of an initial set of operating parameters which makes the network immediately workable with said path. As a result, no measurements of dispersion characteristics of the path switched to are required for re-establishing the data transfer within the network, making the restoration time particularly short.
A set of operating parameters comprises information about the chromatic dispersion characteristics of the respective path. By suitable classification in categories, a set of operating parameters may be reduced to a single variable, in accordance with the invention.
By determining and storing a set of operating parameters for each path, a large number of alternative paths can be made workable within the network.
A highly preferred variant of the inventive method is characterized in that in case of a link break in one path, the network switches automatically to another path in good order. The short restoration time makes the inventive method particularly suited for this path protection by protection switching.
In another variant of the inventive method, the set of initial operating parameters is identical with the set of predetermined operating parameters of the respective path. In this case, no derivation operation, i.e. no calculation, is necessary to find the initial operating parameters, keeping the variant simple and quick to perform.
An alternative and preferred variant of the inventive method is characterized in that upon switching from a first path to a second path, a network variation is determined by comparing the last set of operating parameters of the first path with the set of predetermined operating parameters of the first path, and the set of initial operating parameters of the second path is derived from the set of predetermined operating parameters of the second path taking into account the determined network variation. The characteristics of the network may be subject to significant changes during operation, such as a seasonal change of temperature. Then the operating parameters of the network need to be adapted to these changes for optimal network performance. The network variation indicates the current network characteristics. In case of a switching of paths, taking into account the network variation allows the derivation of more accurate initial operating parameters for the path to switch to, i.e. the network performance immediately after switching is improved. The derivation may include a weighting of the network variations with path lengths.
In a preferred variant of the inventive method, the operating parameters of the network include a receiver threshold and/or a receiver phase. By this means, network performance immediately after switching can be improved.
The invention comes to show to advantage in a variant wherein the optical network has a data transfer rate of 40 Gbit/s or more. The variant can provide chromatic dispersion compensation with short restoration times even at these rates of data transfer, with just one TDC.
Further preferred is a variant characterized in that the TDC is adapted to the set of initial operating parameters by means of a software signal. This adaptation is particularly quick and easy to realize.
In another advantageous variant of the inventive method, after having adjusted the network to the set of initial operating parameters upon a switching of paths, an optimization process for the operating parameters is launched. Although the initial operating conditions make the network immediately workable, they may differ from the best possible operation parameters, e.g. because of unbalanced degradation of data transfer lines over time. By means of the variant, better operating conditions can be found and used, increasing the network performance. Said optimisation process may include FEC measurements of BER.
The scope of the invention also comprises a transponder for an optical data network, comprising at least two ports for providing connections to at least two respective paths of the network, further comprising means for switching between the ports, characterized in that the transponder comprises a common tuneable dispersion compensator (=TDC) for all of the at least two ports, further comprising storage means, storing a set of predetermined optimized operating parameters of the transponder for each port, wherein the operating parameters of the transponder include operating parameters of the TDC, and means for automatically adjusting the transponder, upon switching to a port, to a set of initial operating parameters based upon the set of operating parameters predetermined for the respective port. Such an inventive transponder can be used when performing the inventive methods described above.
Furthermore, the scope of the invention also comprises the use of an inventive transponder as described above in an optical network with path protection.
Further advantages can be extracted from the description and the enclosed drawing. The features mentioned above and below can be used in accordance with the invention either individually or collectively in any combination. The embodiments mentioned are not to be understood as exhaustive enumeration but rather have exemplary character for the description of the invention.
The invention is shown in the drawing.
The nodes A, B comprise optical switches 2, 3 for choosing a path for data to be transferred from node A to node B. Regularly, the direct path AB is used, being the shortest connection of node A and node B. In the state shown in
Node B comprises a tunable dispersion compensator (=TDC) 4, which is connected to the right end of the switch 3. Thus, the TDC 4 is used in any position of the switch 3, i.e. with any path data traffic can be transferred from. The transferred traffic is, thanks to the TDC 4, within an acceptance window of a receiver (not shown) at the receiving node B. Node B basically constitutes an inventive transponder in the sense of the invention, with the left connections of the switch 3 being its ports.
A set of optimized operating parameters for both paths AB and A to C to B (=ACB) were determined upon installation of the network 1. For this reason, it is known that path AB has an accumulated dispersion of about +100 ps/nm, and path ACB has an accumulated dispersion of about +50 ps/nm. Moreover, an optimized receiver threshold and phase were determined and recorded for each path AB and ACB upon installation.
When path AB is in use, as shown in
During the life of the network 1, a link break 5 in the direct path AB linking nodes A and B can occur. The network 1 in such a damaged state is shown in
In order to get better network performance, in particular a lower BER, the operating parameters of the network 1 can be optimized during the use of the network 1. Typical optimization algorithms include FEC measurements of BER, for example. A change of the best operating conditions can occur due to temperature variation. Then, when switching between paths, the current network variation should be taken into account.
As an example, immediately before the link break 5 was detected, the TDC 4 compensated for an accumulated dispersion of only +80 ps/nm in path AB, instead of the +100 ps/nm. This means the network 1 had a network variation of
(80 ps/nm−100 ps/nm)=−20 ps/nm
with respect to accumulated dispersion of path AB. Therefore, the initial operating parameter for the TDC 4 for path ACB should not be the predetermined operating parameter of +50 ps/nm, but be adapted accordingly. Assuming that the network variation is merely an adding number, an initial operating parameter for path ABC results as
(50 ps/nm−20 ps/nm)=30 ps/nm.
As part of the installation procedure, a correlation function between network variations in various paths can be determined. When the network variation of the path switched away from is known, the correlation function can be used for calculating the initial operating parameters from the predetermined optimized operating parameters of the path to switch to.
Of course, the initial operating parameters may be a starting point for further optimization of the operating parameters of the path switched to, e.g. using FEC measurement of BER.
| Number | Date | Country | Kind |
|---|---|---|---|
| 05290767.2 | Apr 2005 | EP | regional |
