The invention relates to a method of determining a spectral route for a given connection in an optical telecommunications network, and it also relates to a node for implementing the method. It relates more particularly to wavelength division multiplex (WDM) optical networks that use a plurality of wavelengths to transmit a plurality of data streams simultaneously on the same optical fibre.
To set up a connection in this kind of network, it is necessary to determine not only a spatial route, consisting of a sequence of route segments connecting the starting node to the destination node, but also a spectral route, since each segment is able to support a plurality of wavelengths each constituting a spectral route segment. Selecting a spectral route entails selecting the wavelength to be used, or the wavelengths to be used successively, on different segments along the spatial route. It is sometimes necessary to carry out operations on the signal and/or the information transported, necessitating the addition of dedicated equipment to the network and the use thereof. However, such processing operations are costly, and it is therefore desirable to avoid them as much as possible. They may relate to regeneration and/or wavelength conversion, which may be carried out by purely optical means or using optical-electrical and electrical-optical conversion means.
The expression “wavelength continuity” refers to the use of the same wavelength from the starting node to the destination node, even if operations on the signal and/or the information transported require optical-electronic-optical conversion or 1R, 2R, or 3R regeneration.
The term “transparency” is used, a distinction being drawn between different types of transparency, depending on whether optical-electronic-optical conversion, wavelength conversion, 1R, 2R or 3R regeneration, or a combination of these operations are avoided. The aim is to avoid the route for the signal passing through “non-transparency points” or, if this is not possible, to minimize the number of non-transparency points the signal passes through. For example, in the case of a form of transparency consisting in the absence of optical-electronic-optical conversion, the aim is to minimize the number of times the optical signal passes through opto-electronic and electronic-optical interfaces. If it is not possible to avoid conversion completely, then a route is looked for that minimizes the number of conversions needed.
There may additionally be connection capacity constraints or quality-of-service constraints that influence the selection of the spatial route and the spectral route.
At present, there is no satisfactory method of determining a transparent route of the above kind. One method was proposed to the IETF in Generalized MPLS—Signaling Functional Description, chapters 3.4 and 3.5, Expiration date: November 2001, Network Working Group, Internet Draft.
That prior art method consists in:
The object of the above prior art method is only to assure wavelength continuity. The nodes do not propagate the sets of wavelengths because they do not need to do this to find a spectral route assuring wavelength continuity (should one exist end to end).
Each node of the route may retain or reduce the set of wavelengths that it inherits from the upstream node, according to the resources available for the connection to the downstream node. A transparent route is finally set up if the resulting set contains at least one wavelength. A drawback of that method is a high probability of blocking, since the selection made locally at each node may reduce options in downstream nodes. That method therefore constitutes a sub-optimum solution, or even no solution at all in some cases, when there is in fact an acceptable solution, although it is not transparent end to end. Also, the above prior art method takes account of only one parameter: the continuity of a given wavelength.
The object of the invention is to propose a method that does not have the above drawbacks.
The invention consists in a method of determining a spectral route in an optical telecommunications network between a starting node and a destination node of the network, the method being characterized in that it consists in:
The above method has the advantage of reducing the probability of blocking resulting from the impossibility of finding a route, since it provides, at any given time, knowledge of the parameter values for all the segments constituting one or more candidate routes (especially transparency parameter values). This complete knowledge is used to select more efficiently a transparent route or a route comprising a minimum number of non-transparency points. This overview leads to real optimization, i.e. avoiding possible solutions being abandoned during the process.
In a preferred embodiment, to collect parameter values characterizing all route segments along each candidate spatial route, a route set-up request message is sent from the starting node to the destination node and parameter values are collected in the message as it passes through each node along the candidate spatial route.
It should be observed that, in order to maximize the chances of finding a transparent route, this method may be applied several times simultaneously on separate spatial routes to satisfy the same connection set up request.
A preferred embodiment of the method of the invention is implemented in the node at the end of the route whose setting up has been requested, which is called the destination node.
The method of the invention uses the signaling means of the network to transmit transparency parameter values, providing up-to-date values for each new route set up request.
These parameter values do not relate only to wavelengths, and may relate to all other physical parameters of the connections between the nodes of the network.
In one embodiment, the parameters characterizing all the spectral route segments along each candidate spatial route take account of transparency constraints.
In one embodiment, the parameters characterizing all the spectral route segments along each candidate spatial route take account of connection capacity constraints.
In one embodiment, the parameters characterizing all the spectral route segments along each candidate spatial route take account of quality of service constraints.
The invention also provides an optical network node for implementing a method according to the invention, the node being characterized in that it comprises management means for:
The invention will be better understood and other features will become apparent in the light of the following description and the accompanying drawings, in which:
The network T shown by way of example in
ON1-ON2
ON1-ON3
ON1-ON4
ON2-ON6
ON3-ON5
ON3-ON6
In this example, the network T interconnects three client networks CNA, CNB, CNC which are connected to nodes ON1, ON4, ON6, respectively, at the edge of the network T. The method of the invention is executed in the network T and is totally independent of the number and nature of the client networks.
In one implementation of the method of the invention, a connection set-up request CSR is sent by the client network CNA to the management means of the node ON1 in order to set up a connection between the client networks CNA and CNC. The request contains the identity of the requesting client network CNA and the identity of the requested client network CNC and indicates constraints on transparency, capacity, quality of service, etc. The network T must determine a transparent route or, if this is not possible, a route comprising as few points of non-transparency as possible but still conforming to the capacity and quality of service constraints set for the connection. The optical transparency parameter constraints may be values of wavelength, spectral spacing, tolerance of non-linear effects (four-wave mixing, etc.), the obligatory absence of regeneration, etc.
Using a conventional routing method, the management means determine one or more spatial routes, referred to as candidate routes, connecting the client network CNA to the client network CNC, as a function of the topology and the connectivity of the network T. For example, they find two candidate spatial routes, Route 1 and Route 2, which are valid because they satisfy all the routing constraints referred to in the original connection set-up request CSR.
Route 1=ON1, ON4, ON5, ON6
Route 2=ON1, ON3, ON6
The management means of the node ON1 then forward the route set-up request RSR to the node ON6, and two copies RSR1 and RSR2 of the request are routed simultaneously on the two routes Route 1 and Route 2. The routing along these two routes is controlled by the management means of the starting node ON1 on furnishing the request to the signaling means of the network T. The copy RSR1 passes first through the node ON4 on the route Route 1 and the copy RSR2 passes through the node ON3 on the route Route 2.
Each node ON1, ON4, ONS through which the route set-up request RSR1 passes adds to the content of the request parameter values relating to the route segment immediately upstream and/or downstream of the node on the spatial route concerned, together with parameter values concerning the interfaces of the node, these values corresponding to parameters referred to in the request, in particular optical transparency parameters. Each node ON1, ON3 through which the route set-up request RSR2 passes performs the same action on the request RSR2. the link between ON3 and ON6. These parameter values may equally relate to the downstream connection and the upstream connection of the node. Finally, the two copies RSR1 and RSR2 of the route set-up request reach the destination node ON6.
In a first embodiment, the data collected in these two copies is processed by the management means of the destination node ON6 to determine an optimum combination of spectral route segments between the nodes ON1 and ON6 along each of the spatial routes taken by the connection requests. This data is processed using an optimization algorithm that minimizes a cost function taking into account all the parameter values collected. This algorithm may employ a shortest path algorithm such as the Dijkstra algorithm.
If at least one transparent route is possible, the optimization algorithm finds a transparent route. If no transparent route is possible, the optimization algorithm determines a route comprising the fewest possible points of non-transparency, i.e. an optimum combination of transparent sub-paths.
The destination node ON6 sends a route release message RR to all the other immediately adjacent nodes along spatial routes that have not been adopted, in this instance Route 1. In this example, the route release message RR is sent to the node ON5, which forwards it in the direction of the node that was the source of the route set-up request (node ON1).
If the algorithm concludes that no route is possible (even a non-transparent route), the destination node ON6 sends a route release message to all of the nodes on the routes Route 1 and Route 2, i.e. the nodes ON1, ON3, ON4, ON5.
In a second embodiment, the two copies RSR1 and RSR2 of the route set-up request (containing all the data collected) are not processed in the destination node ON6, but are returned from the destination node ON6 to the starting node ON1, or to a central unit somewhere in the network T, to be processed there. If the algorithm concludes that no route is possible (even a non-transparent route), the node ON1, respectively the central unit, sends a route release message to all of the nodes situated along the routes Route 1 and Route 2, i.e. the nodes ON1, ON3, ON4, ON5.
The first embodiment has the advantage that it avoids forwarding all of the data collected to the starting node or a central processing unit. This avoids occupying network resources for such forwarding.
Filing Document | Filing Date | Country | Kind | 371c Date |
---|---|---|---|---|
PCT/FR02/00652 | 2/21/2002 | WO | 00 | 10/26/2004 |
Publishing Document | Publishing Date | Country | Kind |
---|---|---|---|
WO03/079596 | 9/25/2003 | WO | A |
Number | Name | Date | Kind |
---|---|---|---|
5541756 | Chang-Hasnain et al. | Jul 1996 | A |
6205117 | Doshi et al. | Mar 2001 | B1 |
6584071 | Kodialam et al. | Jun 2003 | B1 |
6594268 | Aukia et al. | Jul 2003 | B1 |
6842463 | Drwiega et al. | Jan 2005 | B1 |
6982951 | Doverspike et al. | Jan 2006 | B2 |
7031299 | Chaudhuri et al. | Apr 2006 | B2 |
7039009 | Chaudhuri et al. | May 2006 | B2 |
7096176 | Hess | Aug 2006 | B1 |
7167443 | Dantu et al. | Jan 2007 | B1 |
7215640 | Matsubara | May 2007 | B2 |
7263100 | Hauser et al. | Aug 2007 | B2 |
20030198227 | Matsuura et al. | Oct 2003 | A1 |
Number | Date | Country |
---|---|---|
WO 9946951 | Sep 1999 | WO |
Number | Date | Country | |
---|---|---|---|
20050069316 A1 | Mar 2005 | US |