Optical communication network configuration

1. A method for configuring an optical communication network, comprising:

obtaining a network graph representing the optical communication network, the network graph including:

network graph edges corresponding to communications links in the optical communication network, each of the network graph edges being associated with an available subset of a set of frequency slots for the optical communication network, the set of frequency slots comprising overlapping frequency slots; and

network graph vertices corresponding to nodes of the optical communication network;

obtaining a demand on the optical communication network, the demand having a corresponding source network graph vertex and terminal network graph vertex on the network graph;

constructing layer graphs using the network graph, each of the layer graphs corresponding to one of the frequency slots and including:

layer edges corresponding to the network graph edges that include, in the available subsets for the network graph edges, the one of the frequency slots;

layer vertices corresponding to the network graph vertices connected by the network graph edges that include, in the available subsets for the network graph edges, the one of the frequency slots; and

wherein the layer vertices include a source layer vertex corresponding to the source network graph vertex and a terminal layer vertex corresponding to the terminal network graph vertex;

selecting a path over a first layer of the layer graphs that connects the source layer vertex of the first layer to the terminal layer vertex of the first layer, the selection based on a cost associated with the selected path; and

configuring the optical communication network to satisfy the demand using the selected path over the first layer.

2. The method of claim 1 , wherein:

the method further comprises:

determining layer edge costs, the determination comprising:

identifying affected layer edges for a first layer edge; and

determining a first layer edge cost for the first layer edge based on the layer edges affected by the first layer edge; and

determining the cost associated with the selected path using the layer edge costs.

3. The method of claim 2 , wherein:

determining the first layer edge cost compromises:

determining the first layer edge cost based on scores for the layer edges affected by the first layer edge.

4. The method of claim 3 , wherein:

the scores are equal; or

a metric value for one of the layer edges depends on, at least in part, a topology of the network graph or of a one of the layer graphs including the one of the layer edges.

5. The method of claim 3 , wherein:

a metric value for one of the layer edges depends on, at least in part,

a one of the layer graphs including the one of the layer edges; or

the metric value for one of the layer edges depends on, at least in part, a central frequency or a frequency slot width associated with the one of the layer graphs including the one of the layer edges.

6. The method of claim 1 , wherein:

selecting the path over the first layer of the layer graphs comprises:

constructing a combined layer graph by joining the source layer vertices or joining the terminal layer vertices; and

searching the combined layer graph to identify at least one path from one of the source layer vertices to one of the terminal layer vertices.

7. The method of claim 1 , wherein:

The method includes selecting a transceiver type for the demand, the transceiver type specifying a frequency slot width; and

the first layer is further selected based on the specified frequency slot width.

8. The method of claim 1 , wherein:

each frequency slot corresponds to a center frequency and a frequency slot width.

9. The method of claim 1 , further comprising:

updating the network graph or the layer graphs based on the selection of the path.

10. A method for configuring an optical communication network, comprising:

obtaining layer graphs, each of the layer graphs corresponding to one of a set of frequency slots on the optical communication network and including:

layer vertices corresponding to a subset of network graph vertices, the network graph vertices corresponding to nodes of the optical communication network, the layer vertices including a source layer vertex and a terminal layer vertex, the source layer vertex corresponding to a source node of a demand on the optical communication network and the terminal layer vertex corresponding to a terminal node of the demand on the optical communication network;

layer edges corresponding to a subset of network graph edges, the network graph edges corresponding to communication links in the optical communication network, each layer edge connecting two of the layer vertices; and

wherein layer edge costs are associated with the layer edges, a first layer edge cost for a first layer edge based on the layer graphs or layer edges affected by the first layer edge;

determining a path from the source layer vertex to the terminal layer vertex based on the layer edge costs; and

configuring, based on a determined cost of the path, the optical communication network to use the path to satisfy the demand.

11. The method of claim 10 , wherein:

obtaining the layer graphs comprises constructing the layer graphs using a network graph of the optical communication network.

12. The method of claim 11 , wherein:

the network graph includes:

the network graph edges;

the network graph vertices; and

wherein each of the network graph edges is associated with an available subset of the set of frequency slots;

a first layer graph of the layer graphs corresponds to a first frequency slot in the set of frequency slots; and

constructing the layer graphs comprises, for the first layer graph:

identifying edges in the network graph that include, in the available subsets, the first frequency slot; and

identifying vertices in the network graph vertices connected by the network graph that include, in the available subsets, the first frequency slot.

13. The method of claim 10 , wherein:

obtaining the layer graphs comprises determining the layer edge costs, determining the first layer edge cost for the first layer edge comprising:

identifying ones of the layer graphs or ones of the layer edges affected by the first layer edge; and

determining the first layer edge cost based on the identified layer graphs or layer edges.

14. The method of claim 13 , wherein:

determining the first layer edge cost comprises identifying the ones of the layer graphs affected by the first layer edge; and

the ones of the layer graphs affected by the first layer edge are identified based on:

a first frequency slot of a one of the layer graphs containing the first layer edge; and

second frequency slots of ones of the layer graphs containing the ones of the layer edges affected by the first layer edge.

15. The method of claim 14 , wherein:

the layer edges affected by the first layer edge are identified based on the first frequency slot overlapping with the second frequency slots.

16. The method of claim 13 , wherein:

determining the first layer edge cost for the first layer edge comprises determining the first layer edge cost based on the identified layer edges; and

the first layer edge cost is determined based on scores of the identified layer edges.

17. The method of claim 16 , wherein:

the scores are equal; or

a metric value for one of the identified layer edges depends on, at least in part, a topology of a network graph of the optical communication network or of the layer graph including the one of the identified layer edges.

18. The method of claim 16 , wherein:

a metric value for one of the identified layer edges depends on, at least in part, the layer graph including the one of the identified layer edges; or

the metric value for one of the identified layer edges depends on, at least in part, a central frequency or a frequency slot width associated with the layer graph including the one of the identified layer edges.

19. A method for configuring an optical communication network comprising:

obtaining a network graph representing the optical communication network, the network graph including:

network graph edges corresponding to communications links in the optical communication network;

network graph vertices corresponding to nodes of the optical communication network;

wherein a demand on the optical communication network has a corresponding source network graph vertex and terminal network graph vertex on the network graph, and is associated with a frequency slot width of two; and

wherein:

the network graph comprises k subnetworks, each subnetwork comprising two of the graph vertices, the two of the graph vertices connected by three of the network graph edges;

for each 2≤j<k, one of the two of the graph vertices of subnetwork j is connected to subnetwork j−1 through an intermediate subnetwork j−1 and the other of the two of the graph vertices of subnetwork j is connected to subnetwork j+1 through an intermediate subnetwork j, the one of the two of the graph vertices of subnetwork 1 not connected through intermediate subnetwork 1 to subnetwork 2 is the source network graph vertex, and the one of the two of the graph vertices of subnetwork k not connected through intermediate subnetwork k−1 to subnetwork k−1 is the terminal network graph vertex;

a set of frequency bands {c 1 , c 2 , . . . , c n } is associated with the optical communication network and a set of overlapping frequency slots comprises consecutive pairs of these frequency bands; and

one of the three of the network graph edges e j in each subnetwork j has an available subset of the overlapping frequency slots that excludes the frequency slots {{c 4 , c 5 }, {c 5 , c 6 }, {c 6 , c 7 }, {c 8 , c 9 }, {c 9 , c 10 }, {c 10 , c 11 }} and for each of the remaining network graph edges, an available subset of the overlapping frequency slots includes the entire set of overlapping frequency slots; and

determining a frequency slot and a path through the network graph, the frequency slot being {c 7 , c 8 } and the path includes edges e 1 to e k .