Spectral Defragmentation for Optical Network Capacity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Determining optimal routes in optical communication networks that satisfy demands is an NP-hard problem, leading to inefficient assignments and reduced network capacity as the network size or number of demands increases, resulting in unserved additional demands due to conflicts and resource inefficiencies.
Innovation Solution
A network management method that represents the communication network as a graph, updates paths to satisfy cost and network improvement conditions, and configures the network using updated paths to optimize resource usage and reduce Shannon entropy, thereby improving route assignments and network capacity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional routing methods are used to assign demands to routes in optical communication networks, then the network can handle basic communication demands, but the network capacity and resource utilization deteriorate as the network size or number of demands increases due to the NP-hard nature of the optimization problem
Solution Approach 1:
The patent segments the routing optimization problem by introducing a spectral dimension that divides the frequency spectrum into multiple slots. This segmentation transforms the complex NP-hard routing problem into a more manageable structure where demands can be assigned to specific spectral slots, enabling scalable network capacity without exponential complexity growth.
Solution Approach 2:
The patent adds a spectral dimension to the traditional routing problem by incorporating frequency slot assignments. This dimensionality change transforms the problem from purely spatial path selection to a combined space-frequency resource allocation problem, which resolves the complexity bottleneck by providing additional degrees of freedom for demand satisfaction.
2Productivity
If the network assigns more demands to existing routes, then network utilization increases, but resource conflicts and inefficiencies worsen leading to unserved demands
Solution Approach 1:
The patent changes the assignment parameters by introducing spectral slot indices in addition to spatial path selection. This parameter expansion allows multiple demands to be assigned to the same spatial route with different frequency slots, eliminating resource conflicts and enabling higher demand satisfaction rates without degrading assignment efficiency.
Solution Approach 2:
The patent creates virtual copies of network resources by allocating different spectral slots along the same physical path. This copying approach allows multiple demand flows to coexist on the same physical infrastructure without interference, effectively increasing the network's demand satisfaction capacity while maintaining reliable route assignments.
3Quantity of substance
If the network uses more frequency spectrum resources, then more demands can be served, but unused frequency spectrum portions increase when demands are not optimally assigned
Solution Approach 1:
The patent introduces dynamic spectral allocation where frequency slot assignments can be adjusted based on current network conditions and demand patterns. This dynamic approach allows the network to optimize spectral utilization in real-time, ensuring that frequency resources are actively used when needed and minimizing unused spectral portions through flexible reassignment.
Solution Approach 2:
The patent makes the frequency spectrum universally usable across multiple spatial paths by allowing the same spectral slots to be allocated on different physical routes for different demands. This multi-functionality of spectral resources eliminates waste by ensuring that each frequency slot serves a productive purpose, maximizing overall spectrum utilization without increasing unused portions.
Data Source
AI summary
An assignment of demands to routes on a communication network can be improved by selecting and updating a subset of the routes. A network management system (or other communication network component) can obtain a representation of the communication network. The representation can include a network graph, sets of resources, and a set of paths for the network graph. The network management system can update the set of paths by selecting a path subset and updating the paths of the path subset. The updating can satisfy a total change condition and a network improvement condition. The updating can be limited to unused edge and resource combinations or edge and resource combinations used by paths in the path subset. The network management system can configure the communications network to satisfy the demand using routes corresponding to the updated set of paths.