Multi-Rail Spectrum Assignment for Parallel Link Routing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional routing and spectrum assignment (RSA) processes in multi-rail networks face challenges with exponentially increasing path numbers, leading to long processing times and underestimation of network capacity due to the inability to evaluate all routes, especially in large networks with parallel rails, resulting in suboptimal spectrum assignment and reduced network throughput.
Innovation Solution
A routing and spectrum assignment algorithm that utilizes edge-groups, such as parallel rails, to optimize spectrum assignment by considering all routes and minimizing fragmentation, eliminating the need for iterative path evaluations, and providing a graph transformation technique for efficient spectrum management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional RSA processes evaluate all possible routes in multi-rail networks, then network capacity estimation accuracy improves, but processing time increases exponentially
Solution Approach 1:
The patent segments the multi-rail network into individual rail segments between ROADM nodes. Instead of evaluating all possible end-to-end routes simultaneously, the algorithm processes each rail segment independently, determining spectrum availability and routing decisions at each segment level. This segmentation reduces the combinatorial explosion of route evaluations while maintaining accurate capacity estimation.
Solution Approach 2:
The algorithm performs preliminary spectrum availability assessment on each rail segment before conducting end-to-end route evaluation. By pre-computing which frequency slots are available on each segment, the system eliminates the need to re-evaluate spectrum constraints during subsequent route optimization, significantly reducing processing time while preserving accuracy.
2Productivity
If conventional RSA algorithms limit path evaluation to k-shortest paths, then processing time is reduced, but network capacity is underestimated
Solution Approach 1:
The patent introduces a new dimension of evaluation by considering spectrum availability across multiple frequency slots simultaneously, rather than evaluating routes sequentially. The algorithm uses multi-dimensional spectrum maps that capture availability across all rails and all frequency slots, enabling comprehensive capacity assessment without being constrained to a limited number of shortest paths.
Solution Approach 2:
The algorithm changes the evaluation parameters from route-based (k-shortest paths) to spectrum-slot-based assessment. Instead of asking 'which k routes should I evaluate?', the system asks 'which frequency slots are available on which rails?', fundamentally changing the problem formulation to avoid the k-shortest path limitation while maintaining computational efficiency.
3Adaptability or versatility
If parallel rails are deployed to increase network capacity, then spectrum flexibility improves, but the number of possible routes grows exponentially
Solution Approach 1:
The patent merges the evaluation of multiple parallel rails into a unified spectrum availability model. Instead of treating each rail as a separate entity requiring independent route evaluation, the algorithm combines spectrum information from all parallel rails into a single comprehensive view, allowing simultaneous consideration of all rails without exponential complexity growth.
Solution Approach 2:
The algorithm creates a universal routing framework that handles both single-rail and multi-rail scenarios using the same core logic. The spectrum availability model serves multiple functions: it tracks individual rail status, identifies parallel rail opportunities, and determines optimal spectrum assignment across all rails, eliminating the need for separate evaluation mechanisms for different network configurations.
Data Source
AI summary
Aspects of the subject disclosure may include, for example: obtaining first data indicative of first parallel links between a first pair of network nodes, the first data comprising a first indication of first available frequency slot(s) for a first one of the first parallel links and a second indication of second available frequency slot(s) for a second one of the first parallel links; obtaining second data indicative of a second link between a second pair of network nodes, the second data comprising a third indication of third available frequency slot(s) for the second link; selecting (as a selected link) the first one or the second one of the first parallel links, the selecting being in accordance with an availability in the selected link of a particular available frequency slot; and outputting a route including the selected link and the second link. Other embodiments are disclosed.