Sequence Processing Method for Network Path Similarity with Alternative Routers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Matrix-based processing techniques, such as dynamic sequence alignment, are limited in evaluating the similarity between two network paths when multiple routers are used as alternatives to a single router, as they rely on a fixed structure and cannot accommodate combination alternative routers.
Innovation Solution
A method using a complex substitution matrix and score matrix calculations, where matrix elements represent the likelihood of routers being used interchangeably, allowing for the evaluation of similarity between network paths with combination alternative routers by constructing a trace-back matrix and calculating a similarity value based on optimal path alignment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a fixed structure matrix is used for processing sequences, then the processing method is simple and straightforward, but it cannot accommodate combination alternative routers when evaluating network path similarity
Solution Approach 1:
The patent transforms the fixed static matrix structure into a dynamic one where rows can be merged based on alternative router relationships. The matrix structure adapts during processing by combining rows that represent alternative routers, allowing the system to handle combination alternative routers while maintaining computational feasibility.
Solution Approach 2:
The patent creates a universal processing framework that can handle both simple alternative routers (one-to-one replacements) and combination alternative routers (one-to-many replacements) using the same matrix structure. The method unifies different types of router alternatives under a single processing paradigm, making the system versatile without requiring separate specialized structures.
2Reliability
If traditional dynamic sequence alignment is used, then the evaluation process is straightforward for simple alternatives, but it fails when multiple routers are used as alternatives to a single router
Solution Approach 1:
The patent segments the network path comparison problem into distinct phases: initial matrix construction with individual router rows, identification of alternative router relationships, and merging of corresponding rows. This segmentation allows the system to process complex combination alternatives by breaking them down into manageable steps while maintaining overall evaluation accuracy.
Solution Approach 2:
The patent merges rows in the matrix that represent alternative routers into a single combined row. When multiple routers are alternatives to a single router, their corresponding rows are merged to create a unified representation, enabling the dynamic sequence alignment algorithm to correctly evaluate paths with combination alternatives while preserving the reliability of similarity measurement.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
This invention relates to a method of processing a sequence in a computing system, and a device for implementing said method, the method comprising the steps of: a) receiving data of a plurality of historical sequences in a computing system, wherein the data for each historical sequence identifies each point in the historical sequence; b) analysing the historical data to identify a first and second alternative sub-sequence within a first and second historical sequence, wherein the first alternative sub-sequence includes a first common point identified in both the first and second historical sequence, a first intermediate point identified in the first historical sequence, and a second common point identified in both the first and second historical sequences, and the second alternative sub-sequence includes the first common point, a plurality of intermediate points identified in the second historical sequence, and the second common point; c) defining a virtual point as the plurality of intermediate points and a virtual sub-sequence as the first common point followed by the virtual point followed by the second common point; d) processing a first sequence in the computing system by: i) identifying the second alternative sub-sequence in the first sequence; and ii) defining a first variant sequence as the first sequence with the second alternative sub-sequence replaced with the virtual sub-sequence.