Delay Predictability Routing for Low Power Lossy Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Implementing time-sensitive routing strategies in low power and lossy networks (LLNs) is challenging due to varying delays caused by environmental changes and limited resources, making it difficult to control packet delivery times effectively.
Innovation Solution
Distributing machine learning processes to network devices to predict delays along communication segments, which are then used as routing constraints to select paths that optimize delay predictability, rather than just minimizing delays.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If routing decisions minimize average delay in LLNs, then speed is improved, but reliability deteriorates due to high delay variability from environmental changes
Solution Approach 1:
The patent changes the routing parameter from minimizing average delay to optimizing delay predictability. By using statistical measures (variance, standard deviation) of delay rather than mean delay alone, the system selects routes with more predictable timing characteristics, improving reliability for time-sensitive applications in LLNs where environmental conditions cause high delay variability
Solution Approach 2:
The patent implements feedback mechanisms where nodes collect and analyze delay measurements from multiple packets over time. This statistical feedback about delay patterns enables dynamic route selection based on predictability metrics, allowing the network to adapt to changing environmental conditions and maintain reliable timing even in lossy network environments
2Measurement precision
If machine learning processes are distributed to network devices for delay prediction, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent segments the machine learning functionality into distributed components across multiple network nodes. Each node performs localized delay measurements and predictions for its adjacent links, rather than requiring centralized processing. This segmentation enables accurate delay prediction while keeping individual device complexity manageable through division of computational tasks
Solution Approach 2:
Network devices perform their own delay measurements and predictions using locally collected data, rather than relying on external controllers. Each node independently builds statistical models of its connected links' delay characteristics, enabling self-service delay prediction that improves measurement precision without requiring complex centralized infrastructure
Data Source
Figure 1
Figure 2
Figure 3A
AI summary
In one embodiment, a method is disclosed in which a device receives delay information for a communication segment in a network. The device determines a predictability measurement for delays along the segment using the received delay information. The predictability measurement is advertised to one or more devices in network and used as a routing constraint to select a routing path in the network.