LLN Topology Stabilization via Learning Machine Profiling

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Low Power and Lossy Networks (LLNs) face challenges such as lossy links, low bandwidth, and large node numbers, leading to inefficient routing, Quality of Service (QoS) management, and traffic engineering, where traditional approaches are inadequate due to the complexity and scale of these networks.

Innovation Solution

The implementation of learning machines that determine topological profiles of nodes and likelihoods of becoming a floating topology root, using a Topology Profiler and Moderator to proactively manage network connectivity and objective functions, reducing the occurrence of floating Directed Acyclic Graphs (DAGs) and stabilizing the network topology.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional routing approaches are used in LLNs, then implementation is simple, but network stability deteriorates due to floating DAGs and traffic disruptions

Engineering Contradiction:
Improvenetwork stabilityVSAvoidrouting management complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system performs preliminary actions by proactively identifying nodes susceptible to becoming floating DAG roots using topological profiling and likelihood calculations. Keep-alive messages are activated in advance for these identified nodes before actual topology failures occur, preventing floating DAG formation before it happens. This shifts the approach from reactive to preventive network stabilization.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system implements continuous feedback loops by monitoring network topology changes, updating topological profiles of nodes, and recalculating likelihoods of nodes becoming floating DAG roots. This feedback mechanism allows the system to adapt to dynamic network conditions and adjust keep-alive message activation accordingly, improving network stability through continuous optimization.

Inventive Principle:
Principle #23Feedback

2Reliability

If learning machines are deployed to predict floating DAGs, then network stability improves, but processing requirements and complexity increase

Engineering Contradiction:
Improvetopology stabilityVSAvoidprocessing energy consumption
Core Design Contradiction:
ReliabilityVSUse of energy by moving object

Solution Approach 1:

The system applies local quality by implementing learning machine functionality selectively only at network nodes identified as susceptible to becoming floating DAG roots, rather than uniformly across all nodes. The Topology Profiler and likelihood calculation mechanisms are activated locally at strategic points in the network, reducing overall processing energy consumption while maintaining effective prediction and prevention capabilities.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The learning machine system performs self-service by automatically profiling node topologies, calculating likelihoods, and activating keep-alive messages without requiring external intervention or manual configuration. The system autonomously adapts to network changes and manages its own operation, reducing the need for external control infrastructure and associated energy overhead.

Inventive Principle:
Principle #25Self-service

3Reliability

If keep-alive messages are activated for all nodes, then floating DAG detection is comprehensive, but network traffic overhead increases

Engineering Contradiction:
Improvefloating DAG detection accuracyVSAvoidnetwork traffic volume
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

The system applies partial action by activating keep-alive messages only for the subset of nodes identified as susceptible to becoming floating DAG roots, rather than for all network nodes. The Topology Profiler and likelihood calculation mechanisms identify a targeted subset of high-risk nodes, and keep-alive monitoring is applied selectively to these nodes, reducing network traffic overhead while maintaining effective detection capability for the most critical cases.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP3090595B1Reducing floating directed acyclic graphs (DAGS) and stabilizing topology in low power and lossy networks (LLNS) using learning machines
Publication Date: 2020.02.05 CISCO TECHNOLOGY INC
  • EP3090595B1 patent drawingFigure 1
  • EP3090595B1 patent drawingFigure 2
  • EP3090595B1 patent drawingFigure 3

AI summary

In one embodiment, a device determines a topological profile of individual nodes in a shared-media communication network, and also determines a respective likelihood of the nodes in the network to become a root of a floating topology based on the topological profiles. Accordingly, the device may provide instructions to particular nodes in the network based on the respective likelihoods.