Industrial Network Orchestration Using Centrality-Based Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Industrial communication systems face scalability issues and complex management due to the joint planning of contention and schedule-based medium access techniques, leading to increased data collection and actuator command delays, as well as inflexibility in network topology changes.
Innovation Solution
A method involving centrality-based node clustering, where nodes with similar centralities are grouped into clusters, with a cluster head responsible for resource allocation, and centrality metrics are calculated to optimize network orchestration, reducing packet collisions and latency by coordinating transmissions and allocating resources effectively.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If contention based MAC strategy is applied to wireless networks in ICS, then flexibility to topology change is improved, but additional interference and packet collision increase
Solution Approach 1:
The patent segments the network into clusters based on node centrality metrics. Each cluster is managed independently with its own cluster head, allowing the system to maintain flexibility in topology changes while reducing interference within clusters. This segmentation resolves the contradiction by localizing collisions to specific clusters rather than the entire network.
Solution Approach 2:
The patent introduces cluster heads as intermediary nodes that coordinate communication within clusters. These intermediaries manage packet transmission and reduce direct collisions between end nodes, thereby maintaining the flexibility of contention-based access while reducing overall network interference through localized coordination.
2Object-generated harmful factors
If schedule based MAC strategy (TDMA/FDMA) is applied to wired networks in ICS, then packet collisions are minimized, but flexibility to topology change deteriorates
Solution Approach 1:
The patent implements dynamic cluster formation and reconfiguration based on real-time centrality metrics. When topology changes occur, clusters are dynamically adjusted and re-planned only for affected areas rather than the entire network. This dynamic approach maintains the collision-minimizing benefits of scheduled access while restoring flexibility through adaptive reconfiguration.
Solution Approach 2:
The patent applies scheduled MAC strategies locally within clusters rather than globally across the entire network. This localized application of TDMA/FDMA reduces packet collisions within clusters while limiting the impact of topology changes to only the affected cluster, thereby maintaining overall network flexibility through selective re-planning.
3Productivity
If joint planning of contention and schedule based medium access techniques is performed, then resource allocation is optimized, but system complexity and management difficulty increase
Solution Approach 1:
The patent segments the complex joint planning problem into smaller, independent cluster-level planning problems. Each cluster head performs localized resource allocation based on cluster-specific centrality metrics, rather than requiring global orchestration. This segmentation dramatically reduces computation complexity while maintaining resource allocation efficiency within each cluster.
Solution Approach 2:
The patent enables cluster heads to autonomously perform resource allocation and coordination within their respective clusters using local centrality information. This self-service approach eliminates the need for complex centralized orchestration of the entire network, reducing management difficulty while still achieving optimized resource allocation through distributed decision-making.
4Productivity
If centralized orchestration is used to manage ICS, then global resource allocation is optimized, but scalability and response time to topology changes deteriorate
Solution Approach 1:
The patent divides the centralized orchestration function into distributed cluster-level orchestration. Each cluster head independently manages its cluster based on local centrality metrics, eliminating the need for global re-planning when topology changes occur. This segmentation improves scalability by allowing clusters to be added or removed without reconfiguring the entire network, while maintaining efficient resource allocation locally.
Data Source
AI summary
A method for providing network orchestration for industrial communication system with multiple network slices, comprising: —calculating centralities of nodes in the industrial communication system; —grouping nodes with similar centralities of nodes into clusters; —selecting cluster head for each cluster that is responsible for the resources allocation for the members of the cluster; and—calculating centrality metric for cluster centralities, so as to orchestrate the industrial communication system.