Virtual Network Topology Control via Langevin Fluctuation Equation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing virtual network control methods struggle to adapt to unpredictable traffic demand fluctuations and network component failures without relying on network state information, particularly in large-scale communication networks with diverse applications and services.
Innovation Solution
A virtual network control method and device utilizing a fluctuation equation based on Langevin equations to dynamically adjust virtual network topologies in response to environmental changes, allowing for adaptive control without requiring cross-traffic information or network state data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If online control dynamically reconstructs virtual network based on periodical measurement, then the virtual network can adapt to traffic fluctuations, but the system requires precise traffic demand matrix which is difficult to acquire in large-scale networks
Solution Approach 1:
The virtual network control device autonomously adjusts the virtual network topology using its own internal fluctuation equation without requiring external traffic demand information. The system serves itself by using observable performance metrics (link loads, throughput) to trigger topology reconstruction, eliminating the need for complex traffic matrix measurements while maintaining adaptability to traffic fluctuations.
2Measurement precision
If conventional control methods use network state information for optimization, then the control precision is improved, but the device complexity and information requirements increase
Solution Approach 1:
The invention extracts only the essential observable metrics (link loads, throughput) from the complex network state, discarding the need for complete traffic demand matrix and detailed state information. By taking out only the necessary performance indicators, the system achieves control precision without the complexity of processing comprehensive network state data.
Solution Approach 2:
Instead of maintaining complex, long-term traffic demand matrices, the system uses simple, short-term observable metrics (current link loads, throughput) that can be easily measured and discarded after use. These lightweight information objects replace the heavy, persistent state information requirements of conventional methods.
3Adaptability or versatility
If the virtual network topology is frequently reconstructed to adapt to environmental changes, then the adaptability improves, but the network stability and convergence time are affected
Solution Approach 1:
The system employs periodic evaluation of network performance metrics to determine when topology reconstruction is necessary. Rather than continuous or frequent reconstruction, the fluctuation equation triggers topology changes only when performance degradation exceeds thresholds, providing periodic stabilization while maintaining adaptability to significant environmental changes.
Solution Approach 2:
The fluctuation equation incorporates hysteresis and threshold mechanisms that cushion against premature topology changes. By setting appropriate performance thresholds and evaluation periods, the system buffers against minor fluctuations that would otherwise trigger unnecessary reconstructions, thereby maintaining topology stability while preserving adaptability to genuine environmental changes.
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
A virtual network control method, for adaptively controlling a topology of a virtual network formed on a physical network in response to environmental changes in the virtual network, is provided with: a step of storing the Langevin equation, as a fluctuation equation, which models the dynamics of the topology of the virtual network as a variable for controlling the number of wavelength paths on the physical network; a step of designing control parameters included in the fluctuation equation; and a step of controlling the topology of the virtual network by applying the control parameters to the fluctuation equation to change an order parameter included in the fluctuation equation when environmental changes occur in the virtual network, and by transitioning the solution of the fluctuation equation between attractors determined by the deterministic term of the fluctuation equation due to the fluctuation term of the fluctuation equation.