Mesh Network Provisioning via Graph Path Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network simulation techniques require exhaustive specification of states where performance is adversely affected, leading to over-engineering or failure to meet service levels due to errors in simulating rare failures and high service availability demands.
Innovation Solution
A computer-implemented method models network design as a graph data structure, iteratively adjusting for failures and repairs, and evaluates performance through path analysis to ensure the mesh network meets predetermined criteria, using stochastic generation of failure and repair times and least cost routing algorithms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If exhaustive specification of adversely affected network states is used, then measurement precision of service availability is improved, but device complexity and error risk increase
Solution Approach 1:
The patent segments the network into modular components (nodes, links, interfaces) that can be independently modeled and simulated. Each component has its own failure modes and repair processes, allowing the system to simulate network behavior without exhaustively listing all possible failure states. This modular approach maintains measurement precision while reducing the complexity of state specification.
Solution Approach 2:
The patent implements dynamic simulation where network states evolve over time through stochastic failure and repair processes. Instead of statically specifying all adverse states, the system dynamically generates states through time-dependent simulations of component failures and repairs. This allows accurate service availability measurement without requiring complete pre-specification of adverse states.
2Reliability
If exhaustive specification of failure states is performed, then reliability prediction accuracy is improved, but loss of time in simulation setup increases
Solution Approach 1:
The patent performs preliminary configuration of network components, their failure modes, and repair processes during network design. Failure rates, repair times, and component dependencies are pre-defined, allowing rapid simulation execution. This preliminary setup maintains reliable service level predictions while significantly reducing the time required for network provisioning compared to exhaustive state specification.
Solution Approach 2:
The patent uses parameter-based modeling where network reliability is determined by adjusting key parameters (failure rates, repair times, component capacities) rather than enumerating states. This allows flexible and rapid exploration of different network configurations and their reliability implications, maintaining accuracy while reducing provisioning time.
3Quantity of substance
If traditional simulation methods are used, then coverage of network states is improved, but manufacturing precision of network design decreases
Solution Approach 1:
The patent implements feedback mechanisms where simulation results directly inform network design adjustments. Service availability measurements from simulations feed back into the design process, allowing iterative refinement of network configuration. This feedback loop maintains manufacturing precision by using actual simulation outcomes rather than relying on exhaustive state coverage assumptions.
Solution Approach 2:
The patent replaces traditional mechanical enumeration of network states with computational simulation methods. Instead of physically listing and analyzing each possible state, the system uses algorithmic simulations to estimate service availability. This substitution maintains design accuracy while avoiding the errors inherent in manual state specification.
Data Source
AI summary
A method of provisioning mesh communication networks is disclosed. The method involves simulating the performance of a proposed network design to ensure the proposed network design meets service level criteria before provisioning a network in accordance with the proposed network design. Such simulations are required to be comprehensive because highly improbable events can be sufficient to result in a mesh network not meeting the stringent performance criteria which such networks are required to meet. Known methods of provisioning rely on exhaustively listing the mesh network states which would adversely impact the service offered by a proposed network design as part of simulating the performance of the proposed network design—this is an error prone exercise since relevant network states can be missed. A simulation technique is proposed in which the network state after each event is represented by a weighted graph indicating a measure of path cost for each of the links in the mesh network. A graph searching algorithm is applied to seek a path across the graph, thereby systematically exploring paths over mesh network which could provide a suitable route for the service in the simulated network state represented in the graph. Networks are thus provisioned which meet stringent performance criteria without being over-engineered.