Data-Driven Network Port Delay Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Monitoring individual network ports in large communication networks is economically unfeasible due to the resource-intensive nature of physical sensors, making it difficult to detect and address degraded port performance causing packet loss or delay.
Innovation Solution
A data-driven estimation method that models network ports, links, and paths using vectors and matrices to calculate delay times, allowing for the identification of bad ports without physical sensors, utilizing existing latency information and quadratic programming to solve for individual port delays.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If physical sensors are deployed to monitor each network port, then measurement precision of port delay is improved, but device complexity and resource usage increase significantly
Solution Approach 1:
The patent creates a virtual model (copy) of the physical network that replicates port delay characteristics. Instead of deploying physical sensors to each port, the system uses software-based network modeling and path analysis to estimate port delays. The binary matrix A represents the network topology, and the vector x contains estimated port delays derived from path delay measurements, eliminating the need for physical sensing hardware at each port.
Solution Approach 2:
The patent replaces the mechanical/physical sensor-based measurement system with a data-driven computational approach. Instead of using physical sensors to directly measure port delays, the system uses quadratic programming to solve the equation Ax=b, where path delay measurements are processed mathematically to infer individual port delays. This substitution eliminates the need for physical sensing infrastructure while achieving the same measurement objective.
2Reliability
If physical sensors are installed at each network port, then reliability of port performance monitoring is improved, but loss of energy and resource consumption increase
Solution Approach 1:
The network system monitors its own performance using existing infrastructure and data. The patent leverages path delay measurements that are already being collected by the network for routing and performance management purposes. By processing this existing data through quadratic programming, the system achieves reliable port-level monitoring without requiring additional energy-consuming sensors or dedicated monitoring hardware at each port.
Solution Approach 2:
The patent makes existing path delay measurement infrastructure serve multiple functions. The same path delay data used for routing decisions and performance management is also utilized to estimate individual port delays through the mathematical model. This multi-functional use of existing data reduces the need for separate monitoring resources and energy consumption.
3Measurement precision
If individual port monitoring is implemented using traditional methods, then detection precision of degraded ports is improved, but ease of operation deteriorates due to large number of ports
Solution Approach 1:
The patent segments the monitoring problem into a mathematical framework where the network topology is represented as a binary matrix A, and port delays are represented as elements of vector x. This segmentation allows the system to handle large numbers of ports systematically through matrix operations and quadratic programming, rather than requiring manual monitoring of each individual port. The structured approach automates the detection process while maintaining precision.
Solution Approach 2:
The patent transforms the monitoring approach by changing from direct physical measurement parameters to mathematical estimation parameters. Instead of directly measuring each port with sensors, the system estimates port delays as parameters in a mathematical model that satisfies the path delay equations. This parameter transformation enables automated processing of large-scale networks while maintaining detection precision through the constraints of the mathematical model.
Data Source
AI summary
A computational method and system for estimating port delays in a network may use a data-driven estimation with quadratic programming based on available network path data that is already collected. In this manner, port delays for each individual port in the network may be estimated without having to measure each individual port using sensors.