Bandwidth Test Probe Rate Model Jitter Adaptation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Active network measurement methods, such as the Probe Rate Model (PRM), face challenges in achieving accurate bandwidth estimation while minimizing network resource consumption and reaction time to dynamic changes in available bandwidth, often leading to congestion and decreased service quality due to the fixed number of probe samples and lack of adaptation to network jitter.
Innovation Solution
The method involves using a variable number of probe sample packets based on the probe rate, calculated using the formula N(R)={50, if R<100 Mbps; 32.57·ln(0.046·R), if R≥100 Mbps}, and adapting the inter-packet time interval using a step decreasing factor SDF(JtI) to account for network jitter, optimizing the measurement process for reduced overhead and faster response to network changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If a fixed number of probe samples are used in bandwidth measurements, then the measurement process is simple to implement, but the accuracy of bandwidth estimation deteriorates when network conditions vary
Solution Approach 1:
The patent applies dynamics by making the number of probe samples variable rather than fixed. The system dynamically adjusts the number of probe samples based on network conditions, specifically adapting to network jitter characteristics. This allows the measurement system to maintain high accuracy across varying network conditions while preserving reasonable implementation complexity through automated adaptation.
2Measurement precision
If a large number of probe samples are injected into the network, then the accuracy of bandwidth measurement is improved, but network congestion and service quality deterioration occur
Solution Approach 1:
The patent applies parameter changes by dynamically adjusting the number of probe samples based on network jitter parameters. Instead of using a constant number of probe samples, the system modifies this parameter according to measured network conditions. This allows the system to use fewer probe samples when network jitter is low (reducing congestion risk) and more probe samples when network jitter is high (maintaining measurement accuracy).
Solution Approach 2:
The patent implements feedback by measuring network jitter and using this information to adaptively adjust the number of probe samples for subsequent measurements. The system continuously monitors network conditions and feeds this information back into the measurement process, creating a closed-loop system that optimizes probe sample injection to balance accuracy and network impact.
3Measurement precision
If the measurement duration is extended to capture more network variations, then the accuracy of available bandwidth estimation is improved, but the reaction time to dynamic network changes deteriorates
Solution Approach 1:
The patent applies dynamics by adapting the measurement duration and probe sample count based on real-time network jitter characteristics. When network conditions are stable (low jitter), the system uses shorter measurement durations with fewer probe samples, enabling faster reaction to changes. When network conditions are variable (high jitter), the system extends measurement duration and increases probe sample count to capture sufficient variation data, thereby maintaining accuracy while optimizing response time to dynamic changes.
Data Source
AI summary
A method for performing a bandwidth test for communications from a first network station to a second network station of a communication network is based on the probe rate model (PRM) in which iteratively a train of probe sample packets is transmitted from the first station to the second station with a constant packet rate per iteration. The packet rate is adapted to the available bandwidth on the data pipe per iteration. According to the disclosure the number of probe packets per train of probe sample packets is varied with the available bandwidth per iteration. This increases the accuracy of the test, in particular, in the range where a high data rate is available on the data pipe. A step decreasing factor (SDF) is applied with which the performance test is improved. Such step decreasing factor is dependent on the network jitter and further improves the performance of the test.