Concave Hull Bandwidth Estimation for Network Traffic
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for estimating the bandwidth required to maintain quality of service in data networks are inefficient and require complex statistical modeling, making them impractical for real-time implementation and dynamic traffic conditions.
Innovation Solution
A method that uses a concave hull function to analyze traffic patterns, allowing for accurate and computationally efficient estimation of bandwidth requirements without assuming statistical properties of the traffic flow, and dynamically adjusts service rates to prevent packet loss or delay.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If complex statistical modeling methods are used to estimate bandwidth requirements, then measurement precision is improved, but device complexity and computational requirements increase significantly
Solution Approach 1:
The patent extracts only the essential statistical characteristics (mean and variance) from the complex traffic patterns, discarding unnecessary higher-order moments and complex distribution assumptions. This allows accurate bandwidth estimation using simple second-order statistics rather than complex full-distribution modeling.
Solution Approach 2:
The patent transforms the bandwidth estimation problem from a complex statistical inference problem into a simpler parameter calculation problem. By changing the approach from modeling entire traffic distributions to calculating bandwidth based on mean and variance parameters, the complexity is significantly reduced while maintaining estimation accuracy.
2Reliability
If conservative bandwidth allocation is used to eliminate packet loss, then reliability is improved, but productivity and network efficiency deteriorate
Solution Approach 1:
The patent applies partial action by providing bandwidth that is sufficient for the actual traffic requirements rather than conservative over-provisioning. The effective bandwidth calculation provides just enough capacity to handle the traffic with acceptable packet loss probability, avoiding excessive bandwidth allocation.
Solution Approach 2:
The patent uses feedback from actual traffic measurements (mean and variance estimation) to dynamically adjust bandwidth allocation. This feedback mechanism allows the system to adapt to actual traffic conditions and provide appropriate bandwidth without conservative over-provisioning, optimizing both reliability and efficiency.
3Productivity
If real-time bandwidth estimation is implemented, then productivity is improved, but computational overhead and processing requirements increase
Solution Approach 1:
The patent extracts only the necessary computational operations for real-time bandwidth estimation, removing complex statistical calculations. By focusing on calculating mean and variance from traffic samples and using these simple parameters for bandwidth determination, the computational overhead is minimized while enabling real-time operation.
Solution Approach 2:
The patent uses simple, lightweight statistical parameters (mean and variance) that can be computed quickly from traffic samples, replacing expensive complex modeling operations. These simple parameters serve as disposable approximations that enable real-time calculations without significant computational burden.
Data Source
AI summary
Embodiments of the invention relates to bandwidth requirements within a packet data network. A method and system for analysing traffic is described herein. The method may utilises an estimation of a concave hull function of the arrived traffic within the buffer. The use of such a concave hull representation may allows for more efficient data processing and for a direct measurement of desired service rates for differing predetermined control parameters.