Multi-Scale Network Traffic Generation Using MMPP
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network traffic generation methods fail to accurately simulate self-similar network traffic patterns across multiple scales, which are crucial for testing network performance due to their bursty and long-range dependent nature, as they typically rely on Poisson distribution assumptions.
Innovation Solution
A multi-scale network traffic generation method using an n-state Markov modulated Poisson process (MMPP) model with transition windows to simulate self-similar traffic by determining state transitions and computing inter-packet times for each scale, enabling the generation of packets that accurately reflect varying traffic patterns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If Poisson distribution assumptions are used for traffic generation, then the modeling and analysis process is simplified, but the accuracy in simulating self-similar network traffic patterns deteriorates
Solution Approach 1:
The patent transitions from using a fixed Poisson distribution to an n-state Markov modulated Poisson process that dynamically changes parameters based on traffic state. The system defines multiple states (e.g., idle, low-rate, high-rate) and transitions between them using Markov chains, allowing the traffic generation model to adapt to self-similar patterns while maintaining computational feasibility through parameterization of transition probabilities and rate constants.
2Device complexity
If single-scale traffic generation is used, then the generation process is simple, but the ability to capture multi-scale burstiness and long-range dependence deteriorates
Solution Approach 1:
The patent segments the traffic generation process into multiple independent scales, each handled by its own Markov modulated Poisson process. Instead of attempting to model all temporal scales simultaneously, the system divides the time domain into different scales (e.g., short-term bursts, medium-term variations, long-term trends) and applies the MMPP model to each scale separately, then combines the results to generate realistic multi-scale traffic patterns.
3Reliability
If extensive equipment and human testers are used for network performance testing, then the testing comprehensiveness is improved, but the cost and time consumption increase
Solution Approach 1:
The patent creates virtual copies of real network traffic patterns using the Markov modulated Poisson process model. By capturing the essential statistical characteristics (burstiness, self-similarity, long-range dependence) through the MMPP model, the system generates synthetic traffic that replicates real-world behavior without requiring physical network deployments or human operators. This virtual replication enables comprehensive performance testing through software-based simulation rather than costly hardware-based testing.
Data Source
AI summary
Embodiments of the present invention provide a method, system and computer program product for multi-scale network traffic generation. In one embodiment of the invention, a network traffic generation method can be provided. The method can include defining multiple, different scales in an n-state MMPP model to accommodate a full characteristic response of a modeled traffic scenario. The method further can include establishing a transition window for each of the scales and determining a state through the transition window for selected ones of the scales. Finally, the method can include computing an inter-packet time according to the determined state for each of the selected ones of the scales and generating and transmitting packets for the selected ones of the scales utilizing a correspondingly computed inter-packet time.