Cloud Network Evaluation Service for Real-Time Traffic Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current network testing systems lack flexibility and accuracy, particularly in evaluating network conditions and predicting application performance across various devices and protocols, which is crucial for optimizing network resources and detecting negative behavior in modern broadband cellular and wireless networks.
Innovation Solution
A cross-platform network evaluation service that collects and intelligently analyzes live network traffic, using machine learning and AI to predict application performance, identify negative behavior, and optimize network usage by distributing workload based on network load, filtering raw data, and varying time resolution for seamless analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional network testing systems are used, then device complexity is reduced, but measurement precision and reliability of network condition evaluation deteriorate
Solution Approach 1:
The patent introduces a cloud-based network evaluation service as an intermediary between network devices and analysis tools. This service collects raw network traffic data from multiple devices, performs centralized intelligent analysis using machine learning models, and returns processed results. This mediator approach enables high-precision network condition evaluation without requiring complex testing systems at each device, as the analytical complexity is offloaded to the cloud service.
Solution Approach 2:
The patent replaces traditional mechanical/network testing hardware and manual analysis methods with AI-based machine learning models. These models automatically analyze network traffic patterns, predict application performance, and identify negative behaviors without requiring complex physical testing equipment. The substitution of AI algorithms for traditional testing mechanisms achieves high measurement precision while reducing device complexity.
2Measurement precision
If comprehensive network traffic data is collected and stored for analysis, then measurement precision improves, but loss of substance (storage resources) increases
Solution Approach 1:
The patent extracts only the essential and relevant features from raw network traffic data for storage and analysis. Instead of storing complete raw traffic captures, the system extracts key parameters such as traffic patterns, application performance metrics, and anomaly indicators. This extraction approach maintains measurement precision by preserving critical analysis information while dramatically reducing storage requirements by eliminating redundant raw data.
Solution Approach 2:
The patent implements selective data collection and processing, focusing on partial but critical aspects of network traffic rather than analyzing every single packet. The system collects and stores only the portion of data necessary for accurate network condition evaluation, using intelligent sampling and filtering techniques. This partial action approach achieves sufficient measurement precision without the excessive storage burden of complete data retention.
3Productivity
If real-time network analysis is performed on all devices, then productivity improves, but use of energy increases
Solution Approach 1:
The patent uses a cloud-based evaluation service as a mediator that performs the computationally intensive real-time analysis tasks. Individual network devices only need to collect local traffic data and send it to the cloud service, avoiding the need for high-power local processing. This intermediary approach enables real-time network-wide analysis productivity while keeping energy consumption at individual devices minimal, as the heavy computational workload is offloaded to the cloud infrastructure.
4Adaptability or versatility
If distributed system architecture is used for network evaluation, then adaptability improves, but device complexity increases
Solution Approach 1:
The patent implements a universal cloud-based evaluation service that can handle multiple network protocols, device types, and traffic patterns through a single standardized interface. The service is designed to be protocol-agnostic and can adapt to different network conditions without requiring device-specific implementations. This universality approach improves adaptability across diverse networks while reducing device complexity, as individual devices don't need to implement complex protocol-specific analysis logic.
Data Source
AI summary
A method, a device, and a non-transitory storage medium provide a network evaluation service. The service collects live network traffic data for a client device in a network; stores a benchmark pattern model; determines a category of the live network traffic data based on a segment size; detects a first traffic pattern of the live network traffic data based on measured segment parameters for the category; matches the first traffic pattern to a second traffic pattern in the benchmark pattern model to identify a result; compares the live network traffic data with a benchmark application pattern from the benchmark pattern model; and identifies, based on the comparing, a level of degraded performance in the network.