Application Type Classification via Bearer Metrics and Label Maps

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional methods for identifying application types in mobile networks rely on deep packet inspection (DPI), which is costly and ineffective for encrypted data flows and requires significant hardware upgrades, limiting the ability to differentiate and optimize Quality of Service (QoS) for various applications.

Innovation Solution

A method that classifies application types by computing statistics vectors from bearer metrics and using label maps to identify application types without the need for DPI, enabling efficient resource allocation and improved Quality of Experience (QoE) by distinguishing between different application contexts.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If deep packet inspection (DPI) is used to identify application types, then application type identification accuracy is improved, but hardware cost and system complexity increase significantly

Engineering Contradiction:

Improveapplication type identification accuracyVSAvoidhardware cost and system complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts only the essential statistical features (buffer size, resource allocation, time patterns) from bearer data flows, rather than performing full deep packet inspection. This extraction approach achieves sufficient application type identification accuracy while avoiding the hardware complexity and cost of DPI systems.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent replaces the mechanical DPI inspection system with a statistical analysis approach using machine learning classifiers. Instead of mechanically examining packet contents, the system uses computational algorithms to analyze statistical patterns in bearer metrics, reducing hardware requirements while maintaining identification accuracy.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Adaptability or versatility

If deep packet inspection (DPI) is used to identify application types, then application type identification capability is improved, but implementation cost and hardware upgrades increase

Engineering Contradiction:

Improveapplication type identification capabilityVSAvoidimplementation cost and hardware upgrades

Core Design Contradiction:

Adaptability or versatilityVSEase of manufacture

Solution Approach 1:

The patent uses inexpensive statistical metrics (buffer size, resource blocks, time intervals) as proxies for application identification, replacing expensive DPI hardware. These statistical measurements are cheap to obtain and process, enabling widespread deployment without significant hardware investment.

Inventive Principle:

Principle #27Cheap short-living objects (Disposable)

Solution Approach 2:

The patent changes the identification approach from content-based DPI parameters to statistical parameters of bearer behavior. By measuring buffer sizes, resource allocations, and temporal patterns instead of packet contents, the system achieves application type identification with existing network infrastructure, avoiding hardware upgrades.

Inventive Principle:

Principle #35Parameter changes

3Productivity

If conventional bearer metrics are used without classification, then system simplicity is maintained, but Quality of Service optimization for different applications is limited

Engineering Contradiction:

ImproveQuality of Service optimizationVSAvoidclassification system complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent performs preliminary classification of bearers into application types using statistical analysis before QoS optimization is applied. This preliminary categorization enables the system to apply appropriate QoS policies for different application types (e.g., video streaming, file transfer, voice) without real-time complexity, improving overall service quality.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent implements a feedback mechanism where classification results inform QoS policy adjustments. The system continuously monitors bearer statistics, classifies applications, and adjusts resource allocation accordingly, creating a closed-loop system that optimizes QoS based on actual application behavior patterns.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS9780997B2Method and system for controlling an operation of an application by classifying an application type using data bearer characteristics

Publication Date: 2017.10.03 ALCATEL LUCENT SA

AI summary

An application type of a bearer is classified by computing statistics vectors of bearer metrics and locating points on a label map corresponding to the statistics vectors to obtain application type information. The application type information is exported to a network node to control an operation of application. The bearer metrics include bearer identifier information and bearer condition information, where the bearer condition information includes channel condition information and cell congestion level information. The bearers are paired, such that uplink and downlink bearers for a same application are identified, so that paired bearers are classified together. The label map is produced using previously classified bearer information to calculate cluster centroids and cluster regions that define portions of the map for particular application types. The bearer is classified by determining which cluster region is closest to points on the label map that are associated with the statistics vectors for a particular bearer.