Low-Level Mobile Activity Detection for MVNO Bandwidth Forecasting

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Mobile Virtual Network Operators (MVNOs) lack sufficient insight into UE usage patterns due to limited data access from Mobile Network Operators (MNOs, hindering accurate prediction and management of bandwidth usage.

Innovation Solution

A system that utilizes artificial intelligence (AI) to analyze low-level information, create embeddings, and cluster activities to predict bandwidth usage, enabling MVNOs to determine optimal plans and request necessary bandwidth from MNOs.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If MVNOs rely on limited data access from MNOs, then data sharing complexity is reduced, but bandwidth usage prediction accuracy deteriorates

Engineering Contradiction:

Improvebandwidth usage prediction accuracyVSAvoiddata processing complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent introduces an AI-based processing layer as an intermediary between MNO data sources and MVNO analysis needs. This intermediary automatically extracts meaningful patterns from raw MNO data, transforming the complexity burden from MVNOs to a dedicated processing system that bridges the data accessibility and analysis accuracy requirements

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces manual or traditional statistical analysis methods with AI/ML-based pattern recognition systems. This substitution enables more accurate prediction of bandwidth usage patterns from limited data by leveraging neural networks and machine learning algorithms that can extract complex non-linear relationships automatically

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If MVNOs access more detailed UE data from MNOs, then bandwidth usage prediction accuracy improves, but data sharing complexity increases

Engineering Contradiction:

Improvebandwidth usage prediction accuracyVSAvoiddata sharing implementation ease

Core Design Contradiction:

Measurement precisionVSEase of manufacture

Solution Approach 1:

The patent extracts and processes only the essential features and patterns from the full MNO datasets using AI/ML techniques. By taking out only the meaningful information needed for prediction rather than transferring complete raw data, the system achieves high prediction accuracy while minimizing data sharing complexity and privacy concerns

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The AI processing system acts as an intermediary that receives detailed MNO data, processes it through pattern recognition algorithms, and delivers simplified prediction results to MVNOs. This intermediary approach enables MVNOs to access high-quality predictive insights without directly handling complex raw data from MNOs

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS12413990B2Determining an activity associated with a mobile device based on a low-level information representing the activity

Publication Date: 2025.09.09 BOOST SUBSCRIBERCO LLC

AI summary

The system obtains a first low-level information representing a first multiplicity of activities. The first low-level information includes a first log including a first multiplicity of entries representing the first activity. An entry among the first multiplicity of entries can represent various activities. The system obtains a second low-level information representing a second multiplicity of activities. The second low-level information includes a second log comprising a second multiplicity of entries. The system obtains a correspondence between the second multiplicity of activities and the second low-level information. The correspondence maps a second activity among the second multiplicity of activities to a subset of entries in the second log. The system determines that the subset of entries in the second log and a subset of entries in the first log are similar and the subset of entries in the first log indicates the second activity.