IoT Device Identification via Behavioral Pattern Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In 5G networks, identifying Internet-of-Things (IoT) devices based on their predictive behaviors is challenging due to the use of common international mobile equipment identity (IMEI) types, making it difficult to distinguish between devices without physical inspection, as current methods rely on serial TAC values associated with embedded modules rather than integrated devices.
Innovation Solution
A master aggregation IMEI database (MAID) aggregates data from various sources, including TAC and IMEI databases, to predict device identities by analyzing behavior patterns, such as mobility, connectivity frequency, and location, using artificial intelligence and machine learning to assign confidence levels and group unknown devices with known types.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If common IMEI types are used for IoT devices, then device manufacturing is simplified and cost is reduced, but device identification precision deteriorates making it difficult to distinguish between devices
Solution Approach 1:
The patent segments device identification into multiple layers: first identifying device type using common IMEI/TAC values, then further distinguishing individual devices through behavioral pattern analysis. This segmentation allows manufacturing to use cost-effective common IMEI types while achieving precise identification through the additional behavioral layer.
Solution Approach 2:
The patent introduces behavioral patterns as an intermediary between the common IMEI type and device identification. Instead of relying solely on hardware identifiers, the system uses behavioral characteristics (mobility, connectivity patterns, location) as a mediating factor to achieve precise device distinction without changing the fundamental IMEI structure.
2Measurement precision
If physical inspection is performed to identify devices, then identification accuracy is improved, but time consumption and operational complexity increase
Solution Approach 1:
The patent performs preliminary action by continuously collecting and analyzing behavioral data in the background before identification is needed. Device behavioral patterns are pre-characterized and stored, allowing rapid matching and identification without requiring real-time physical inspection or manual intervention.
Solution Approach 2:
The patent replaces the mechanical approach of physical device inspection with an automated electronic system that analyzes behavioral data from network communications. This substitution eliminates the need for manual physical inspection while maintaining high identification accuracy through automated pattern recognition.
3Reliability
If more data is collected for prediction, then identification confidence level is improved, but system complexity and data processing requirements increase
Solution Approach 1:
The patent applies partial action by selecting and analyzing only the most relevant behavioral parameters (mobility, connectivity frequency, location patterns) rather than attempting to process all possible device data. This focused approach achieves high identification confidence while managing system complexity through selective data collection and analysis.
Data Source
AI summary
Internet-of-things (IOT) devices can be identified based on specific behavioral patterns when their identification data is unknown. Previously identified IOT devices with similar behavioral patterns can be used as a baseline from which to compare data that is available about unknown IOT devices. For example, an IOT device can be pooled with a group of IOT devices based on the frequency with which they connect to a wireless network. Additionally, a confidence level of the unknown device being associated with the group of IOT devices can be generated based on such comparison data.