Merging Heterogeneous Rulesets for IoT Device Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
As the Internet of Things (IoT) expands, existing device classification systems face challenges in accurately identifying device types due to the proliferation of diverse IoT devices, leading to a high percentage of devices being classified as 'UNKNOWN' and varying effectiveness across different classification rulesets, which complicates network access control and security policies.
Innovation Solution
A device classification service that merges and optimizes heterogeneous rulesets by resolving conflicts and training a machine learning-based classifier using unified rulesets, enabling more accurate and comprehensive device type classification across a broader range of devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple heterogeneous device classification rulesets are used to cover diverse IoT devices, then the coverage of device types increases, but the conflict between conflicting device characteristics from different rulesets increases
Solution Approach 1:
The patent merges multiple heterogeneous device classification rulesets into a single unified ruleset. The device classification service receives rulesets from different sources, resolves conflicts between them by establishing a priority scheme, and consolidates them into one unified ruleset that maintains comprehensive device type coverage while eliminating contradictions.
Solution Approach 2:
The unified ruleset serves as a universal classification mechanism that handles diverse IoT device types. Instead of maintaining separate heterogeneous rulesets for different device categories, the system creates a single multi-functional ruleset that can classify various device types through a common framework with resolved conflicts.
2Ease of operation
If traditional rule-based classification systems are used for device classification, then the system is easy to understand and implement, but the classification accuracy decreases with increasing device diversity
Solution Approach 1:
The patent replaces traditional mechanical rule-based classification systems with a machine learning-based classification system. The unified ruleset is used to train a machine learning model that automatically learns classification patterns, achieving higher accuracy for diverse IoT devices while maintaining operational simplicity through automated decision-making.
Solution Approach 2:
The system transforms static rule-based classification into dynamic machine learning-based classification. By changing from fixed rules to adaptive learning parameters, the system can automatically adjust to new device types and patterns, improving classification accuracy without increasing operational complexity for end users.
3Measurement precision
If machine learning-based classification is implemented, then classification accuracy improves, but the computational resources and training time increase
Solution Approach 1:
The patent performs preliminary action by pre-processing and unifying the ruleset before training the machine learning classifier. By resolving conflicts and consolidating rules in advance, the training process becomes more efficient and requires less time. The unified ruleset serves as pre-prepared training data that reduces the computational burden during the actual training phase.
Solution Approach 2:
The system extracts and separates the ruleset unification and conflict resolution processes from the main training process. By handling ruleset consolidation as a preliminary step, the actual classifier training focuses only on learning from clean, unified data, reducing overall training time and computational resource requirements.
4Adaptability or versatility
If comprehensive device classification rulesets are maintained, then classification coverage improves, but the size and complexity of the rulesets increase
Solution Approach 1:
The patent combines multiple smaller heterogeneous rulesets into one unified ruleset, eliminating redundancy and conflicts. Instead of maintaining separate large rulesets for different device types, the system merges them into a single compact unified ruleset that achieves comprehensive coverage through consolidated logic and resolved contradictions.
Data Source
AI summary
In one embodiment, a device classification service receives a plurality of device classification rulesets, each ruleset associating a set of device characteristics with a device type label. The device classification service forms a unified ruleset by resolving a conflict between conflicting device characteristics from two or more of the device classification rulesets. The device classification service trains a machine learning-based device classifier using the unified ruleset. The device classification service classifies, using telemetry data for a device in a network as input to the trained device classifier, the device with the device type label.