IoT Gateway Learning Control for Model Accuracy and Load
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing IoT gateway systems face challenges in balancing the accuracy and convenience of normal communication models, as learning for a short period may decrease accuracy, while prolonged learning degrades convenience and increases time requirements.
Innovation Solution
A control device and method that include a learning unit to create and manage normal communication models for IoT devices, with a determination unit to dynamically interrupt, finish, continue, or resume learning based on the load of the learning environment, ensuring accuracy and convenience.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If learning is performed for a long period of time to ensure accuracy, then model accuracy is improved, but creation time increases and convenience is degraded
Solution Approach 1:
The system performs learning periodically rather than continuously, applying the normal communication model at regular intervals. This allows the model to be updated with fresh data while avoiding excessive creation time, resolving the contradiction between accuracy and time consumption.
Solution Approach 2:
The learning period and frequency are made dynamic and adjustable based on system load conditions. When resource load is low, longer learning periods can be used to improve accuracy; when load is high, shorter periods maintain convenience. This dynamic adjustment resolves the fixed trade-off between accuracy and creation time.
2Ease of operation
If learning is performed for a short period of time, then convenience is improved, but model accuracy decreases
Solution Approach 1:
By implementing periodic learning cycles, the system ensures that even short learning periods contribute to gradual model improvement. The periodic application of the model allows multiple short learning cycles to accumulate accuracy benefits while maintaining operational convenience between cycles.
Solution Approach 2:
The system maintains continuous operation with the learned model between learning cycles, ensuring convenience is not interrupted. The useful action of communication control continues uninterrupted, while learning occurs in scheduled periods, resolving the contradiction between short learning duration and model accuracy.
3Measurement precision
If learning processing is continuously performed, then model accuracy is maintained, but system load increases
Solution Approach 1:
Learning processing is executed periodically rather than continuously, reducing system load during non-learning periods. The model maintains accuracy through these periodic updates while the system enjoys reduced resource consumption between cycles, resolving the contradiction between continuous accuracy maintenance and system load.
Solution Approach 2:
The communication control function continues operating continuously using the learned model, maintaining accuracy without requiring continuous learning. This separation of model maintenance (continuous) from model creation (periodic) reduces energy usage while preserving measurement precision.
Data Source
AI summary
Provided is an IoT GW (10), including a learning unit (131) configured to create, for each IoT device connected to the IoT GW (10), a normal communication model (122) that has learned a normal communication pattern of the IoT device; and a determination unit (132) configured to: determine, for learning by the learning unit (131), whether to interrupt, finish, continue, and resume learning based on a load of a learning environment; and control learning processing by the learning unit (131) based on a result of determination.