Machine Learning Device for BLE Propagation State Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for simulating radio wave propagation states between HVAC devices equipped with radio-wave transmitters/receivers are either complex, time-consuming, or lack accuracy, particularly in narrow spaces like those above ceilings.
Innovation Solution
A machine learning apparatus that learns the radio wave propagation state between radio devices by acquiring information related to the distance and objects between them, using this information to learn state variables and radio wave propagation states in association, thereby improving accuracy and simplifying the process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual association work is performed to store network address and physical arrangement information, then data accuracy is improved, but time consumption and labor cost increase
Solution Approach 1:
The system enables self-service by having HVAC devices automatically perform radio wave transmission and reception, and automatically store their own network address and physical arrangement information in the management device, eliminating the need for manual association work while maintaining data accuracy
Solution Approach 2:
The patent replaces the manual mechanical process of data entry and association with an automated radio wave-based system. The management device automatically obtains physical arrangement information by receiving radio wave strength data from HVAC devices, substituting human labor with automated electronic measurement and processing
2Measurement precision
If radio wave propagation simulation is performed using existing methods, then propagation state estimation is achieved, but system complexity and computation time increase
Solution Approach 1:
The patent extracts only the essential elements needed for propagation estimation: radio wave strength measurements from multiple HVAC devices and basic distance calculations. By taking out only the necessary components and discarding complex simulation parameters, the system achieves accurate propagation state estimation with minimal complexity
Solution Approach 2:
Instead of performing complex physical simulations, the system creates a simplified mathematical model that copies the essential behavior of radio wave propagation. The management device calculates propagation states by comparing measured radio wave strengths against a simplified distance-based model, achieving accurate results without complex computation
3Productivity
If radio wave strength information is used to estimate HVAC device arrangement, then manual work is reduced, but measurement accuracy decreases in narrow spaces
Solution Approach 1:
The patent merges information from multiple HVAC devices by having each device transmit radio waves that are received by all other devices. The management device combines the radio wave strength data from multiple transmission-reception pairs to calculate distances and determine physical arrangements, improving accuracy in narrow spaces through multi-point measurement fusion
Solution Approach 2:
The system implements feedback by using the measured radio wave strengths to calculate distances, then using these distance calculations to determine and verify the physical arrangement of devices. This feedback loop allows the system to iteratively refine its estimation of device locations, improving accuracy in challenging environments like narrow ceiling spaces
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A computer (10) is a machine learning apparatus that learns a radio wave propagation state between a BLE module and another BLE module, and includes an acquisition unit (20) and a learning unit (30). The acquisition unit (20) acquires air conditioner arrangement information (21) and beam arrangement information (22) as information for obtaining state variables. These pieces of information are information related to something between the BLE module and the other BLE module. The learning unit (30) learns the state variables and the radio wave propagation state between the BLE module and the other BLE module in association with each other.