Radio Tag Positioning with Learned Model
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing communication devices face accuracy issues in determining the position of radio tags due to environmental disturbances when using a fixed threshold value for phase difference determination.
Innovation Solution
A communication device that includes an antenna, a driving unit, a first acquisition unit, an input unit, and a second acquisition unit, which acquires and processes tag data using a learned model generated through machine learning to determine whether a radio tag is within a specific range, improving accuracy by adapting to varying environmental conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a fixed threshold value is used for phase difference determination, then the determination process is simple, but determination accuracy deteriorates due to environmental disturbances
Solution Approach 1:
The patent replaces the fixed threshold value with a dynamic threshold that is learned through machine learning from phase difference data collected under various environmental conditions. The learning unit continuously adapts the threshold based on actual measurement patterns, allowing the system to automatically adjust to changing environmental factors while maintaining high determination accuracy.
Solution Approach 2:
The system implements feedback by using the learning unit to process actual phase difference measurements and update the threshold value accordingly. This closed-loop approach allows the system to learn from real-world data and improve its determination accuracy over time, compensating for environmental disturbances that would otherwise degrade performance.
2Measurement precision
If a learned model is used for position determination, then position determination accuracy is improved, but device complexity increases
Solution Approach 1:
The learning unit performs preliminary learning during an initial phase by collecting and processing phase difference data to establish the threshold model before actual position determination begins. This preliminary action prepares the system with optimized determination parameters, allowing accurate position determination to proceed without requiring complex real-time calculations during operation.
Solution Approach 2:
The system creates a simplified representation (copy) of the complex environmental factors through the learned threshold model. Instead of directly processing all environmental variables in real-time, the system uses the pre-learned threshold as a simplified decision criterion, reducing computational complexity while maintaining accuracy.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The solution enhances the accuracy of determining the position of radio tags by using a learned model based on machine learning, effectively mitigating the impact of environmental disturbances and improving the precision of range determination.
Implementation Method 1
receiving a radio wave transmitted from the radio tag attached to an article with an antenna
Implementation Method 2
measuring a phase of the radio tag... determines whether the radio tag is within the predetermined range or out of the predetermined range based on a phase difference which is a change amount of the measured phase
Data Source
AI summary
A communication device includes an antenna, a driving unit, a first acquisition unit, an input unit, and a second acquisition unit. The driving unit moves a position of the antenna. The first acquisition unit acquires tag data of each radio tag based on a radio wave of each radio tag received by the antenna at a plurality of positions of the antenna. The input unit inputs the tag data of each radio tag at the plurality of positions of the antenna into a learned model. The second acquisition unit acquires, from the learned model, data indicating whether each radio tag is included in a first range or a second range based on the input of the tag data of each radio tag into the learned model by the input unit.