Wearable Sensor Data Segmentation for Validity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional wearable devices have a high user operation requirement and low data collection success rate due to strict contact duration limitations, leading to frequent and prolonged data collection processes.
Innovation Solution
A data collection method that selects valid data segments from original collected data, even if the contact status between the human body and sensor is not continuous, allowing for the use of partial valid data to improve collection success rates and reduce the number of data collection times.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the collected part of the human body is required to be in continuous contact with the sensor for preset duration, then the data quality is improved, but the user operation requirement increases and collection success rate decreases
Solution Approach 1:
The patent segments the collected data into valid data segments and invalid data segments based on contact quality. Instead of requiring continuous contact for the entire preset duration, the system identifies and extracts valid data segments that meet quality standards from the collected data, allowing flexible combination of multiple short valid segments to satisfy the total duration requirement.
Solution Approach 2:
The patent changes the evaluation parameter from continuous contact duration to valid data segment duration. The system evaluates data quality based on the cumulative duration of valid data segments rather than requiring uninterrupted contact, thereby changing the fundamental parameter for assessing successful data collection.
2Measurement precision
If the collected data contains invalid data exceeding preset threshold, then the data quality is maintained, but the total data collection duration increases
Solution Approach 1:
The patent extracts valid data segments from the collected data that contains both valid and invalid portions. Instead of discarding entire datasets with invalid segments, the system extracts and utilizes only the valid portions, thereby reducing the need for repeated collection attempts and shortening total collection duration.
Solution Approach 2:
The patent recovers valid data from previously discarded collections. By identifying valid data segments in collections that originally failed quality checks, the system recovers usable data that would otherwise be thrown away, reducing redundant collection operations and total time investment.
3Measurement precision
If the collected data contains invalid data exceeding preset threshold, then the data quality is maintained, but the number of data collection times increases
Solution Approach 1:
The patent segments collected data into valid and invalid portions, allowing the system to identify usable data even when overall collection quality is poor. This segmentation enables higher collection success rates by recognizing that partial valid data can be extracted from previously failed collections.
Solution Approach 2:
The patent changes the success criterion from binary (entire collection valid/invalid) to cumulative (total valid segment duration meets requirement). This parameter change increases the probability of achieving successful data collection by allowing flexible aggregation of valid segments across multiple collection attempts.
Data Source
Figure 1
Figure 2~3
Figure 4~5
AI summary
A data collection method, a terminal device (100, 110, 130), and a storage medium are provided, and applicable to the field of electronic technologies. The data collection method includes: obtaining at least one piece of original collected data of a target user collected by a sensor in a first preset time period, where the original collected data includes a valid data segment (S31); selecting, from the at least one piece of original collection, at least one target valid data segment (DP1, DP2, DP3, DP4) whose total duration meets a preset duration requirement (S32); and obtaining target collected data of the target user based on the target valid data segment (DP1, DP2, DP3, DP4) (S33). According to the data collection method, a user operation requirement in a data collection process can be reduced, a collection success rate of the target collected data for data analysis can be improved, a total quantity of data collection times can be reduced, and total data collection duration can be shortened.