Wearable Sensor Data Collection Using LSTM Sample Skipping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Wearable medical devices face significant energy consumption due to wireless transmission and high sampling rates for physiological data, necessitating energy-efficient data collection methods that maintain diagnostic accuracy without frequent data transmissions.
Innovation Solution
Implementing a sample prediction mechanism using advanced machine learning, specifically Long Short-Term Memory (LSTM) networks, to forecast upcoming physiological data samples, allowing for the skipping of predictable samples and optimizing packet formation to reduce energy consumption and data volume.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high sampling rates are used to collect physiological data, then measurement precision is improved, but energy consumption increases
Solution Approach 1:
The system performs preliminary actions by predicting future sensor data samples using RNN-LSTM models before actually collecting them. The prediction model is trained offline on historical data, and during operation, it forecasts upcoming samples, allowing the system to skip actual high-rate sampling and wireless transmission of predictable data, thereby reducing energy consumption while maintaining measurement precision for significant events.
2Productivity
If wireless transmission frequency is increased to improve data throughput, then information delivery speed is improved, but energy consumption increases
Solution Approach 1:
The system applies skipping by selectively omitting the wireless transmission of sensor data samples that are predicted to be redundant or highly correlated with previous samples. The RNN-LSTM model identifies which samples can be skipped, and only significant or unpredictable samples are transmitted, thereby maintaining data throughput for critical information while dramatically reducing the energy consumption associated with frequent wireless transmissions.
3Use of energy by moving object
If data compression is applied to reduce transmission volume, then energy consumption is reduced, but measurement precision deteriorates
Solution Approach 1:
The RNN-LSTM prediction model serves as an intermediary between the sensor data collection and wireless transmission processes. Instead of directly compressing the raw sensor data (which would lose information), the prediction model generates estimated samples that serve as a proxy for the actual data. This intermediary approach allows the system to reconstruct signals with high accuracy by combining received samples with predicted samples, thereby reducing transmission volume and energy consumption without sacrificing measurement precision.
Data Source
AI summary
A wearable garment with sensors attached to obtain physiological data. The sensors are incorporated to form a body area sensor network to obtain the data. This provides patients with improved health monitoring by aggregating multiple interconnected nodes on a human body for sensorimotor measurements and provides patients with quantitative measurements of their progress. The data is obtained in a way that allows for the number of transmissions to be reduced thereby conserving the energy of the wearable devices. This is made possible by each sensor reducing the number of samples by eliminating predictable samples and configuring the sensors to pack the data efficiently. A neural network can determine whether a sample can be skipped or needs to be reported. A long short term memory architecture creates a waveform for a given snapshot of samples based on the previous samples regardless of whether these samples were reported or predicted.