Wearable Sensor Network Data Packing With Predictive 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 increasing computational overhead.
Innovation Solution
Implementing a sample prediction-based energy optimization (SPEO) using LSTM networks at sensor and gateway nodes to forecast data samples, reducing unnecessary transmissions and optimizing packet formation for energy conservation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high sampling rates are used for collecting physiological data, then measurement precision is improved, but energy consumption increases
Solution Approach 1:
The system performs preliminary actions by predicting future data samples using LSTM networks before actual measurement and transmission occurs. The prediction model forecasts physiological data trends, allowing the system to skip transmission of predictable samples while maintaining diagnostic accuracy, thus reducing energy consumption without sacrificing measurement precision.
Solution Approach 2:
The system dynamically changes the sampling and transmission parameters based on predicted data variability. When predictions indicate low variability or high confidence in forecasted values, the system reduces sampling rate or skips transmission entirely. When variability increases, the system increases sampling rate, thereby adapting measurement precision and energy consumption to actual physiological conditions.
2Loss of information
If wireless transmission frequency is increased for real-time monitoring, then information completeness is improved, but energy consumption increases
Solution Approach 1:
The system employs feedback mechanisms where the LSTM prediction model continuously learns from actual transmitted data to improve future predictions. The gateway node also uses received data to update its prediction model, creating a closed-loop system that adapts to individual patient patterns. This feedback enables the system to maintain information completeness by transmitting only when predictions diverge from actual measurements, reducing unnecessary transmission energy consumption.
Solution Approach 2:
Instead of transmitting all sampled data, the system applies partial action by selectively transmitting only those samples that deviate significantly from predicted values or meet specific transmission criteria. This partial transmission approach maintains sufficient information completeness for diagnostic purposes while dramatically reducing the number of wireless transmissions and associated energy consumption.
3Use of energy by moving object
If data compression is applied to reduce transmission volume, then energy consumption is reduced, but computational overhead increases
Solution Approach 1:
The system introduces an intermediary prediction layer (LSTM network) between data acquisition and transmission. This intermediary forecasts future data points and identifies which samples warrant transmission, effectively pre-processing data to minimize transmission volume. The computational overhead is concentrated in the prediction model rather than in transmission protocols, and the gateway node performs complementary prediction to reconstruct skipped samples, reducing overall system complexity compared to traditional compression methods.
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.