Multi-Sensor Sleep Stage Detection via Data Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current sleep stage detection solutions rely on single sensors, such as UWB or microphones, resulting in low accuracy, and embedding EEG sensors in commercial devices is challenging, making it difficult to achieve precise sleep stage monitoring.
Innovation Solution
A sleep monitoring apparatus utilizing multiple sensors, including radar and piezoelectric sensors, with a processor that performs raw data fusion, feature extraction, and decision fusion to accurately determine sleep stages, enhancing accuracy through multi-sensor data integration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If single sensor (UWB or microphone) is used for sleep stage detection, then device complexity is reduced, but measurement precision deteriorates with accuracy around 70%
Solution Approach 1:
The patent combines multiple sensors (UWB, microphone, radar, piezoelectric sensors) into a single sleep monitoring apparatus, merging their detection capabilities to achieve accurate sleep stage classification without requiring invasive EEG sensors. This integration allows the system to leverage complementary strengths of different sensors to overcome individual limitations.
Solution Approach 2:
The sleep monitoring apparatus is designed with multi-functional sensors that can detect various physiological parameters simultaneously. The UWB sensor detects motion and breathing patterns, the microphone captures audio signals, and the radar detects chest movements, all within a single device that serves multiple detection purposes.
2Measurement precision
If EEG sensors are embedded for accurate sleep stage detection, then measurement precision is improved, but ease of manufacture deteriorates due to difficulty in embedding
Solution Approach 1:
The patent extracts the sleep stage detection function from invasive EEG sensors and implements it using non-invasive alternative sensors. By removing the requirement for EEG electrode embedding, the system achieves comparable accuracy through a combination of UWB, microphone, radar, and piezoelectric sensors that can be easily integrated into commercial devices.
Solution Approach 2:
The system copies the functional capability of EEG-based sleep stage detection using non-invasive sensor arrays. Instead of directly measuring brain waves with EEG, the system captures equivalent physiological information through motion, audio, and radar signals, replicating the diagnostic value without the manufacturing complexity of EEG integration.
3Measurement precision
If multiple sensors are integrated for sleep detection, then measurement precision is improved, but device complexity increases requiring raw data fusion and feature extraction
Solution Approach 1:
The patent segments the data processing pipeline into distinct stages: raw data collection from multiple sensors, feature extraction from each sensor type, feature fusion combining extracted features, and final sleep stage classification. This segmentation manages complexity by organizing processing tasks into modular, manageable steps rather than handling all raw data simultaneously.
Solution Approach 2:
The system introduces feature extraction and feature fusion as intermediary processing layers between raw sensor data and final sleep stage classification. These intermediaries transform complex multi-sensor raw data into meaningful features that can be effectively combined and classified, reducing the complexity of the final decision-making process.
Data Source
AI summary
A sleep monitoring apparatus includes a plurality of sensor modules, a transceiver, and a processor operatively coupled with the plurality of sensor modules and the transceiver. The processor is configured to receive, from the plurality of sensor modules, raw sensor data for each of the plurality of sensor modules related to a sleep session of a user of the sleep monitoring apparatus, and perform raw data fusion of the raw sensor data. The raw data fusion generates a fused raw data signal. The processor is further configured to, based on the fused raw data signal, perform feature extraction for the plurality of sensor module; based on the feature extraction, perform feature fusion for the plurality of sensor modules; based on the feature fusion, perform decision fusion for the plurality of sensor modules; and determine a sleep stage of the user based on the decision fusion.