Millimeter-Wave Radar Sleep Stage Detection Without Polysomnography
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current sleep monitoring products are unable to accurately detect sleep stages without the inconvenience of polysomnography in a sleep laboratory.
Innovation Solution
An electronic device using a millimeter wave radar to extract physiological signals, perform fast Fourier transform, and utilize machine learning models to generate accurate sleep stage predictions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If polysomnography is performed in a sleep laboratory to accurately detect sleep stages, then measurement precision is improved, but device complexity and ease of operation deteriorate
Solution Approach 1:
The patent extracts the core measurement function from the complex polysomnography system by using a simple millimeter wave radar to detect only respiratory signals, which are then processed to infer sleep stages. This extracts the essential measurement capability while removing the complexity of multiple sensors and electrodes required in traditional PSG.
Solution Approach 2:
The patent replaces the mechanical contact-based measurement system (electrodes and wires in traditional PSG) with a non-contact millimeter wave radar system. This substitution maintains measurement precision while dramatically reducing device complexity and improving ease of operation by eliminating the need for physical contact with the subject.
2Measurement precision
If polysomnography is performed in a sleep laboratory to accurately detect sleep stages, then measurement precision is improved, but ease of operation deteriorates
Solution Approach 1:
The patent replaces the cumbersome mechanical setup of traditional polysomnography with a non-contact millimeter wave radar system. This allows subjects to be monitored without wearing electrodes or being connected to equipment, dramatically improving ease of operation and convenience while maintaining accurate sleep stage detection through respiratory signal analysis.
Solution Approach 2:
The system enables self-service monitoring where the millimeter wave radar automatically detects and processes respiratory signals without requiring subject participation or adjustment. The subject simply needs to be in the monitoring area, and the system automatically performs sleep stage detection, eliminating the need for technical personnel to set up and adjust complex equipment.
3Ease of operation
If simple monitoring products are used to monitor sleep status, then ease of operation is improved, but measurement precision deteriorates
Solution Approach 1:
The patent introduces an intermediary processing layer that transforms simple radar respiratory signals into accurate sleep stage classifications. The signal processing module extracts respiratory features and the machine learning model interprets these features to determine sleep stages, acting as an intermediary that bridges the gap between simple measurement and precise classification.
Solution Approach 2:
The patent changes the analysis parameter from direct sleep stage measurement to respiratory signal frequency analysis. By performing fast Fourier transform on respiratory signals and analyzing peak area ratios at different frequencies, the system infers sleep stages indirectly, achieving high precision with simple initial measurements through parameter transformation.
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
Generates accurate and easily interpretable sleep stage predictions, including wakefulness, REM, and deep sleep stages, by combining radar signal processing with machine learning models.
Implementation Method 1
A transceiver receives a radar signal
Implementation Method 2
perform fast Fourier transform on the physiological signal by using a first window to obtain a transformed signal
Data Source
AI summary
The disclosure provides an electronic device and a method of detecting a sleep stage. The method includes the following. A radar signal is received, and a physiological signal is extracted from the radar signal. Fast Fourier transform is performed on the physiological signal by using a first window to obtain a transformed signal. A peak area ratio corresponding to the transformed signal is obtained according to a peak of the transformed signal. A first prediction result of the sleep stage is generated according to the peak area ratio by using a first machine learning model. The first prediction result is outputted.