Energy-Efficient Real-Time 4-Stage Sleep Classification at 10-Second Resolution: A Comprehensive Study

TL;DR

This study develops an energy-efficient, real-time sleep stage classification system using ECG signals, achieving high accuracy with low power consumption suitable for wearable devices.

Contribution

The paper introduces a novel low-power deep learning model, SleepLiteCNN, and two windowing strategies for near-real-time sleep staging from ECG data.

Findings

MobileNet-v1 achieves 92% accuracy but consumes significant energy.

SleepLiteCNN attains 89% accuracy with only 5.48 microjoules per inference.

Quantization and FPGA deployment further reduce power consumption.

Abstract

Sleep stage classification is crucial for diagnosing and managing disorders such as sleep apnea and insomnia. Conventional clinical methods like polysomnography are costly and impractical for long-term home use. We present an energy-efficient pipeline that detects four sleep stages (wake, REM, light, and deep) from a single-lead ECG. Two windowing strategies are introduced: (1) a 5-minute window with 30-second steps for machine-learning models that use handcrafted features, and (2) a 30-second window with 10-second steps for deep-learning models, enabling near-real-time 10-second resolution. Lightweight networks such as MobileNet-v1 reach 92 percent accuracy and 91 percent F1-score but still draw significant energy. We therefore design SleepLiteCNN, a custom model that achieves 89 percent accuracy and 89 percent F1-score while lowering energy use to 5.48 microjoules per inference at 45 nm. Applying eight-bit quantization preserves accuracy and further reduces power, and FPGA deployment confirms low resource usage. The proposed system offers a practical solution for continuous, wearable ECG-based sleep monitoring.