Problem
Research questions and friction points this paper is trying to address.
on-device cardiac feature extraction
ultra-low-power FPGA
Seismocardiography
resource-constrained wearable sensors
real-time health monitoring
At the Edge of the Heart: ULP FPGA-Based CNN for On-Device Cardiac Feature Extraction in Smart Health Sensors for Astronauts
🤖 AI Summary
This work proposes an ultra-low-power, real-time cardiac feature extraction method tailored for resource-constrained wearable health monitoring in aerospace environments. Implemented on a Lattice iCE40UP5K FPGA, the approach employs quantization-aware training to construct a fully integer-based convolutional neural network and integrates a systolic array accelerator to enable entirely on-chip inference. For the first time, this solution achieves high-accuracy, radiation-tolerant seismocardiogram (SCG) feature classification on an extremely low-power FPGA, consuming only 8.55 mW and utilizing 2,861 LUTs (with just 7 DSPs). The system delivers a classification accuracy of 98% with an inference latency of 95.5 ms, striking an exceptional balance among energy efficiency, hardware footprint, and suitability for space applications.
📝 Abstract
The convergence of accelerating human spaceflight ambitions and critical terrestrial health monitoring demands is driving unprecedented requirements for reliable, real-time feature extraction on extremely resource-constrained wearable health sensors. We present an ultra-low-power (ULP) Field-Programmable Gate Array (FPGA) based solution for real-time Seismocardiography (SCG) feature classification using Convolutional Neural Networks (CNNs). Our approach combines quantization-aware training with a systolic-array accelerator to enable efficient integer-only inference on the Lattice iCE40UP5K FPGA, which offers an ideal platform for battery-powered deployments -- particularly in space environments -- thanks to its power efficiency and radiation resilience. The implementation achieves a validation accuracy of 98% while consuming only 8.55 mW, completing inference in 95.5 ms with minimal hardware resources (2,861 LUTs and 7 DSP blocks). These results demonstrate that fully on-device SCG-based cardiac feature extraction is feasible on resource-constrained hardware, enabling energy-efficient, autonomous health monitoring for astronauts in long-duration space missions.
Innovation
Methods, ideas, or system contributions that make the work stand out.
ULP FPGA
CNN
Seismocardiography
quantization-aware training
systolic-array accelerator
🔎 Similar Papers
No similar papers found.
