Hardware Accelerator or ISA Extension: A Measured Co-Design of a PPG Heart-Rate Accelerator and a Multicycle RISC-V Soft Core

Wearable pulse oximeters must turn a photoplethysmogram (PPG) into heart rate (HR) and oxygen saturation(SpO2) on very small processors. Adding a dedicated acceleratoris a common answer, but its benefit is rarely measured againsta fair software baseline or against a cheaper instruction-setextension. This paper presents a system on chip (SoC) in which amulticycle RV32I soft core and a purpose-built PPG accelerator(moving-average filter, slope, peak detector and HR divider) sharea memory-mapped interface and read a MAX30102-class opticalsensor through an I2C controller. The same firmware algorithmwas built in four configurations (RV32I, RV32IM, and eachwith the accelerator) and characterised for latency, area, energyand accuracy. Against the RV32IM baseline the acceleratorreduces beat-to-result latency by 1.75× (476 to 272 cycles at50 MHz) and makes it constant, while costing 165 adaptive logicmodules (ALMs) and no DSP blocks on a Cyclone V, against 357ALMs and three DSP blocks for RV32M. The two additions arecomplementary: RV32M speeds the division-heavy SpO2 path by9.9×. The accelerator closes timing at 50 MHz in all corners ofthe SkyWater 130 nm process in 0.066 mm2 and 1.55 mW, ofwhich 99% is clock power. On 53 BIDMC records a model of thefirmware’s detector gives a mean absolute HR error of 2.40 BPM.

Authors

Institutions

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-18
DOI
https://doi.org/10.5281/zenodo.24350653
Primary Topic
Embedded Systems Design Techniques
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Hardware Accelerator or ISA Extension: A Measured Co-Design of a PPG Heart-Rate Accelerator and a Multicycle RISC-V Soft Core

R. Jayagowri, Aditya Sri Vasishta Vasishta, Manasi M M, Asher Raj
1 links
Zenodo (CERN European Organization for Nuclear Research)
Embedded Systems Design Techniques
preprint

Hardware Accelerator or ISA Extension: A Measured Co-Design of a PPG Heart-Rate Accelerator and a Multicycle RISC-V Soft Core

R. Jayagowri, Aditya Sri Vasishta Vasishta, Manasi M M, Asher Raj
preprint en

Abstract

Wearable pulse oximeters must turn a photoplethysmogram (PPG) into heart rate (HR) and oxygen saturation(SpO2) on very small processors. Adding a dedicated acceleratoris a common answer, but its benefit is rarely measured againsta fair software baseline or against a cheaper instruction-setextension. This paper presents a system on chip (SoC) in which amulticycle RV32I soft core and a purpose-built PPG accelerator(moving-average filter, slope, peak detector and HR divider) sharea memory-mapped interface and read a MAX30102-class opticalsensor through an I2C controller. The same firmware algorithmwas built in four configurations (RV32I, RV32IM, and eachwith the accelerator) and characterised for latency, area, energyand accuracy. Against the RV32IM baseline the acceleratorreduces beat-to-result latency by 1.75× (476 to 272 cycles at50 MHz) and makes it constant, while costing 165 adaptive logicmodules (ALMs) and no DSP blocks on a Cyclone V, against 357ALMs and three DSP blocks for RV32M. The two additions arecomplementary: RV32M speeds the division-heavy SpO2 path by9.9×. The accelerator closes timing at 50 MHz in all corners ofthe SkyWater 130 nm process in 0.066 mm2 and 1.55 mW, ofwhich 99% is clock power. On 53 BIDMC records a model of thefirmware’s detector gives a mean absolute HR error of 2.40 BPM.

Zenodo (CERN European Organization for Nuclear Research)
B.M.S. College of Engineering (IN)
Embedded Systems Design Techniques
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon