HOME Meet the Team! Our Approach Meet Our Sponsors! Semester 1 Semester 2 Bibliography

The Heart Rate Monitors

Welcome to Our Portfolio!


Meet the Team!

Jacob Gardner

I'm a Senior at Northern Arizona University studying Electrical Engineering. I started college in 2017 at community college in Phoenix, and finally transferred to University in 2020. My main focus is system integration and semiconductors. After graduation I plan on getting a Masters Degree and working in the semiconductor industry. In my spare time I enjoy hiking, motorsports, and spending time with my family.

Email: jng247@nau.edu

Zachary Price

I am a Senior at Northern Arizona University majoring in computer engineering. I came to NAU on the lumberjack scholarship and have been here for the last three years. Throughout my years I have taken more and more interest in coding, and may come back for another degree in computer science. In my free time I enjoy coding, tinkering with computers and playing games with my friends.

Email: ztp33@nau.edu

Steven Provencio

Hello my name is Steven Provencio and I am from Chandler, Arizona. I am currently studying electrical engineering. I'm a senior here at NAU and am expected to graduate winter of 2023. In my free time, I enjoy spending time with friends and family. I also enjoy playing and watching soccer as well as messing around with electronic devices such as the Raspberry Pi.

Email: snp82@nau.edu

Our Approach

The Problem

Photoplethysmography (PPG) is a noninvasive technique used to detect vital signs such as heart rate (HR), saturation of peripheral oxygen (SpO2), and blood pressure (BP). Recently, wearable finger-type PPG devices are increasingly developed for the convenience of the person under monitoring (PUM). The two most critical features of wearable PPG devices are high accuracy and long operation time. To enhance these functions, this project aims to develop a method to select, process, and transfer only high-quality PPG signals. Hence, data quality is significantly improved and power consumption on the wireless module is minimized. For our results to be evaluated as an exemplary project, we are expected to develop an HR estimation framework utilizing a deep learning model based on a convolutional neural network (CNN) and long short-term memory (LSTM) network. If successful, the outcomes of this project will improve personal healthcare.

Statement of Needs

For the heart rate monitor there are key needs that need to be met. The heart rate monitor needs to be affordable and 90% accurate with the readings. It needs to be able to clip on the finger of any person and transfer the signals to a computer wirelessly. This device needs to have a battery and be rechargeable. The device also needs to be able to recognize any signal it's given and filter out the heart rate and blood oxygen level. The filtration and signal tuning needs to be done with machine learning.

List of Requirements

Engineering Requirements
  • At least 90% accurate when compared to industry standard.
  • Use Machine Learning to filter out bad reads.
  • Successfully connect MAX30102 sensor and ESP 32 Wifi/BLE module.
  • Connect to a computer to collect heart rate data.
  • Limit noise in order to ensure accurate data.
  • Low power/low current.
Safety Requirements
  • Comfortably grabs the finger.
  • Device is able to be cleaned for sanitary precautions.
  • Compact.
  • No long wires.
  • Transmitter strapped to user's arm.
  • Clean wiring and strong solder joints.
Marketing Requirements
  • Cost Efficient, below $250.
  • Comfortable and compact transmitter.
  • Compact.
  • No long wires.
  • Transmitter strapped to user's arm.
  • Clean wiring and strong solder joints.

Our Fall Semester Summary

This semester our group wanted to focus on research and prototyping our device. We were able to recieve all of our parts necessary to get started on a working prototype. Our first priority was getting the MAX30102 to communicate with the ESP32 in order to send signals to a computer. In order to accomplish this, we used Arduino IDE which helped show us our unfiltered data. What we noticed right off the bat was that our sensor is very sensitive so we needed to house the device in a finger clamp that would not allow ambient light to penetrate into the sensor. Our clamp was made using a 3D printer along with a spring and a couple screws. To end our semester, we were able to showcase our project at the Engineering Fest where we were able to share our progress and plans for our project. We were happy that our sponsors, as well as our fellow classmates and professors, were excited to see our progress.

Fall Semester Progress


Testings and Findings

Test 1


Results: After improving our finger clamp design, we were able to reduce the amount of ambient light shining through to the sensor which resulted in a significant change in quality raw data collected. Our testing was a success even though our ppg signals collected were reversed compared to the good data we referenced. All we were tasked with this semester was making sure our ppg sensor communicates with the microcontroller then translating that to a computer. Next semester we are tasked with cleaning up and filtering the ppg data and this experiment points us in the right direction.

Test 2


Results: We were reluctant to have multiple sensors in order to freely experiment with soldering and incorporating our design in different applications. Our first two sensors were not able to produce the most quality signals most likely due to the breadboard-to-wire connections. We found our best results when the ppg sensor and the wires were soldered to each other. The cleaner our connections were, and the less disturbance between each connection, it resulted in producing a higher quality ppg signal.

Plans for next semester

We are very proud of our progress thus far and cannot wait to continue next semester! We plan to work researching ways we can successfully filter the incoming signals using machine learning in matlab. We also plan on adding a wireless charging unit to our device in order to make our design more compact. When doing this, we will need to use the wifi module on the ESP32 in order to send the data instead of relying on the wired connection from the ESP32 directly to our computer. Also, we will be integrating an IC for our wireless charging unit that will be PCB designed. Our last priority is making sure all aspects of our device, like our finger clamp, look nice and clean while still being able to function like expected.


Below we have a quick showcase of our finger clamp design, unfiltered ppg signals, poster from Engineering Fest, and our PCB for our wireless charging unit we plan to work on next semester.

First prototype of finger clamp

finger clamp

First time testing sensor on Arduino IDE

proto waves

First poster for presentations at Engineering Fest

engineering fest

Our Spring Semester Summary

Our Plan

After acquiring a better understanding of how the rest of our project will play out, we had a couple things that we wanted to accomplish. First we ordered a pre-soldered wireless charging module that uses the BQ51050b that we can go ahead and connect to our ESP32. Once this was completed, it simplified our design as well as complete our requirement of making our system wireless. Our next task was to improve our data filtration code. Our code now includes a moving average filter and a bandpass filter, which is meant to limit the noise coming from the sensor. After polishing up our code, the next step was testing.

Features that were tested
  • Wireless Charging
  • Battery Operation
  • Bluetooth Capability
  • Data Collection
  • Filter Quality
  • Heart Rate Accuracy
  • Signal Quality Reporting

When testing was completed and our code and prototype were polished the way we wanted it, the last steps were to prepare a poster and presentation for UGRAD! After UGRAD we will be writing an Instruction Manual and our Final Report as well as recieving our Client Evaluation. We feel so grateful for the opportunity to be a part of Professor Nghiem and Professor Nguyen's project and we had so much fun over these last two semesters!


Below we have a quick showcase of our accomplishments from this spring semester

Prototypes over the course of both semesters with the final product in the lower right.


Our working filtration algorithm to eliminate noisy PPG signals

rawPPG filtPPG

Meet Our Sponsors!

We are very excited and honored to be working under Prof. Nguyen and Prof. Neghiem! Here is a little bit of information on each of our professors.

Dr. Truong Nghiem, Assistant Professor of Electrical Engineering

Truong X. Nghiem is an Assistant Professor at the School of Informatics, Computing, and Cyber Systems at Northern Arizona University (NAU). He received his Ph.D. in Electrical and Systems Engineering from the University of Pennsylvania. Before joining NAU, he was a postdoctoral scientist in the Automatic Control Laboratory at EPFL (Switzerland), and a postdoctoral researcher at Penn. During his time at Penn, he was a member of the General Robotics, Automation, Sensing and Perception (GRASP) Laboratory, the Real-time and Embedded Systems Lab (mLab), and the Penn Research In Embedded Computing and Integrated Systems (PRECISE) center.


Dr. Tuy Nguyen, Assistant Professor of Electrical Engineering

Dr. Nguyen is an expert in software and hardware design for post-quantum cryptography and error correction codes

Email: tuy.nguyen@nau.edu


[1] D. Biswas, N. Simões-Capela, C. Van Hoof and N. Van Helleputte, "Heart Rate Estimation From Wrist-Worn Photoplethysmography: A Review," IEEE Sensors Journal, vol. 19, no. 16, pp. 6560-6570, 15 Aug.15, 2019. https://doi.org/10.1109/JSEN.2019.2914166.

[2] S. S. Chowdhury, R. Hyder, M. S. B. Hafiz and M. A. Haque, "Real-Time Robust Heart Rate Estimation From Wrist-Type PPG Signals Using Multiple Reference Adaptive Noise Cancellation," IEEE Journal of Biomedical and Health Informatics, vol. 22, no. 2, pp. 450-459, March 2018. https://doi.org/10.1109/JBHI.2016.2632201.

[3] A. Temko, "Accurate Heart Rate Monitoring During Physical Exercises Using PPG," IEEE Transactions on Biomedical Engineering, vol. 64, no. 9, pp. 2016-2024, Sept. 2017. https://doi.org/10.1109/TBME.2017.2676243.

[4] J. Azar, A. Makhoul, R. Couturier, J. Demerjian. Deep Recurrent Neural Network-based Autoencoder for Photoplethysmogram Artifacts Filtering. Computers & Electrical Engineering, vol. 92, 2021, 107065. https://doi.org/10.1016/j.compeleceng.2021.107065.