College - Author 1

College of Engineering

Department - Author 1

Computer Engineering Department

Degree Name - Author 1

BS in Computer Engineering

College - Author 2

College of Engineering

Department - Author 2

Computer Engineering Department

Degree - Author 2

BS in Computer Engineering

Date

6-2026

Primary Advisor

Kun Hua, College of Engineering, Electrical Engineering Department

Abstract/Summary

This report goes over the wearable health monitoring system that we built that uses an activity classification system to tell what you are doing real time. We used a Shimmer3 sensor to collect accelerometer data and streamed it via Bluetooth to our laptops where we processed the data and ran a machine learning classifier to detect if someone is resting, walking or running. We collected our own data to train the Random Forest classifier that we used from scratch and ended with 98% accuracy. The system has a live dashboard implemented into it that shows sensor data and the current activity alongside a confidence percentage. This project shows that someone can build an activity recognition system with limited hardware and time. With more time a classifier similar to one used in an Apple Watch or Fitbit can be made.

Share

COinS