Real-Time Signal Processing for Wearable Electrocardiogram Device

Technologies Used

• We used nRF51 with BLE 4.0 support and Cortex-M0 core as MCU in this project.
• FreeRTOS was used for firmware development.
• The firmware was implemented using C/C++.
• Microsoft Visual Studio with Visual GDB plugin was used for firmware implementation.
• The android application was implemented in the Android Studio using Java language.
• The device has a microphone, ECG, accelerometer, and a thermometer sensor

How it works

Participants in the experiment (children and adults) are equipped with a wearable ECG device placed on their skin. As they go about their day, electrical impulses released by the heart are picked up, processed by the microprocessor and recorded. At the end of the day, the participants come back to the lab where the researchers take off their ECG devices and study the heart's electrical activity in response to environmental noise, including from road traffic and aircraft.

Challenges solved: Real-time data processing

The main job of the firmware is to enable data transfer from the wearable ECG device to Android smartphones via Bluetooth 4.0+ in real time. As soon as the data arrives at the smartphone, the GPS coordinates of the phone are added to the data package. The device collects the data from the following sensors:

• ECG: 250Hz
• Accelerometer: 30Hz
• Thermometer: 1Hz
• Microphone: 11,2kHz

We needed to implement a continuous data transfer. To make it possible for the ECG device to keep the data when the connection with a smartphone is lost, we implemented temporary data storage in the flash memory.

Another major tech challenge was related to the data transfer speed that Bluetooth channels could handle. The required speed was 2MB/sec. But after we made our calculations, the realistic speed appeared to be lower – 2-3KB/sec in that particular case. To optimize the data transfer rate our team implemented compression achieved by encoder and decoder.

Result

With the wearable ECG device and a user-friendly mobile app, it became possible for scientists to perform their research and investigate the effects of noise exposure on children and adults’ blood pressure and heart health.

Integra's team helped us clarify our requirements and change systems architecture to achieve the final goal. I'd say that the flexibility and enthusiasm of both engineering and management teams throughout the whole project were some of the things we value most in our collaboration.

Dr. Sam Wass, University of East London