Breeze Environmental Platform

Making Monitoring Environmental Conditions a Breeze

Our third-place winner (line-powered category) in the Challenge Climate Change contest was Attila Tőkés, who created an extremely versatile environmental sensor platform he calls “Breeze.” This platform features a carrier board which can connect multiple sensor boards, a communications module, the QuickLogic QuickFeather board, and a solar panel/charging system along with a Li-Ion battery. One of the most clever aspects of his system is that different sensor boards and different communications modules can easily be swapped in and out to create a fully customizable environmental sensing platform. 

Another clever aspect of his design was to use the SensiML Analytics Toolkit to implement Machine Learning so that his system can pre-process sensor data prior to sending it on to a larger network or to the cloud. This approach reduced the amount of communications bandwidth and power consumption his platform requires to operate efficiently and effectively.

His prototype board was designed to accommodate multiple Mikroe Click Board sensor modules.

They offer an amazing number of options, all using the same connector. Attila chose a UV sensor and an air quality sensor. These were complemented by a third lightning sensor module (also mounted on the carrier board), as well as the mCube accelerometer and Infineon PDM digital microphone and temperature/pressure sensors already available on the QuickFeather board. The combination of all those different sensors gives the Breeze the ability to monitor and analyze a wide range of environmental factors and the flexibility to easily adapt to new or changing requirements.

The communications module was also from Mikroe, and again multiple options are available from the same company in the same connection footprint – making it easy to change should the need arise. In Attila’s implementation he chose a Wi-Fi module to send locally pre-processed data to cloud-based applications. 

In his example application, he trained his system to recognize the vibration patterns associated with earthquakes using the Machine Learning capabilities built into the SensiML Analytics Toolkit, and then used the accelerometer on the QuickFeather board to collect vibration data. QuickFeather’s EOS S3 analyzed the data using AI, and then the processed results were sent to the communications module.