https://za.pycon.org/talks/how-to-submit-a-talk/
Sensor data processing on microcontrollers with MicroPython
Being able to measure physical phenomena is critical to many areas of science and engineering; from measuring pollution in public spaces to monitorning of critical machinery. Over the last decades the capabilities and costs of sensor system has become much better, driven by improvements in microprocessors, MEMS sensor technology, and low-energy wireless communication. Thanks to this Wireless Sensor Networks and "Internet of Things" (IoT) sensor systems are becoming common.
Typical sensor nodes use microcontroller-based hardware, and the firmware developed primarily using C (or C++). However, it is now becoming feasible to write microcontroller firmware using Python. This is thanks to the MicroPython project, combined with affordable and powerful hardware from the last couple of years. Using the familiar and high-level Python programming language makes the process of creating sensor nodes more accessible to an engineer or scientist.
In this talk we will discuss developing microcontroller-based sensors using MicroPython. This includes a brief introduction to MicroPython, how to do efficient data processing, and share our experience applying this to process accelerometer and microphone data, using both Digital Signal Processing and Machine Learning techniques.
Intended audience: Any developer or data scientist curious about sensor data processing, IoT, and how Python scales down to the smallest of devices.
Related open-source project: https://github.com/emlearn/emlearn-micropython
Remote Talk: A conference talk that will be presented remotely. Talk slots are 30 minutes long, with an additional 10 minutes for discussion at the end.
- Data Science: Talks on using Python to process and extract meaning from data.
- Scientific Computing: Using python for scientific computing applications, not covered by the Data Science track
Jon is a Machine Learning Engineer specialized in IoT systems. He has a Master in Data Science and a Bachelor in Electronics Engineering, and has published several papers on applied Machine Learning, including topics like TinyML, Wireless Sensor Systems and Audio Classification.
These days Jon is co-founder and Head of Data Science at Soundsensing, a leading provider for condition monitoring solutions for commercial buildings and HVAC systems. He is also the creator and maintainer of emlearn, an open-source inference engine for microcontrollers and embedded systems.