LoRaWAN devices are typically simple, they grab some sensor data and deliver it back to the network. By adding some embedded machine learning we can make them a lot more intelligent!
4. Typical LoRaWAN sensor in 2020
Vibration sensor (up to 1,000 times per second)
Temperature sensor
NFC
Water & explosion proof
Processor capable of running >20 million
instructions per second
5. But... what does it actually do
Once an hour:
• Average motion (RMS)
• Peak motion
• Current temperature
6. 99% of sensor data is discarded due to
cost, bandwidth or power constraints
https://www.mckinsey.com/~/media/McKinsey/Business%20Functions/McKinsey%20Digital/Our%20Insights/
The%20Internet%20of%20Things%20The%20value%20of%20digitizing%20the%20physical%20world/The-Internet-of-things-
Mapping-the-value-beyond-the-hype.ashx
9. On-device intelligence is the only solution
Vibraon pa-ern
heard that lead to fault
state in a weekTemperature
varies in a way that
I've never seen
before
Machine
oscillates different
than all other
machines in the
factory
10. On-device intelligence is the only solution
Temperature varies in a way
that I've never seen before
(0x1)
11. Machine learning is great at finding
pa1erns in messy data
(anything you can't reason about in Excel)
13. TinyML
Inspired by "OK Google"
Focus on inferencing, not training
Machine learning model is just a mathemacal funcon with lots of
parameters
Accuracy vs. speed, reducing parameters, hardware-opmized paths
Targeng ba-ery-powered microcontrollers
Pete Warden
Neil Tan
14. What is it good for?
https://www.flickr.com/photos/oceanyamaha/7091324605
Recognizing sounds Detecting abnormal vibration
https://pixabay.com/photos/washing-machine-wash-cat-4120449/
Biosignal analysis
https://www.flickr.com/photos/sheishine/16696564563
Anything with messy, high-resolu9on sensor data
22. Conclusions back to TTN
♻
Sample for four seconds
Classify
Result differs? Message.
Sheep is walking
h-ps://pixabay.com/photos/sheep-curious-look-farm-animal-1822137/
26. Edge Impulse - TinyML as a service
Embedded or edge
compute deployment
options
Test
Edge Device Impulse
Dataset
Acquire valuable
training data securely
Test impulse with
real-time device
data flows
Enrich data and
generate ML process
Real sensors in real time
Open source SDK
Free for developers: edgeimpulse.com