Table of Contents

Introduction

Hi,

the project idea I pitched for this challenge is one of wireless, self-powered sensor nodes reporting to a central unit which can, if needed, trigger air filtering systems in specific rooms/areas.

I’ve worked with Arduino, Raspberry Pi and BeagleBone Black before, but not with the TI Launchpads. This seems to be the perfect opportunity to get started with them and the associated tooling such as Energia.

Remote sensor node

Self-powered

The Launchpads consume very little power. By having a small solar power charging circuit, the Launchpad’s battery pack could be recharged during the day.

In order to validate the idea, I would first use off the shelf modules for the charging circuit, more specifically the "USB / DC / Solar Lithium Ion/Polymer charger” from Adafruit.

Next, I would try using the Würth Electronics kit to build my own circuit, gradually learning more about the different components involved.

Sensors

The Launchpads will be equipped with some sensors (e.g. air quality an dust) by means of a custom BoosterPack. This BoosterPack will be designed using Eagle and its design shared with the community.

Case

The different components will be tucked away in a case which can be attached to windows by means of suckers, ensuring the solar panel is facing the outside.

I plan to design and 3D print this case for every remote node. I already have some red ABS available which will suit the Launchpad’s color very well.

MQTT

The remote nodes will report the measurements from the different sensors back to the central unit using a lightweight protocol, such as MQTT.

I am not familiar with MQTT yet, but I have seen some interesting resources lately on how to use it with the Launchpads, which would be a great introduction to this technology. Different approaches could be used for reporting: only report when a value changes, or periodically report all values.

Central unit

The BeagleBone Black will be at the heart of the central unit, gathering all sensor data from the remote nodes and from one locally attached node (MSP-EXP430FR5969). The sensor data will be aggregated and made available locally via openHAB, but also online via Sierra Wireless’ AirVantage.

Network

The central unit will access two networks:

- on one side the public internet, to upload data and make it accessible remotely

- on the other side, a private network with access limited only to the remote nodes

By splitting the networks, the remote nodes will not be directly connected to the internet. Should internet access be required for the remote nodes at any point, firewall policies and routing could be adapted on the BeagleBone Black in order to ensure this access.

Display

An LCD touch screen will ensure local access to the available data with openHAB. It is possible to easily create a UI with drill-down capabilities using openHAB.

I have used openHAB in the Forget Me Not challenge and found it incredibly easy and intuitive to work with!

OpenHAB and AirVantage

I plan to use openHAB for:

- local access to data

- visualisation via charts

- monitoring of the remote nodes (i.e. are they still alive ?)

I am unfamiliar with the AirVantage solution from Sierra Wireless, but it does seem to offer powerful tools to interpret data that has been pushed to it. This will be a good opportunity to try it out and compare it with other tools such as openHAB.

Filtering system

The filtering system would consist of three elements:

- a remote controlled power socket

- a fan, connected via the remote controlled power socket

- a filter (HEPA, …) attached to the fan

By remotely controlling the power socket, the fan can be turned on and off. Ideally, speed variation should be possible.

I’d like to test this by using dimmable power sockets. Should this not work, alternatives will be searched and tested.

Conclusion

This challenge will be a great opportunity to work with hardware and software I haven’t worked with before (Launchpads/AirVantage/...) and share my experiences about them.

I'm looking forward to a challenge with lots of sharing, learning, experimenting and having fun!

Frederick