AMIS Mobile SIG – rain sensors pouring data into the mobile network

Last night, the AMIS Mobile Special Interest Group convened. This session had as objective to explore the many different mobile agents and devices that we encounter, the types of signals, information and functions that are relevant with these various agents and the impact their inclusion in the mobile sphere of our customers has on the enterprise applications and architecture.

Our guest speaker – Rolf Hut of the Faculty of Civil Engineering and Geosciences – did an excellent job of introducing a new class of mobile agents with his presentation on hydro-meteorological stations. He even had us all build rain-sensors, from easily accessible and very cheap materials. 10 minutes of super-glueing, soldering and kindergarten-level handcrafting was all it took for 6 teams to create a sensor that could be connected to a laptop to register simple signals.

Image

From this very practical example, it was not hard to envisage a wide network with hundreds or thousands of devices that report findings – to be processed by the backend infrastructure.

Image

See Rolf presenting on YouTube: http://www.youtube.com/watch?v=Nh7GDD3Ssr8&feature=player_embedded#at=44.

Rolf represents the Tahmo project – http://www.tahmo.org/ – Trans-African Hydro-Meteorological Observatory. This project attempts to distribute some 20000 weather stations all over Africa – located every 30 km, ideally one near every school. These weather stations will measure rain fall, sun radiation and other meteorological values.

As Thamo’s website states: “Monitoring Africa’s environment is an important challenge if the continent’s resources are to be used in an optimal and sustainable manner. Food production and harvest predictions would profit from improved understanding of water availability over space and time. Presently, the African observation network is very limited. National governments and regional planners do not have the data to make proper decisions regarding investments in water resources infrastructure.

The idea behind this project is to build a dense network of hydro-meteorological monitoring stations in sub-Saharan Africa; one every 30 km. This asks for 20,000 of such stations. By applying innovative sensors and ICT, each station should cost not more than $200. The stations would be placed at schools and integrated in the educational program. The data will be combined with models and satellite observations to obtain a very complete insight in the distribution of water and energy stocks and fluxes.”

The distribution of weather stations across the world would drastically change once the Tahmo plans have been realized:

Image

The rain sensor is a fairly simple component. It is designed with several constraints in mind: low (or no) maintenance requirements, no moving parts (like the common tipping bucket sensor has), no slits, gaps or other open areas. And low cost. The cosy ‘raindrops on the tent canvas’ are one way to represent the concept of the sensor: using a piezo electric element that translates pressure (including sound waves) into an electrical signal (simply put), the sensor will register raindrops (pressure) and translate them to a small signal. A simple electronic circuit can process the signal and put on the wire(less); a simple copper wire acts as antenna and an HTTP Post request containing a single byte of data that describes the raindrop’s size is sent to the base station.

When it comes to handling the signals from the sensors – the mobile agents in this example – as well as signals and interactions from various other types of devices, the enterprise architecture is extended and faced with several interesting challenges.

Image

This picture shows some of the device types that we are dealing with. It also illustrates the types of interactions – from fire and forget reporting (like the sensors do), high volume and frequence, typically low data content, through retrieve (query) and update (lower frequency, higher data content) to push and alert (where the back end pushes out messages to devices that have registered for such messages).

The enterprise back end needs to be prepared for these challenges, in terms of volume and load, availability, security, data formats and communication protocols, relevant, appropriate, enriched content, performance and scalability. And integration with the traditional enterprise resources is required – although the timing (synchronous/asynchronous) and level of detail (micro-level or aggregation only) need to be determined.

For the case of the rain sensors, it would seem that individual drops do not need to be recorded in a database. The (near) real-time continuous processing and the desire to record aggregates derived from the individual signals suggest the use for example of Complex Event Processing. I think we will further explore that avenue.

This session was altogether useful and inspiring for our effort in the Mobile arena. Both functional opportunities, non-functional questions and challenges and impact on and integration with enterprise architecture came together. Thanks Rolf for guiding us and thanks Rutger for putting the session together.

Resources

Arduino – http://www.arduino.cc/ – Arduino is an open-source electronics prototyping platform based on flexible, easy-to-use hardware and software. Arduino can sense the environment by receiving input from a variety of sensors and can affect its surroundings by controlling lights, motors, and other actuators.

Instructables – http://www.instructables.com/ – Instructables is the Biggest How To and DIY community where people make and share inspiring, entertaining, and useful projects, recipes, and hacks. Contains many examples of constructing sensor-configurations using for example Arduino or the WII mote controls

Gap Minder – http://www.gapminder.org/ – Gapminder is a non-profit venture – a modern “museum” on the Internet – promoting sustainable global development and achievement of the United Nations Millennium Development Goals. Gapminder exposes statistical data on many aspects of humanity, across countries and regions.

Tags:,