Being on vacation when the announcement of the winners went out means that I'm about to a late start. But on the other hand that alows me to start with fresh energy into this design challenge...

In contrast to the last one, this time the focus will be more on the software side of things.. So lets start with what I intend to work on:

My journey into home automation, smart home and whats now known as the Internet of things started already several years ago. When we moved into our new house, one of the questions that regularly came up was "we want to leave - did we forget to close one of windows?". So I wanted to have a central place to look this up. Together with the need to have a (central) overview about all the appointments of a given day, tasks to do this lead me into my first project - "The Wall" (see It started as project for the NXP mbed design contest. Unfortunately it proved to be unstable, and one of the APIs I used back then was discontinued - that meant it did never go into production. To solve at least the "family information" problem, I just mounted an Android tablet onto our kitchen wall, where it now serves as information central (appointments, tasks, weather, tram schedule and the occasional Google lookup during dinner).

But this did not solve the 'Smart home' problem. I still didn't know whether the windows are open or not (and installing sensors throughout the house scared me off - laying so many cables has a really low wife-acceptance-factor). But an incident we had a while ago led me to another project. When we where away for our skiing vacation, our heating system stopped working while there were -20°C outside, and our house nearly froze down (we came back when it was at about 1°C), and nothing broke. To have better chance at handling this situation better in the future (its difficult to ensure it won't fail again, since the culprit most likely was fluctuations in the gas line) I started a project to monitor to temperature online: . This allows me to see any problems and have somebody handling them even while being on vacation.

So this was another step down the road to a more intelligent home - but I still didn't solve the original problem. I still check the windows before leaving the house...

So this challenge will be the next step to move towards a smarter home. I would like to use the EnOcean reed switch sensors to monitor the most important (if not all) windows in our house to detect which ones are open (so I will probably use the budget to buy a whole lot of them). Two features are important for me: first, they are wireless so no cables are needed (this increases the Waf). Second, they run with energy harvesting,so I never have to change batteries another bonus point. So they are whats needed to equip a whole house with sensors.

The RaspBerry Pi should serve as a middle-man between the sensors and the tablet. Since I want to have a very special view of the sensor data (just a quick glance whether there are any windows open) which should be shown as widget on the home screen, I need to investigate my options. Maybe I can use the OpenHAB Android software directly, maybe I can use OpenHAB as middle-ware. If not, I will use the EnOcean SDK to write my own software running on the RasPi.

So the next steps will be to get everything running for some first tests - fortunately there are already some excellent blog posts about that. Having that done I need to do some range tests to determine whether I can use a central EnOcean receiver or need multiple ones.