Hardware Reference
In-Depth Information
Z-Wave
Z-Wave was created by Zen-Sys, founded in 1999, as a proprietary protocol for controlling products wirelessly in
the home automation space. In 2008 the company, and the technology, was taken over by Sigma Designs who
still manufacture and supply the lions share of the chips. This has meant that all the manufacturers producing
Z-Wave branded equipment must buy from Sigma (or their affiliate, Mitsumi) and all have to go through a testing
and verification procedure to be allowed to brand their product as “Z-Wave.” These vendors are part of the Z-Wave
Alliance, and currently total 200 manufacturers with around 600 different products.
System Design
The communication channel used in Z-Wave is a 900MHz RF signal (which varies slightly according to your
geographic location), transmitting around 40kbits/second. (Although newer devices with the 400-series chips can
support up to 100kbs.) While the signal strength suffers the usual vagaries of walls and floors, each device has a range
of up to 20-30 meters, and can be connected in a mesh formation, so that one device can repeat messages to the next,
thereby extending the range. As well as sending messages between themselves, these devices also communicate
with a hub, or primary controller, which is used to connect them to a wider network, either via a ethernet router, or
computer. Typical controllers include the Tricklestar and Z Wave Aeon USB adapters, which requires a PC, and the
VeraLitz Z Wave Controller which doesn't. There is also a RaZberry Z Wave controller which connects directly to the
GPIO controller of your Raspberry Pi, if you wish to adopt that path.
When each new Z-Wave device is added to the system, it is paired with a nearby controller that determines the
signal strength between them, which is then used in the routing algorithms. This makes it possible to send signals to
devices that are placed underground and in garages, provided there is at least one device that is in range of it, and the
rest of the network. However, with such a low data rate, it is easy to overload nodes if you try to force them into acting
like repeaters.
For security, Z-Wave relies on this pairing arrangement, since buttons are pressed on each unit in a similar
fashion to Bluetooth pairing to ensure each unit may talk to another. It is far from ideal, but given the comparative low
uptake the chance of a man-in-the-middle attack is rather low.
Like X10, Z-Wave has a house code (which is called a Network ID), and a unit code (its Node ID) for addressing.
These allow for 232 different nodes on a single network which, although slightly fewer than the number available with
X10, is enough for the majority of houses, although it is not difficult to bridge between networks, should the need arise.
Z-Wave's biggest advantage has always been compatibility, since the interoperability labs at the Z-Wave Alliance
ensured that virtually all devices could communicate with every other. This closed shop, however, ensured that no
open source software could work. But that has now changed . . .
Bypassing NDAs
Because every member of the Z-Wave Allinace must sign an NDA and confidentiality agreement to access the formal
specification and whitepapers, it was impossible for open source developers to build a true Z-Wave device—until
hackers started to reverse engineer the protocols! There are now a couple of options.
Open Z-Wave
With such a direct and typically open source name, it's no surprise that this project has garnered a lot of attention. It
comes in two main parts, the library, and the web-based configuration interface, which comes with its own mini web
server. The library depends on
libudev
, which depending on your system may need to be explicitly installed.
For example:
apt-get install libudev-dev
Search WWH ::
Custom Search