It is currently Thu Aug 22, 2019 3:02 am

Post Reply  Page 1 of 1 [ 1 post ] 
Author Message
PostPosted: Tue Dec 03, 2013 4:36 am 

Joined: Sat Jul 13, 2013 7:27 pm
Posts: 7
I'm a home automation enthusiast, openHAB contributor, hobbyist and a dreamer. Some months ago I started to think about building my own home automation wireless nodes and to start a business around them. I started a research of the best and the most promising software/hardware technology available. And so I found DASH7. It seems to be exactly what I have looking for. At least on paper.

We think it is a great technology if you need good indoor coverage, and you don't have needs to transfer a lot of data. The other technologies that you might consider for home automation are WiFi, Bluetooth, and maybe some 802.15.4-based solutions. But really, I think WiFi and Bluetooth and better choices than 802.15.4. The downside of all of these technologies is poor coverage. WiFi is also power-hungry and expensive, which is a problem for some things but not others. The advantages of WiFi and Bluetooth is that there is billions of dollars of product development behind them, so even a software engineer can build a hardware solution using WiFi or Bluetooth. DASH7 is a smaller technology so far, with much less money behind it. One way to look at this problem is that there is very little competition, though, so if you want to start a company to do home automation products, it might be a better opportunity than WiFi or BT.

What I look for is an open source standard that is relatively widely used (or has a potential for that). So far candidates are ZigBee IP, 6loWPAN with CoAP, S.W.A.P. (PanStamp). I would like to use CC430 family chips as I think they have great price/value ratio and are made to work in sub-Ghz, which is essential for me.

We will support the PanStamp HW as soon as we have some time and we can get a few boards. CC430 is not a great hardware platform, but OpenTag plus some APIs can fit on it. Even OpenTag + CoAP, we think, can fit on a CC430. None of the other "big-player" stacks can fit on CC430, so if you like it you are basically limited to proprietary or OpenTag/DASH7.

More on ZigBee IP and 6LoWPAN:
You should try these yourself and build your own opinion, but I think they are failures. They do not actually use IP, so you need a gateway just like with anything else. The performance is poor at 2.4 GHz, and there is not a lot of standardization. ZigBee is better standardized, so it is a bit easier to set up. DASH7 has some capability for IP translation as well, so practically there is not much difference between ZigBee IP, 6LowPAN, or DASH7 if you are looking from the perspective of a network.

More on CoAP:
CoAP is an application protocol. It can run on any system that is compatible with UDP. DASH7 is compatible with UDP, so it can run CoAP. We have not ported CoAP to OpenTag yet, but I researched CoAP and I know it can be done without too much trouble. But first, we will need to see if there is a good reason to support it. So far, no one is asking for it. I like it, though.

More on SWAP (PanStamp proprietary protocol):
SWAP is very simple. I don't know if it is suitable for commercial/industrial requirements, though. I am hoping that the PanStamp guys decide to migrate to DASH7 in the future, which supports all the features they want to achieve for SWAP2 (and more). In any case, if you want to use PanStamp HW, we will be supporting it in OpenTag fairly soon, so if you want to do some prototyping using PanStamp and SWAP, that is probably a nice place to start.

So, now DASH7. It seems to be very similar to 6loWPAN ideas. I read a lot about it, but still I have some questions and I would be very happy if you find some time to answer them:

The major philosophical difference between DASH7 and 6LP is that DASH7 was designed with performance requirements (low power, long range, low latency) and 6LoWPAN was designed with interface requirements (translate IP to an 802.15.4 MAC). I would go so far as to say this is a typical philosophical difference of electronics engineer vs. software engineer.

1. I would like to know what is the adoption of DASH7 so far? I see there is a certificate program, but I didn't found any consumer products so far. Is it only used by military?

There are some pilots in Europe, and there are a few universities in EU that are building indoor location systems with DASH7. The big consumer product for DASH7 is HayTag (, which is available in the late winter 2014.

2. I didn't quite got the application layer idea. The WIKI says "no more profiles", but instead it expose some kind of a file system. Can you point me to some documentation and/or example of this?

The wiki is the main source. We will be releasing a specification for the upper layers in about 4 or 5 weeks. We are doing stress testing on them right now, so after that is done we can release the spec.

An "application profile" in ZigBee or Bluetooth is a codeword (or just some identification) for a system of parameters and logic that affect the way the system behaves. The implementation of application profiles results in a lot of code, which is one reason why ZigBee and Bluetooth implementations use a lot of code (they are bloated). DASH7 simply standardizes the way that these parameters are stored, what they do, and the ways to change them. It is implemented with a filesystem, because virtual addressing and user privileges become necessary when implementing on different devices and for security.

3. Does it fit on CC430F5137? Is there a port for that chip?

OpenTag started on CC430. We are working on the 0.5 version with Cortex M as the primary focus, but we will backport all the 0.5 improvements to CC430 when we start supporting PanStamp. There should be about 8KB flash and 1.5 to 2KB SRAM free for sketches. It's not a lot, but all you really need is some API glue, so I think it is actually plenty.

Right now I prefer the STM32L + SPIRIT1 platform. HayTag uses this platform, and we will be selling some OEM kits and dev kits. CC430 might work for you, but it uses a lot more power and it has worse packet loss, especially at medium range and indoors.

4. Is there any home automation software that speaks DASH7?

Home automation is not my focus, so I don't know. Software talks to APIs, so I don't think it is a big challenge to integrate some popular API. We have a low-level API we use, so porting to a high level API is just writing some wrapper functions. Maybe you can tell me what you think is a good API to look at.

The serial interface for OpenTag is a packet-based protocol called MPipe. It is quite basic, but it is not TTY (it is packet based). We have a Qt program called OTcom that can interface with OpenTag, but one thing I'm working on is a plain POSIX Std-C client.

Display posts from previous:  Sort by  
Post Reply  Page 1 of 1 [ 1 post ] 

It is currently Thu Aug 22, 2019 3:02 am

Who is online

Users browsing this forum: No registered users and 1 guest

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to: