Does anybody know if Digi’s XBee radios are compatible with other 802.15.4 / Zigbee radios?
From what I understand, since 802.15.4 is the underlying standard for all Zigbee transceivers, they should be able to talk to one another, but whether the information gets through to the computer or MCU depends then on the networking stack used (ie Zigbee).
But how different are they? Would my Xbee 2.5 modules using the Zigbee firmware be able to talk to something like this?
http://www.ember.com/products_zigbee_chips_e260.html
It seems kind of pointless to have a standard if this can’t be done…
ZigBee needs 802.15.4 as it’s MAC and PHY layers.
802.15.4 doesn’t need ZigBee as its Network layer.
as in
802.3 (Ethernet) doesn’t need TCP or IP.
so, brand A 802.15.4 will by definition communicate with brand B using 802.15.4 (if they agree on certain common configuration settings such as long/short addresses), and if they use MAC addresses for source and destination. Much as can be done in Ethernet for MAC layer messaging (done rarely). Or UDP 802.3 or 802.11 since UDP adds very little to to the 802.3 or 802.11 MAC layer.
ZigBee interoperabilty among vendors is what the ZigBee alliance is all about. Getting that complex network layer which sits atop 802.15.4 to work the same among vendors. In truth, it just barely does. But it’s not important now, since about 80% of all 802.15.4 modules/radios are used with other than Zigbee’s protocols. So far.
Alternatives to ZigBee, and ones that also use 802.15.4 include 6LoWPan (an IETF RFC) and ISA 100.11a (draft).
And with ZigBee you also have the politics of the alliance: logo licenses, resale licenses, Alliance membership dues, blah blah.
But again, in simple non-meshing networks, ZigBee is often not used. A simple DIY network is the usual approach. This is very easy if you know the basics such as broadcast v.s. unicast with MAC addresses, put your own addresses in the data payload and do your own forwarding, retransmissions on timeout, and so on. Not hard, maybe 300 lines of code. I’ve done it with '15.4.
Thanks, it makes a lot more sense to me now.
So would it be reasonable to assume that the Zigbee Alliance will eventually succeed in standardizing the protocol between vendors?
I’m working on building an intelligent controller system for mesh networks, and it would be nice to be able to make it usable with devices using alternate Zigbee chips.
I would try your idea of using my own network protocol on top of 802.15.4, but I think for my purposes I’d need Zigbee for it’s meshing and hopefully standard protocol to communicate with other devices.