34
\$\begingroup\$

It's easy to get an introductory explanation of Zigbee.

But technical information on XBee seems harder to find. It's unclear to me what its similarities and differences to Zigbee are.

\$\endgroup\$
38
\$\begingroup\$

ZigBee / ZigBee Pro are mesh communication protocols that sits on top of IEEE 802.15.4 PHY.

XBee / XBee Pro are product names for radio communications modules made by Digi.

The modules can be loaded with various firmwares to support ZigBee / ZigBee Pro / DigiMesh and come in several frequency bands.

DigiMesh is an alternative to ZigBee that changes a few things, and adds some features to make it generally better to work with.

But, you sacrifice compatibility with ZigBee devices.

For example DigiMesh allows routers to sleep, has lower overhead, has 1 node type vs zigbee's 3 leading to a more robust mesh, can run at higher data rates, etc.

Frankly its a better protocol all around, in my opinion. I wouldn't use ZigBee unless i needed interoperability with other vendor's nodes. Not that DigiMesh is the only answer, there are other mesh protocols that sit on IEEE 802.15.4 as well.

\$\endgroup\$
  • 5
    \$\begingroup\$ I retract my post for your much more well explained answer. \$\endgroup\$ – Kellenjb Oct 18 '10 at 3:43
  • \$\begingroup\$ Is there any documentation available for the radio protocol used in AT (serial) mode? Is it some new ZCL cluster running on ZigBee? Or perhaps raw 802.15.4? \$\endgroup\$ – Toby Jaffey Oct 18 '10 at 11:10
  • \$\begingroup\$ I haven't used the XBee Pro dev kit i have for awhile but as i recall the AT mode was just a way to communicate between 2 XBee modules in a very simple mode that mimicked AT modem communications. I assume the radio PHY would still have to be something like 802.15.4 as i would think most of that decode is in hardware, but the rest of the protocol is likely something Digi worked up, proprietary i imagine. \$\endgroup\$ – Mark Oct 18 '10 at 19:04
  • \$\begingroup\$ Have you tried using the synchronous sleep mode, and have you found it useful? Conceptually, having the modules automatically coordinate when they wake up would be great, but requiring that the duration of each waking interval be fixed seems to be both inefficient (in cases where nobody has anything to say) and bothersome (in cases where nodes have more to say than will fit in one wake-up interval). \$\endgroup\$ – supercat Nov 20 '12 at 19:27
  • \$\begingroup\$ I agree that ZigBee is very complicated (I have first-hand experience of that!), but saying that DigiMesh is generally better sounds a bit subjective to me. \$\endgroup\$ – clabacchio Jan 20 '15 at 10:08

Not the answer you're looking for? Browse other questions tagged or ask your own question.