Slashdot Mirror
ZigBee Wireless Standard Ratified
ductormalef writes "Today, the ZigBee Alliance announced the release (pdf) of version 1.0. ZigBee is a standard for low data-rate (250kbps max) wireless personal area networks (WPANs). It utilizes the IEEE 802.15.4 hardware and MAC layers which utilize frequency bands at 898MHz, 902-928MHz, and 2.4GHz. ZigBee supports mesh networking and claims to be 'wireless control that simply works.' They claim to be a solution to everything from wireless home automation to industrial control."
ZigBee supports mesh networking and claims to be 'wireless control that simply works.' They claim to be a solution to everything from wireless home automation to industrial control.
We'll see how this works. The last factory we worked in, we had to use fiber (10MB at that) because cables would have too much interference.
"ZigBee: Wireless Control that Simply Works"
From my days in compsci classes, anything that simply works usually isn't working at all.
A new door opens in the world of aerial communism...
Think about it. Blades became the rage a while ago. Blade = sharp. Bluetooth. Tooth=sharp. I always liked firewire just because of the name. IEEE1394 is just not the same.
ZigBee. Well, I guess Bee=stinger=sharp, but that is stretching it. Especially with a nonsensical "Zig" thrown in.
This might sound funny, but the name is the thing, especially in corporations.
Yeah, right!
"-1 Troll" is the apparently the same as "-1 I disagree with you."
All I could think is that I'm allergic to bee stings.
It's essentially a wireless networking scheme that layers on top of an independant physical platform, yet costs significant dough to get certified for. Very clever scheme. Too bad they haven't included really interesting things in their design. All it lays out is the full node/slave node/coordinator node network. It really should have things like dynamic reconfiguration of the network structure. I think it's around 7500$ to become a 'zigbee partner' and then another indeterminate amount to become zigbee compliant/certified. That doesn't even include the royalties for using the stack commercially. The underlying hardware interface however... is very interesting.
I'm also not sure I want my home devices on an unauthenticated wireless network.
A spread-frequency digital communications system is really useful (802.15.4 standard). It also doesn't have the associated royaly issues.
WPAN.....isn't this the job of Bluetooth ? Great, not only we have HD-DVD vs. Blu-ray, and PSP vs. DS, UMTS vs. WiMax, NOW we have to worry about Bluetooth versus ZigBee!!
Thanks, but no thanks. I'll happily keep my BT appliances.
Vendors shipped 165 million cell phones worldwide in the third quarter of 2004. In-start/MDR predicts 653 million units to be shipped this year. So, even by 2004 numbers, Zigbee will be in less than 1% of new cell phones shipped next year if they hit their target. Bluetooth, on the other hand, ships two million units per week in various devices. Perhaps it "will be big", but you need far stronger numbers to back up your prediction.
Heile says it'll be "on target for 5 million units"? Your own article reports that he also said "analysts are predicting between 5 million and 50 million Zigbee devices in the first year", which means Zigbee might make the low end of predictions.
Also, $5 per unit is a huge cost for cell phone vendors. Nokia, for instance, would have to pay over $1 billion a year (~200 million units, excluding engineering costs) to support this in all their phones. To put that number in perspective, that's about a good quarter's worth of net profit for Nokia.
In other words, like any new technology, it will become much cheaper with wide adoption, but it will not be widely adopted unless it's cheap. Its future may be interesting, but is by no means assured. I simply don't see the evidence for your optimism.
The deal is you can't accomplish the same thing with a Bluetooth radio. Bluetooth radios are legacy frequency hoppers. The two major problems they have are poor receiver sensitivity and slow network synchronization times.
Receive sensitivity matters because half your range is in the last 3dBm. Or better most of you're quality of service is in the last 10-30 db. Bluetooths receiver gain (-70dBm) is horrid as a result of the coding scheme used. Also because Bluetooth is a frequency hopper, once a device powers down it quickly looses synchronization with the network. It can then take a second or two for it to resync when it powers back up. This totally bytes for mice, keyboards and low power sensor networks.
Zigbee uses a direct sequence spread spectrum radio which is very nizely done. Radios designed around the spec can have high (-100dBm vs -70dBm) receiver sensitivity which translates into longer range and and high quality of service than you can get with Bluetooth. And if the device you are trying to talk to is continually powered up, you can start transmitting information immediately, which is good for sensor networks and battery powered devices.
Also the reason R&D is dropping on Bluetooth is that the technology was at a dead end from the start. Once you have functional silicon there isn't anything you can to to improve the performance. Bluetooth also is being squeezed from the top by WIFI which gives you literally 100 times the performance at double the cost and Zigbee which does pretty much everything that Bluetooth was supposed to do at a quarter the cost. Granted there are no Zigbee devices out there yet, but hardware development has been proceeding in parallel with the standard itself.
Finally the NRE cost of using a Zigbee solution will be a fraction of the cost of Bluetooth. That isn't very important for consumer devices which are made by the million, but is very important for industrial devices which are made in the tens of thousands and everything needs to work together for real. (I mean who cares if your Nokia wireless headset works with your Motorolla phone? Motorolla certainly doesn't want it to)