Wireless USB Specification Officially Released - 1st Devices Expected End of 2005

[Ekkie Tepsupornchai]

<div class='os_post_top_link'><a href='http://www.usb.org/press/pressroom/2005_05_24_USBIF2.pdf' target='_blank'>http://www.usb.org/press/pressroom/...5_24_USBIF2.pdf</a><br /><br /></div><i>"[May 24, 2005] - The Wireless USB Promoter Group today announced the completion of the Wireless USB specification. &lt;...> 'We've delivered on the promise we made in 2004 to complete the specification and make it publicly available in 2005,' said Jeff Ravencraft, chairman of the Wireless USB Promoter Group. 'Product development is currently underway, with the first Wireless USB products expected to be available at the end of 2005. The first implementations will be in the form of discrete silicon that is being introduced in a number of form factors such as add-in cards, external adapters and embedded silicon modules for integrated solutions.'"</i><br /><br /><img src="http://www.pocketpcthoughts.com/images/web/2003/ekkie_1444-dongle.jpg" /><br /><br />It's amazing how dependent I've become over the past several years on USB. I have USB hubs both at home and at work (in addition to the USB ports built into my PC and laptop). I can tell you that ALL ports are occupied (Ext HDD, keyboard/mouse, BT adapter, printer, scanner, APC, touchpad, PPC cradle, and more)! Making USB wireless would highly welcomed. At best, I don't expect this to be universally mainstream for another few years at least, but there a large number of manufacterers involved which gives me great hope that this can fill the void that Bluetooth has been unable to fill. What are your thoughts?

[ricksfiona]

I say 'bring it on!'. All the ports on my USB hub are taken up too.. Let's just hope that transfer speeds remain the same.

[Jason Dunn]

I'm looking forward to seeing this spec develop - I have a grudge against seeing cables, so this would allow me to get rid of a lot of cables. Let's hope it gets traction faster and is easier to use than Bluetooth!

[foebea]

lets squeeze one of those into an SD card!

[Jerry Raia]

Could THIS be the death of Bluetooth then??

:devilboy:

[Jason Dunn]

lets squeeze one of those into an SD card!

Maybe, but if Wireless USB is to succeed, it will need to be integrated at the silicon level on all sorts of devices. Though there will certainly be USB-based adaptors for older devices, and who knows, perhaps SD cards for PDAs.

[djdj]

Sounds nice, but there is one problem, though... Wireless USB = batteries or AC power for every device, since USB currently supplies power to a lot of devices. So we still haven't gotten rid of all the cables.

[Ekkie Tepsupornchai]

Could THIS be the death of Bluetooth then??

:devilboy:
Ha!

I would expect BT will continue to survive, but mostly for specialized purposes such as BT headsets, which have become much more mainstream over the past year. For all other uses, BT simply hasn't gained mainstream acceptance.

If wireless USB makes the "pairing" process much more seamless (similar to how wireless keyboards and mice connect to their receivers), then I can see Wireless USB smashing BT into pieces within PC / Laptop / PDA markets.

Today, BT is just too complicated for widespread use. For headsets, it works great. For nearly anything else, you have to worry about too much (e.g. matching BT service profiles between devices... juggling the 1,000 variations of drivers used to support different BT adapters).

I can't wait for Wireless USB to develop!

[Ekkie Tepsupornchai]

Sounds nice, but there is one problem, though... Wireless USB = batteries or AC power for every device, since USB currently supplies power to a lot of devices. So we still haven't gotten rid of all the cables.
I don't think it'll be as big of a problem as it sounds. You're really only talking about devices that strictly rely on the 5V/500mA connection for power.

Looking at the devices I have plugged into my USB ports:

Bluetooth adapter - requires power but becomes less useful with Wireless USB
Keyboard/Mouse - wireless options already exist and both have proven the ability to run for long periods on battery alone
Touchpad - same as above
External HDD - already powered by AC adapter
PPC Cradle - already powered by AC adapter
Scanner - already powered by AC adapter
Printer - already powered by AC adapter
APC - already powered (duh!)
Ext USB hub - already powered by AC adapter

While wires from AC power would still be needed, wireless USB would allow you to take those devices and stop tethering them to your PC. AC sockets are easy to locate and manage, but having 20 cables all tying into your PC is not.

[farnold]

Could THIS be the death of Bluetooth then??

:devilboy:
Or could that be just nonsense because we already have Bluetooth 8O

My feeling is that we just have another nice trans-atlantic (or from here" trans-pacific) seperation as Bluetooth is quite successful everywhere except for the States.

[applejosh]

Isn't the next iteration of BT supposed to address some of the bandwidth issues (i.e. be more than 700kbps) as well as improve some other things? I thought I remembered reading something about the newer BT and wireless USB being able to coexist without interference. If wireless USB was supposed to outright stomp BT, I doubt they would have made such a statement. But then again, I'm full of cold/allegy medicine right now, so my head is floating about 5 feet above the rest of my body (or at least it feels that way). :silly:

[PPCRules]

Could someone remind us (I'm sure someone will) why Bluetooth is so much better than Wireless USB will be? In the last discussion on this someone brought up some [I think, ligitimate] points in that regard.

The key to this is how well Windows natively supports it. If XP SP3 (or a smaller auto-update patch) adds full support around the end of the year, this will exlode in 2006. Otherwise the scene will be a mess until it is fully supported in Windows (which I hope will be before Longhorn).

[surur]

Well, for anything secure you are still going to have to pair eg. keyboards, external drives, usb hubs, web cams etc. else anyone would be able to eavesdrop on your typed password or see through your web cam. I'm Sure the pairing process will look very similar to bluetooth.

The difference will be the absence of profiles. This is good and bad. The good part is if you have drivers for your kit it should work The bad part is that your are still unlikely to get pocket pc drivers for your web cam or printer, a problem that bluetooth addresses.

Surur

[organon]

To answer a prior question, Bluetooth 2.0 + EDR is available now and triples the current data rate. However, compare this to the 480 MB/s of WUSB, and you are still left with a tortoise and a hare.

The WUSB vs. Bluetooth question is made much more confusing, however, when you consider the fact that Bluetooth will now be using the UWB spectrum just like WUSB. A good discussion is available at http://www.techworld.com/mobility/features/index.cfm?featureid=1419 .

So now we have 2 protocol standards that use the same, low power, short distance, ultra-high speed wireless bus. Now the differences become less apparent. Bluetooth has "profiles," but (W)USB has what amounts to the same thing in common device types (i.e., "Generic STORAGE DEVICE USB Device", "HID-compliant device", etc.). (W)USB has the ability to create custom device drivers, but in essence so does Bluetooth by piggybacking on SPP and/or OBEX. Bluetooth is meant to connect mobile devices while WUSB is intended to connect PC's to peripherals, but we're really just talking positioning and semantics as either is capable of performing both tasks.

So, we have Bluetooth with first-mover advantage vs. WUSB with Intel's mass marketing muscle and a "USB" moniker from which to draw name recognition capital. Interesting battle--if it really is a battle. Personally, if WUSB really works, I'm betting on Intel in the long run, but I'm often quite wrong.

[Ekkie Tepsupornchai]

Isn't the next iteration of BT supposed to address some of the bandwidth issues (i.e. be more than 700kbps) as well as improve some other things?
Yep. I posted on that here (http://www.pocketpcthoughts.com/forums/viewtopic.php?t=39883).

I thought I remembered reading something about the newer BT and wireless USB being able to coexist without interference. If wireless USB was supposed to outright stomp BT, I doubt they would have made such a statement.
I have no doubts that the two will be able to coexist, but here's the fact of the matter. Bluetooth has been around for a few years now and the mainstream acceptance has been limited to BT headsets. There's been some success with GPS and keyboards as well, but I consider the success of both to be within tightly niche markets. On top of that, most of BT's success is limited to mobile devices.

OTOH, USB is without doubt, the preferred universal standard for PC/Laptop peripherals. Does making it wireless make it more attractive for mobile devices? I don't know that for sure, but with or without Wireless USB, if BT doesn't improve on its usability, its application might never extend much further than the niche market it currently serves.

Well, for anything secure you are still going to have to pair eg. keyboards, external drives, usb hubs, web cams etc. else anyone would be able to eavesdrop on your typed password or see through your web cam. I'm Sure the pairing process will look very similar to bluetooth.
I can only agree partially. Wireless PC keyboards have been around for years with no concerns over security. You push the button on the tranceiver and push another on the receiver for them to bond. That's it. If you don't have physical access to both, they don't pair. Now, whether this model works for mobile devices or not, I don't know, as there are more use cases to consider. But I don't see any scenario with PCs/Laptops where passkeys and multi-step discovery processes should be required.

[rmasinag]

I'd really like this to work, I really do!

[kkelley]

This would be great! I just bought an HP AIO and was thinking of buying a wireless jetdirect adapter so that I could scan and fax from my computer. The AIO is in the next room. Running a usb cable is not an option - too far. I have a Linksys wireless print server, but I still can't fax or scan without hooking up directly. There appears to be some ambiguity whether even after buying a jetdirect adapter that I could scan or fax wirelessly. :?

Anyway, I was just thinking that I wish they would hurry up and come out with this. Very cool!

:D

[Darius Wey]

Could THIS be the death of Bluetooth then??

I think it'll be years before that happens. So many companies seem hooked on employing BT in every device at the moment. Despite the plethora of companies also supporting Wireless USB, it'll take a while for it to become widely-known and thus become incorporated in a wide range of devices and accessories.

[jimski]

Don't want to change the focus of this thread, but it's interesting to read comments about how Bluetooth has not been accepted by the mainstream and is almost surely doomed. Well, I don't see Bluetooth devices on the "Blue Light" special counter at K Mart and in fact, I have yet to see any specific Bluetooth device (not including PPC's) actually on sale anywhere. Manufacturers have done a very poor job to get the word out. How many BT Stereo Headphones are out there for PC's. How about BT keyboards, mice, cameras, printers, etc. Or how about PC's themselves. Built in BT? What are you talking about. Maybe one, or possibly two of each device type can be found with BT built in, but that is not much of a selection. My attitude is, build it and they will buy (or at least I will).

I have no particular ill feeling towards Wireless USB, but I do look forward to more confusion in the years ahead.

[gibson042]

Could someone remind us (I'm sure someone will) why Bluetooth is so much better than Wireless USB will be? In the last discussion on this someone brought up some [I think, ligitimate] points in that regard.
Well, I brought up the issue of power consumption (http://www.pocketpcthoughts.com/forums/viewtopic.php?p=338328#338328), which will be significantly higher with WUSB than with Bluetooth (120x, ramping down over time to 40x). To be fair though, the speed/power consumption ratio actually favors WUSB; it is about 666 times faster than Bluetooth 1.2 and 222 times faster than Bluetooth 2.0 EDR.

[Dalantech]

I touched on some of this in my Da Wish List (http://www.pocketpcthoughts.com/forums/viewtopic.php?t=40337) thread, but I've got a few things to add to this one...

The BT group seems to be hanging by a thread lately -they know that BT is underpowered in the bandwidth department, but the BT stack (the underlying protocol that allows different BT hardware to communicate) is pretty good, so they've offered their assistance (and the BT stack) to the UWB group. We'll just have to wait and see what happens. If BT gets more bandwidth, then having a BT only wireless sync option in WM5 won't be a big issue...

Power: Since a laptop hard drive doesn't consume a lot of juice it shouldn't be too tough to come up with a portable WUSB hard drive with it's own built in rechargeable battery -maybe one that matches your PDA and even docks in the same cradle for recharging. I really need to find a job in R&amp;D and stop giving out free design ideas *sigh* ;)

The big advantage to any wireless tech is the ability to get withing range of a device and pair with it, sync data, etc. If a device is normally stationary (like a printer) then having to plug it into the wall won't matter. Being able to print without having to physically cable up, or without having to maintain a wireless print server, will be very 8)

[martin_ayton]

For what it's worth:

Wireless USB can only be a viable proposition because of weaknesses (real or perceived) with Bluetooth. That means that WUSB has something to shoot at and something to be better than: Hopefully, the providers of WUSB will learn from the mistakes made in rolling out BT and we'll get a great product.

Up until now, BT has effectively had no competition, so if 'they' made a poor job of it, what the heck? There was no alternative, so the poor consumer had to grind her or his teeth, get expert with the drivers and the hacks and make the best of a bad job. Personally, I have found BT no more (or less) painful than many other first-launch technoloigies.

What I really hope for is that having the new kid on the block will force the BT purveyors to really raise their game so we get a choice of excellent, user friendly wireless solutions. Am I just dreaming?

[surur]

Well, for anything secure you are still going to have to pair eg. keyboards, external drives, usb hubs, web cams etc. else anyone would be able to eavesdrop on your typed password or see through your web cam. I'm Sure the pairing process will look very similar to bluetooth.
I can only agree partially. Wireless PC keyboards have been around for years with no concerns over security. You push the button on the tranceiver and push another on the receiver for them to bond. That's it. If you don't have physical access to both, they don't pair. Now, whether this model works for mobile devices or not, I don't know, as there are more use cases to consider. But I don't see any scenario with PCs/Laptops where passkeys and multi-step discovery processes should be required.

A couple of neighbors in the Norwegian town of Stavenger both had new wireless keyboards from HP. These keyboards, like most other wireless models, transmit keypresses to the computer via radio signals that are beamed to a receiver. But the neighbors quickly discovered that if both of them used their PCs at the same time--even though they were more than a football field's length away from each other in different apartments--both receivers would output anything that either neighbor typed.

http://www.pcworld.com/howto/article/0,aid,108712,00.asp
Further down on the same page.

RF keyboards that use the Bluetooth radio system provide the highest security. Bluetooth keyboards encrypt the radio signal, and synchronize with the receiver connected to the desktop when you start your PC. Even if the keyboard's transmission penetrates walls or crosses large distances, it can't be decrypted by any receiver other than the one that it's matched to. At press time, only Microsoft offers a Bluetooth-enabled wireless keyboard in the United States.


A wireless system needs some kind of shared key to be secure, unless what it transmits is completely unimportant. If security is not built in from the beginning (as with bluetooth) we end up with an evolutionary process and several generations of products of varying security (such as the WIFI mess currently). I sure hope they did not just transfer the USB protocol (with no security to speak of) to a wireless carrier. Imagine I camp out outside your office and suck all the data from your USB hard drive.

Surur

[Ekkie Tepsupornchai]

A wireless system needs some kind of shared key to be secure, unless what it transmits is completely unimportant. If security is not built in from the beginning (as with bluetooth) we end up with an evolutionary process and several generations of products of varying security (such as the WIFI mess currently). I sure hope they did not just transfer the USB protocol (with no security to speak of) to a wireless carrier. Imagine I camp out outside your office and suck all the data from your USB hard drive.
First, if the example of security issues is arising from neighbors in Norway, then to me, the problem is not widespread despite huge acceptance of wireless mice and keyboards all over the globe.

Second, a Bluetooth-style of shared key is of highest security if the key can be personalized. For it to be personalized requires a method of inputting the key on both ends. Assuming that's not realistic (which is the case today for several BT devices), then a set of rotating keys in which the receiver chooses a key and and the tranceiver scans for that key (both prompted by some sort of hardware event like a button push) seems to be the next logical method of execution.

This is not really all that different from how our PPCs get bonded to our BT headsets or GPS units or keyboards. The biggest difference (as illustrated by the Norway-neighbors) is that RF tranceivers can just transmit blindly with no regards to who the receiver is whereas a BT device would be truly bonded with a just one receiver (theoretically).

So like Bluetooth, there does need to be device-specific pairing happening on both sides, but unlike Bluetooth, the process should be more transparent to the user. Manually-entered passcodes are meaningless if they can't be manually entered on both sides.

[surur]

I'm not sure about "scanning a list of rotating keys" but what happens in general with devices without an ability to enter a key is that they come with a preset key thats documented on the packaging. For headsets its normally 0000 . In fact this key is only needed initially at pairing to authenticate, and is not used afterward.

Surur

[Ekkie Tepsupornchai]

I'm not sure about "scanning a list of rotating keys" but what happens in general with devices without an ability to enter a key is that they come with a preset key thats documented on the packaging. For headsets its normally 0000 . In fact this key is only needed initially at pairing to authenticate, and is not used afterward.

FTR, the thought of "rotating keys" was a proposal of how things might work with WUSB, not of how BT executes. In any case, with BT headsets, there is something more involved with the bonding itself than just the 0000-key authentication. You can pair your BT headset with multiple phones, but only one phone gets identified as the preferred device from the headset's standpoint (the "preferred phone" being defined as the one it attempts to connect with upon power-up) and only one phone could be used with the headset at any given time. My point was that the 0000 passcode adds no value. The real value comes with how any given tranceiver connects with one-and-only-one device at a time; thus, making the manual discovery-profiles-passcode process rather pointless in my view. I would think a "button-push" on each side could trigger the same bonding (with the same security levels) if implemented right.