AT News: Wireless USB Primer

We were promised that with the digital age the paperless office will be a reality; we all know how well that turned out. Now we have more printouts of emails and articles than we can keep track of. Another problem of this digital age is the tangle web of wires behind everyone's desk. Efforts to eliminate this tangled web of cables connecting PCs to peripherals have resulted in varying success. However, a new technology might change things in the near future.

At the Intel Developer Forum on February 18^th, Intel announced it has given up on the deadlocked Ultrawidebanded IEEE task group and, with other industry leaders, announced the formation of the Wireless USB Promoter Group to promote a Wireless USB (WUSB) protocol based on ultrawideband technology. The group plans to deliver a completed specification by the end of the year. The first Wireless USB implementations will be in the form of add-in cards and dongles for use on PCs for communication with other WUSB devices.

As the latest implementation of USB technology, WUSB will offer the same functionality as standard wired USB devices but without the cabling. Usage will be targeted at wire replacements at the PCs and PC peripherals, consumer electronics and mobile devices. The new WUSB will have an initial target bandwidth as USB 2.0; 480Mbps over distances up to 10 meters, with future goals over 1000Mbps depending on the adoption of the technology.

WUSB will be implemented in a hub-and-spoke topology. In this topology, the host initiates all the data traffic among the devices connected to it, allotting time slots and data bandwidth to each device connected. The connections are point-to-point and directed between the WUSB host and WUSB device. The WUSB host can connect to a maximum of 127 WUSB devices, called a WUSB cluster. This topology will support a dual role model where a device can also support limited host capabilities similar to USB On-The-Go. Thus, the model allows mobile devices to access services with a central host supporting the services (i.e. downloading a picture from your cell phone to a photo kiosk).

WUSB security will be designed to ensure the same level of security as wired USB; which is basically nonexistant. Connection-level security between devices will ensure a device is associated and authenticated before operation of the device is permitted. Higher levels of security involving encryption will be implemented at the application level, so it does not look like WUSB will become a secure protocol out of the box. The overhead of supporting security should not cause any noticeable performance impacts. Naturally, wireless USB will not pose as severe security situations compared to wireless LAN (due to range), but the issue of your officemate stealing photos from a digital camera could be a real threat. Unfortunately, this threat will not be assessed by the WUSB Group, but by the individual device/application creators.

One of the primary objectives of a WUSB is that it is easy to install and use. Wired connections provide the user with directed expectations. When the device is plugged in, the user see that a connection is made. With wireless connections, connection paths are not always obvious. To facilitate the concept of "turn on and use," WUSB devices that are installed for the first time will automatically install drivers, security features, and etc and associate with systems that they can interact with.

What about Bluetooth? Up until Intel's recent announcement the company was one of the major proponents but the technology was delayed to the market. Furthermore, it has never gained massed acceptance in the PC market. Now based on this new product direction from Intel, its use in the PC centric environment is all but over.

WUSB provides a major performance benefit over the current implementation of Bluetooth. High capacity throughputs are increasingly required by peripherals like Apple's iPod, digital cameras, and removable hard drives. Another oft mentioned problem with Bluetooth is its difficulty to get two Bluetooth devices to talk to each other. By using the USB 2.0 specification Intel can eliminate many of the software compatibility issues that other emerging connectivity solutions like Bluetooth experienced. Unlike Bluetooth, WUSB compatibility with existing USB devices makes it easier for peripheral and PC developers to roll out the technology. Intel's approach to using an existing and accepted specification creates a faster time to market for the related devices. The fact that Intel is also a major motherboard and chipset manufacturer also helps WUSB adoption rate and more importantly, keep costs down (if you adopt Intel based motherboards that is).

Another problem of this digital age is we don't have wireless power yet. While some of these devices pull power from the USB cable, bus-powered WUSB devices such as removable USB hard drives will require a power brick. Until someone manages to broadcast power without killing us, a truly wireless environment may never be possible.

Considering the specification is not finished, nor will be finished until at least January of next year, WUSB still has a long battle to fight. It is very likely that unofficial WUSB products will start appearing before the standard is complete (similar to JEDEC and IEEE standards).