Sunday, December 24, 2006

Can the Nokia factor bring Wibree to the mass?

Nokia launched low rate, low power, small size, low cost cable replacement radio technology.

In October 06, Nokia announced the existence of yet another wireless technology named Wibree developed by Nokia's researchers.

Wibree's properties can be summarized as follows:
- It works on the ISM band, already shared by Wi-Fi, Bluetooth, ZigBee and other proprietary technologies.
- It has a 30-foot (10-meter) range, similar with Class 3 Bluetooth.
- It can achieve speeds of up to 1 Mbps.

By looking at its properties, Wibree is somewhat the earlier Bluetooth version but Nokia argued that Wibree is different because Wibree achieves its performance with one tenth the power consumption of Bluetooth and smaller footprint. Wibree addresses a niche that hasn't been addressed by other wireless technologies.

Technology-wise, nothing new and special about Wibree, since several proprietary short-range radio technologies also work with very low power such as the potential Kleer but Nokia aims to push Wibree to becoming an open standard by taking advantage of Nokia's leading position in mobile phone business. Nokia encourages other companies to license Wibree, creating a Wibree ecosystem emulating the success of Wi-Fi and Bluetooth, and developing standalone Wibree chip and dual-mode Wibree-Bluetooth chip.

In a standalone scenario, Wibree is going to be embedded in devices which require very long battery life therefore need to be recharged infrequently, such as watches, health monitors, sport sensors, keyboards, mice, and toys. However as a number one mobile phone maker, Nokia doesn't hide its ambition to integrate Wibree into its mobile phones by creating a dual-mode scenario that's Wibree on Bluetooth chip. The dual-mode Wibree-Bluetooth chip is targeted for mobile/portable devices such as mobile phone, PDA, notebook.

The typical applications Nokia uses as examples automatically direct my mind to ZigBee (low rate WPAN IEEE 802.15.4). ZigBee is promoted by several industry leaders in sensor, monitoring and automation business. Its specification had already been completed and certification has been started. Whilst Bluetooth has throughput advantage over Wibree, ZigBee has range advantage over Wibree. Moreover, ZigBee has networking capability. Wibree is caught in the middle.

But the fact that ZigBee is still far from replacing the X10 protocol and the likes for one of its target markets, i.e. home automation might have led Nokia to develop yet another radio technology and put Wibree on a distance from ZigBee. In my opinion, ZigBee players lack the drive to push ZigBee to the mainstream and ZigBee doesn't have a news-machine. No one did like what Intel had done for WiMedia-UWB (WUSB) and WiMAX.

Below is a comparison table examining the feasibility of Wibree for today's common applications and the foreseeable future with a mobile phone as the base device. Please note though that a Wibree Specification hasn't been released so this is only a general assessment.


ApplicationWibreeBluetoothZigBee
Business card or other PIM data exchangeYes. The size is small, only text file.Yes. Its primordial use.No. Not designed for this.
Transferring small item like image, song (ringtone), short clip one at a timeYesYes. Already in use.No
Wireless headset or handsfree kitNo. Audio is not in its profiles, although theoretically it can be done.Yes. Has upgraded it to support stereo audio.No
Synch'ing mobile phone with PCNo. Too slow especially for synch'ing many items.Yes. One of its popular applications.No
Using mobile phone as a dial-up modemNo. Not within its targeted application.Yes. Has been used by travelers.No
Image, photo, or video/audio clip folder transfer/sharingNo. Will be very slow.Yes. Still needs speed increase as multimedia files become largerNo
Human Interface Device (HID): keyboard and mouseYes. One of its main targets. Need to switch current Bluetooth based products to Wibree.Yes. Already in use and available from many vendors.Yes. But implementation is almost unheard.
Wireless printingYes. One of its main targets. But over time its speed would keep users away from it.Yes. Already in use. The next generation Bluetooth (WUSB-alike) has greater potential.No
Remote controlling TV, lamp, window, door, AC, etcYes. One of its main targets although its reach has to be optimized. Interesting if a cell phone becomes a universal remote control.No. It theoretically can do it but implementation on this field is rare, many proprietary protocols have accomplished the job.Yes. ZigBee was designed from the beginning for home automation. It has longer reach than Wibree (30 meters or more).
Communicating with wearable things (watch, sport sensor, jewelry)Yes. One of its main targets.No. Consumes too much power and has larger size.Yes. Has advantage in its ultra low power consumption.


Latest update: Wibree finally merged with Bluetooth.

Friday, December 15, 2006

Wouldn't you mind if your PC doesn't have a USB connector?
There comes Wireless USB !

Perhaps, nothing more confusing in the realm of burgeoning wireless standards than Wireless USB (or WUSB for short). The name itself could create misperception. Below are my notes on the Certified Wireless USB (its complete name).

1. Wireless USB does NOT mean USB port plus a wireless dongle such as that for Wi-Fi, IrDA, or Bluetooth. But a type of a Wireless USB implementation and this is the most likely case in the first phase is just wireless business as usual, that's you'll connect a Wireless USB dongle (referred to as HWA or Host Wire Adapter in the spec) to a USB connector on your PC to enable Wireless USB on it, like what you have done with Bluetooth, IrDA and Wi-Fi right now. Meanwhile, legacy USB products such as your printer, scanner, keyboard and mouse can be connected wirelessly to the WUSB-enabled computer via a DWA (Device Wire Adapter).

2. Because every one of us has an intense relation with USB port and cord every day, in our mind USB is always identical with that rectangular connector. It's hard to imagine a USB without cable. When the Wireless USB comes into existence, maybe we should start calling the current USB a "Wired" USB to remove confusion. Wireless USB has similar properties with USB 2.0 minus the cable.

3. To initiate a connection, two wireless USB devices have to associate each other to ensure both are the parties that want to talk to each other. The process which is called pairing in Bluetooth. There are two association models, i.e. numeric association model using number input/display and cable association model by connecting two devices via a short cable before going wireless. Not entirely getting rid of cable.

4. The physical layer as has been discussed in the previous post uses UWB radio (a very wide spectrum in the 3.1 - 10.6 GHz band) based on the WiMedia Multiband OFDM platform. This results in very high throughput (by current standard), that's 480 Mbps at up to 3 meters and 110 Mbps at up to 10 meters.

5. Because of its range limitation, it is categorized as a Wireless Personal Area Network (WPAN) technology, the same category as Bluetooth but added with this citation "high-rate" WPAN. When Bluetooth starts using UWB radio as its PHY and becomes high-rate as well, I wonder how it will be named and positioned on the new wireless map.

6. Similar with Bluetooth, the topology is point-to-point in a hub-and-spoke configuration. In a Wireless USB cluster, there must be a host and up to 127 devices, compared to a master and up to 7 active slaves in a Bluetooth piconet. In an OTG (On-The-Go) capable Wireless USB gadget, it can sometimes act as a host or a device depending on each connection purpose.

7. The targeted usage scenario is home entertainment connectivity, which encompasses TV, PC-related hardware, CE (Consumer Electronics), and mobile/portable communication/multimedia devices. Most are the same devices that have already carried Wi-Fi or Bluetooth.

Why did they (Intel n co) create another wireless, while we're overwhelmed with the existing techno? There is a formal logic behind the standard creation that's the success of USB as a number one interface on PC and other PC-related hardware. We must admit though that some portion of that success comes from that wireless dongle that plugs nicely into a USB rectangular connector.

Whether Wireless USB will be as successful as its counterpart "Wired" USB? Its promoters said yes, because it is easy to use and plug-and-play. But surely, Wireless USB will find competition from older wireless standards.

Sunday, December 10, 2006

New Year's Wishes (1) : Beautify the stack and the spec

To Microsoft:

Please extend Bluetooth capability in Windows XP (SP2) even though you are now busy selling Vista. Do not leave XP users in the dark. Add more services to it with reference to the latest Bluetooth specification.
Did you know that many users find it difficult to work with various third party drivers and software? Yeah, Bluetooth protocol is more complicated than Wi-Fi, but it should be included in the generic Windows functions since it has become an established standard.

To the Bluetooth SIG:

Bluetooth has spread like wild fire and it's present in nearly every new gadget, but the popular use of it is still limited to certain basic services due to - at least in the perception of many users - the impractical process of device authentication and authorization and the confusing and lengthy profile-matching.
That's why please shorten and simplify (or automate) the connection process to make it easy for common users to enjoy all Bluetooth services. Make passing an introperability test a requirement for a compliant Bluetooth product.

Agree or not? I've got to channel many readers' frustrations anyway. Hope you guys stop by and respond. Thank you.