I tend to agree with vtraveller. Wi-Fi and UWB were architected for two very different purposes. Wi-Fi for Wireless LAN (i.e. connecting to the internet), UWB for Wireless PAN (i.e. connecting devices together). Just like Ethernet and USB, Wi-Fi and UWB are very complementary technologies. In any case, more wireless is a good thing IMO...
Well I started developing 802.11 products way back in 1999 and have put them into everything from PCs to mobile phones as well golden reference designs in WiFi.
What I can say with some authority is that while 802.11 works well it doesn't work well enough in numerous real-world applications.
The 2.4GHz band is crowded and suffers co-existence problems. Particularly with BT/WiFi calling in a single handset. People still can't do a BT voip to WiFi call. Exactly the same problem applies to video streaming, which at higher data rates suffers similar co-existence problems. The issue is derived from not enough capacity over-the-air (particularly in mixed-mode backward compatible setups) and mixed support of QOS services both at the 802.11 and TCP/IP levels.
Headline figures have also historically been grossly over rated. Only with 802.11n has throughput criteria been measured in payload data rather than transmission speeds (ignoring NAK, retry and rate-adaptation). I think your claims that 802.11 didn't hype-up data rate are utterly wrong; I lived it and had to fight competitors over inflated claims based on headline data rates.
On the 5GHz band (originally from 802.11a) theory. 802.11 can't get traction in this band. It has an awful market price-point for Access Points and is only in the Intel chipset due to it being a Microsoft Certified criterion. 5GHz still isn't viable for a lot of people.
WiFi power usage is also pretty darn bad. You certainly can't expect a phone to survive beaconing for very long.
Then there's the matter of set up and configuration. As 802.11 is a bridge for 802.3 you typically run it with a TCP/IP stack. Getting peripherals to work with this is cumbersome and unreliable. When my HP printer changes its IP address through DHCP (or I'm not connected at start-up to the network) the HP client software uninstalls my printer driver. No-one had ever considered that the network may appear and disappear periodically when using WiFI. Is 802.11 really that usable for peripherals considering its reliance on high-level IP protocols?
All I'm saying is that 802.11 creeks at the edges. UWB, particularly WirelessUSB, fits well in solving a lot of these problems:
- Good power / bit
- Decent fast data rates; products now achieve 150MB/s of real data throughput (i.e. based on file copy times)
- Fits well into mobile devices
- Great migration path with wired to wireless USB bridging (products can work with existing USB drivers)
I'm glad you're keen to compare with WiFi. It highlights exactly why UWB came along. WiFI has a very clear place and solves a clear set of problems. UWB solves a different set and as such I'm unconvinced there's as much overlap as you suggest.
Low latency video streaming is important for wireless PC docking station, wireless projector and other wireless video applications. I agree with Colombatto about the UWB advantage and WiFi congestion problems associated with the above video applications.
If UWB can deliver stable and inexpensive solution then it have a very good chance against WiFi in PAN applications.
NASA's Orion Flight Software Production Systems Manager Darrel G. Raines joins Planet Analog Editor Steve Taranovich and Embedded.com Editor Max Maxfield to talk about embedded flight software used in Orion Spacecraft, part of NASA's Mars mission. Live radio show and live chat. Get your questions ready.
Brought to you by