Eggheads Search for Wireless Coexistence
If the research pays off, it will remove the problems associated with roaming between wireless LAN hotspots and third-generation cellular networks, allowing the creation of seamless combined systems. But don't hold your breath: A manager at HP stresses that what they are doing is research, and actual bankable technology could be five years away. That is, if it ever makes it out of the lab.
The white-coats at HP and MIT don't want to give away too many details yet, but the research basically involves breaking the golden rule of networking, by allowing the different layers to "see" what is happening higher up or lower down in the network hierarchy.
"Interlayer optimization," is what Susie Wee, R&D manager with the streaming media systems division at HP, calls the idea.
"Your device doesn't have to not know what's going on in the network," she says, no doubt making networking purists everywhere drop their slide rules.
So, how would you get wireless LAN networks and cellular networks to happily "talk" to each other, Unstrung wondered? "Well, you abstract the things that are the same about each network protocol and then make an API [application programming interface] that is common to all of them," Wee said.
Ahhh, we should have guessed.
This is similar to the "plug-and-play" concept that has become popular in desktop computing over the last few years. Plug-and-play gives users the ability to plug a device into a computer and have the computer recognize that the device is there. Interlayer optimization will (ideally) enable different types of data -- voice, text, audio, and video -- to be "plugged" into various types of wireless network, allowing the network to react depending on what data it is handling and what device it is communicating with.
This is the eventual aim of the project, which would enable service providers to build the kind of network where a user could communicate with both cellular networks and public access WLAN hotspots -- with one device and without huge amounts of fuss and bother. However, other wireless network technologies will emerge from the project before then.
For instance, Wee said that by breaking down the separation between the network layers, the researchers could enable smoother audio and video streaming over wireless networks. Presently, multimedia streaming over wireless networks is a herky-jerky experience, with the connection often being dropped midstream.
"[With interlayer optimization], the network layer could tell the application layer, 'Hey, maybe you could take two paths,' " Wee says. Thus, if one stream is disrupted, the other would take up the slack. One thing she didn't explain is whether this means more traffic for wireless networks that are expected to be congested as it is.
MIT has five professors and between 20 and 25 grad students working on the project, while HP has a dozen researchers on it. The wireless research is part of [email protected], a $25 million, five-year research hook-up launched by HP and MIT in June 2000.
— Dan Jones, Senior Editor, Unstrung