Universal Plug and Play (UPnP)
Andrew Liu, Co-Chair UPnP Marketing Committee at Intel, writes:
"It's great to see UPnP starting to get some coverage in your newsletter. As you may be aware, UPnP is going to become very important this year in making home networking more accessible to consumers.
One point that I would like to make about residential gateways is that they all have a fundamental problem that gives big headaches to broadband/home networking applications. They all use NAT, and that causes applications such as VoIP, video-conferencing, hosting multi-player games, and remote assistance to break. Many of these applications are broadband drivers, but what good are they when the residential gateway breaks their usage model. UPnP internet gateways solve this problem. I would like the opportunity to into more detail here and perhaps let you try first hand the benefits of UPnP in solving these problems.
Here's a good reference article: http://www.microsoft.com/windowsxp/expertzone/columns/bowman/december24.asp "
We also heard from Bob Quinn of sockets.com:
"It will be interesting to see what happens in this space! Getting a common API over standard protocols that can use any of the many in-home network medium is key. The fact that it uses XML to provide "self-description" facilitates discovery and zeroconf, which in turn enable the transparency that will make for viable consumer products.
I was involved with the creation of the WinSock API, and this is just as promising for SOHO and factory nets as WinSock was for early Internet applications (personally, I believe WinSock was responsible for the early success of NetScape, for example). Killer home apps here we come!!!"
( www.sockets.com )
We heard from William Watté, General Manager at M-TEC WIRELESS in Belgium:
"I'd like to point out that there is still another standard which was overlooked in your overview. The standard is HiperLAN/2. Opposite to what some people in the yellow press are trying to spread, this standard is far from death! And it is also not "the standard for Europe". HiperLAN/2 can be used all over the world, which is not the case for 802.11a.
However, the discussion of "which standard is going to win" is the wrong discussion. The real discussion should be "which standard should we use for which application". If you are looking for a network where Ethernet has to be replaced by something wireless, an 802.11 standard is clearly the best choice. However, if you are looking at other applications requiring Quality of Service (like video distribution and gaming), the only right choice is HiperLAN/2. More information on the QoS feature in HL/2 can be found on our website ( www.mtecwireless.com/htdocs/products/HL2doormanrev.pdf ).
Since the home environment is full of QoS broadband (with TV on the first place), HiperLAN/2 is probably the first choice for this environment. However, a dual system where also 11a is supported (for the laptops brought home from work) will probably also emerge quite soon.
As you know, our architecture is protocol independent due to its software configurability. Therefore, we don't really need to push any of the standards. We are just providing you this information so you can keep your readers up-to-date. And we would like to keep them from a disillusion which they might experience when transmitting video signals over a non-QoS 11a system. We all don't want this to happen because this could deteriorate the public opinion towards broadband wireless LANs."