[February 4, 2005]

"The Voice of IP" Begins to Speak

BY DAVID DUFFETT
THE VOICE OF IP

Hello, my name is David Duffett and I’m the new guy on the TMC block – in fact I’m the ‘voice of IP’ (thanks to my colleague James for thinking that one up). In this new column I would like not only to consider some of the technology that’s out there and the stuff that’s coming down the line, but also to stand back and take a wider view of things too. If we don’t do that every now and again, we could easily risk falling into the trap of becoming the kind of technology nerds that normal people think need ‘special help.’

Don’t believe the hIPe!

I recently, and unwittingly, uncovered the huge amount of animosity and angst there appears to be between those in the VoIP community who have a strong commitment to H.323 and those who have similar feelings about SIP. There is also a large amount of propaganda circulating from both sides.

During some research for a presentation I delivered just a few days ago, I was looking for a comparison of these two protocols that are both used for call control in the crazy world of voice over IP.

Having found just what I was looking for at sipcenter.com, I converted it into a script (in my presentation, you see, I actually interview Mr. SIP and Mr. H.323!). Then I thought, perhaps I’d better run this by an H.323 person just in case it is biased in any way. The results were amazing. You can try this little experiment yourself…next 4 th July, print out the page at this URL: http://www.sipcenter.com/sip.nsf/html/SIP+and+H.323, hand it to a worthy H.323 supporter and retire to a safe distance – you may get the firework display of your life, or generate enough heat to warm your house until Christmas 2006!

The fact is that, while all the talk seems to be about SIP, most of the action seems to be around H.323.

Your background, to an extent, dictates who – or what – you are. This is the case with SIP and H.323. While H.323 has a telecoms background (borrowing from Q.931 based protocols), SIP is very much a child of the Internet and this is why they approach their jobs so differently.

In the future these guys are going to have to learn to get along – interoperability and co-existence of these two is now a very popular subject with those who are pragmatic enough to concern themselves with real-world deployments.

It will be interesting to see how things work out, and this is a subject I’m sure we will come back to in future editions. Meanwhile, though, it might prove interesting to allow the vehement supporters of H.323 and SIP to settle their differences in a more entertaining way – televised wrestling springs instantly to mind.

Quality of Service (QoS)

Have you come across DiffServ (differentiated services)? If not, let me explain (if you have, feel free to skip the next paragraph):

DiffServ is a standard that specifies a common way to indicate a packet's priority level using a six bit field in the IPv4 packet header's Type of Service (TOS) octet, or the IPv6 Traffic Class octet .

This seems like a great idea, especially given the lack of guarantees that VoIP can be prone to – but it does seem to fly in the face of one of the ideas that we were all sold about the beautiful simplicity of IP networks, i.e., that they are stupid! They just shift bits, they need no intelligence – meaning that feature rollouts just involve upgrading endpoints (where the intelligence resides), not having the time consuming job of fiddling with the networks.

If information about a packet’s priority is contained within a header somewhere then each network node must be able to recognise this prioritization and act accordingly. Now that’s intelligence right there!

I well remember a VoIP enthusiast telling me that when a traditional telephony carrier wanted to implement the caller line ID feature for their customers it took more than a decade to upgrade their network, and rolling out the same functionality in an IP network takes a few months – most of which are used for testing! It was a great story and it made a great point, that you can implement any functionality you want over IP as long as all the endpoints involved are enabled for it – or at least that’s what I thought.

This may be more of a philosophical than a practical point, but it just seems like we’re burning away some of the blank canvas here – what do you think? My contact details are below.

What do you want to talk about?

Over the next few editions I’ve lined up some intensely interesting stuff to discuss. Next edition we’ll talk about host media processing (that’s where you use a host processor to deliver media functions like record and playback instead of using purpose built DSP boards) and the smoke and mirrors used in the information that vendors release about their products in this area. I also have plans to investigate next generation hardware solutions and the exciting way that hardware architecture has changed to allow IP to be inherent in the design, not just an add-on to traditional telephony products. If that wasn’t enough, your appetite will be well and truly whetted when I reveal that if you stick with me we will explode some of the myths around the thorny issue of VoIP enabling legacy equipment.

But, hey, this is your column too – so tell me what you would like to see covered.

If you would like to see one of my live presentations, or just say ‘Hello,’ I will be speaking at the following TMC conferences:

Internet Telephony Expo, Miami:
Wednesday, 23rd February 1.00-1.45pm

Speech World, Dallas:
Tuesday, 24th May 9.00-10.00am
Thursday, 26th May 9.15-10.00am

Remember – host media processing is covered in the next edition. I look forward to talking with you again.

David Duffett
[email protected]

David Duffett, TMCnet's columnist for "The Voice of IP," is a Chartered Engineer and has been in the Telecoms sector for over 14 years with experience spanning air traffic control communications, wireless local loop, mobile networks and computer telephony.

At Aculab for 5 years, David has global responsibility for customer training through the Aculab Academy.