VoIP Peering - Breaking Out Of Walled Gardens

by Brian McConnell

2005 will be remembered as the year that VoIP really (finally) took off. The combination of better VoIP software, simpler configuration procedures and a bewildering array of VoIP ready devices made VoIP a mass market service.

The problem with VoIP is that most networks are closed systems. If you're a Skype user, you can call other Skype users for free, but if you want to call people on other VoIP networks or the PSTN, you have to pay. True, you pay much less than you would for a conventional PSTN call, but if you spend several thousand minutes per month on your phone (if you run a business, you probably do), you can still rack up some pretty big bills, even at a few cents per minute.

The solution to this problem is VoIP peering. The idea behind VoIP peering is to map calls to and from "area codes" that belong to a particular service provider. SipPhone is doing this in the Gizmo Project. So let's say you want to call someone on Iaxtel. You just dial 1-700-xxx-xxxx. If you want to call someone on Blueface in Ireland, you just dial *353-xxxxxx. So calling people on other VoIP networks is as simple as making a phone call. This is one reason I predict that open systems like Gizmo will win in the long term.

As a user, I shouldn't have to care which network someone is using. I don't have to worry about it when I call a public telephone number. The person I am calling could be on a GSM cell phone, a PBX connected to a T1 line, or a old analog rotary phone. The type of connection doesn't matter. The public telephone network is built upon the concept of peering, and wouldn't exist in its present form without it.

VoIP peering is easy to implement with services that support SIP or IAX2 protocols. These peering arrangements are currently managed on an ad hoc basis between service providers. There are efforts underway to create a centrally managed numbering plan (such as E164.info), similar to the North American numbering plan, that will make this process easier to manage. For now, peering arrangements are done informally, with providers on each end routing calls for each other using a mutually agreed upon area or country code. The economics of VoIP peering are compelling. Routing calls from end to end entirely via TCP/IP reduces the need for expensive public telephone network interconnections, and enables completely toll-free calling between endpoints. Watch for peering between VoIP services to take off in 2006. Gizmo is especially well-positioned to benefit from this trend.

VoIP peering will also work the other way. Once there is a uniform numbering plan (area codes for VoIP service providers), conventional fixed-line and mobile operators will be able to route calls from the PSTN to these networks. Want to call an IAXtel user on your cell phone? Just dial +1-700-nnn-nnnn and the cellular network would take care of routing the call via VoIP once it hits the terrestrial network.

Fixed line companies, already hurting from losses to cellular and VoIP providers, will probably shun such relationships. Mobile operators could make a ton of money by embracing VoIP peering. Cellular carriers make money selling airtime, which retails for roughly ten times the per-minute cost of calls placed over fixed-line networks. It will cost them virtually nothing to route calls to VoIP networks. They make their money on the short hop from the handset to antenna. A cellular operator that provides direct calling to VoIP networks will be very popular indeed.

Mobile to Voip (M2V) peering will be a great deal for consumers, and a huge market for carriers that will otherwise lose traffic to end-to-end VoIP calls. It will also make VoIP accessible to any mobile phone, not just high end devices that are capable of running a SIP client. As far as the phone knows, it's just dialing an ordinary telephone number, as the call rides on the voice side of the network from the handset to the base station.

Will customers pay for the convenience? This, I think, is a no-brainer. People already are paying several times more to place mobile calls instead of via cheap fixed line phones. They value convenience more than absolute cost. If they can call friends in far flung places for the cost of a local mobile call, they'll make calls via cellular that they'd otherwise make from another phone. I don't expect Verizon to start offering this service any time soon. VoIP companies will be way ahead of the curve in embracing peering, but sooner or later the telcos will pay attention, and the smart ones will be laughing all the way to the bank.