Why not multicast all of the way down?
May 8, 2007 12:00 PM
Mahmud Noormohamed, Net Insight’s VP business development North America, thinks there’s a much better way for telcos, particularly those in rural areas of the United States, to approach their IPTV roll-outs.
They would be better served by aggregating hundreds of encoded television channels at a headend and routing them all down to DSLAMs, he says. That way issues with IP overhead or those associated with packetizing video can be avoided.
But that’s just half the story. The same aggregated channels from a single headend can be shared via OC-192 links so other rural telcos can launch IPTV services to their customers without the CAPEX required to build their own headends.
IPTV Update: As Net Insight approaches telcos about IPTV service, what are their common requirements at their video headend to manage and aggregate incoming video for IP/MPEG-2 or IP/MPEG4 routing to the network backbone?
Mahmud Noormohamed: In the U.S., we are only talking about the rural telcos. We are looking at independent IOCs really. Our proposition there is consolidate —virtual headends rather than fixed head end per rural telco. The underlying factor is that if you share, it’s a lot cheaper.
So that is the approach we are taking, and it is resonating with the consumer for many reasons. One is the negotiations per rural telco with the content companies is quite laborious and long. So if the rurals can get together and negotiate once and then share an infrastructure, it obviously benefits them.
IPTVU: Who’s currently employing this strategy in the United States.
MN: We have three triple play solutions here in the U.S. — Midwest Tel Net, Vernon Telephone and Matanuska in Alaska.
IPTVU: As you approach telcos in rural settings, how are you finding their interest in MPEG-4 vs. MPEG-2?
MN: From our technology perspective, it doesn’t make any difference whether it’s either of those formats or it’s uncompressed. Our play is if you transport uncompressed, you can share the quality of the technology and when it gets compressed, you’re superimposing the quality that we’ve already demonstrated.
From our perspective, it is irrelevant, but from the rurals’ we definitely are seeing a trend toward MPEG-4. They do need it to offer HD channels, and really the only way they can acquire them and transport them in terms of infrastructure is through MPEG-4. So they are deploying trials in that effort.
IPTVU: Have you found the rural telcos to be up to speed on the QoS demands of video in an IP network?
MN: We guarantee 100 percent QoS, so whatever comes in will come out of the other end without any latency, jitter or wander. Our play is to measure and manage all of that because our infrastructure will guarantee it.
IPTV: How so?
MN: Because we are a Layer 2 protocol, so we don’t have the IP overhead or issues associated with packetizing video. We are a synchronous transport, so we are kind of the TDM of the IP world. That removes 80 percent of the problems. Our play is take video synchronicity to the local loop before you put it through an IP infrastructure. That way, you can avoid a whole bunch of management overhead.
IPTV: How is Net Insight’s Nimbra being deployed at IPTV headends and being used for local distribution?
MN: We are being deployed at headend aggregation, pre- and post ingress, meaning we do support companies that are delivering to the headend — digital media services from their broadcast center. And then post headend we are distributing into the local metro area. We do not do access, so we terminate at either VOD or DSLAM.
| Want to use this article? Click here for options! |  |
