Why AT&T Likes HomePNA

SAN JOSE, Calif. -- IPTV 2007 -- While coaxial cable would be the best medium for a home network, AT&T Inc. (NYSE: T) is happy with using the Home Phoneline Networking Alliance (HomePNA) specification as part of its U-verse IPTV service, AT&T Labs executive Vernon Reed said at the IPTV 2007 conference yesterday.

Reed, sans guitar, gave a presentation explaining the carrier's priorities for home networking, and how those led to the choice of HomePNA, announced in August. (See AT&T: Hold the MoCA.)

The choice was interesting given that Verizon Communications Inc. (NYSE: VZ) had placed an early bet on Multimedia over Coax Alliance (MoCA) , a standard for networking over coax. (See Entropic, Verizon Serve Up MOCA.)

Coax is the best option for home networks, Reed said. It's shielded and unregulated, meaning just about any radio frequency (RF) signal can be sent on the cables without causing or receiving interference.

But AT&T wanted to apply one technology to all homes. "If you go into multidwelling facilities or apartment buildings, you don't always have access to coax," he said.

So AT&T sought a technology that would work on twisted-pair copper, and that ruled out MOCA. Reed said MOCA uses so much RF spectrum that it can't be run on twisted-pair -- at least, not without upsetting the Federal Communications Commission (FCC) .

That made HomePNA a likely choice, but the kicker was the suite of diagnostic tools HomePNA was able to offer. AT&T can pinpoint the exact locations of interference or signal loss within a home, Reed said.

And while Reed didn't mention it, HomePNA 3.1 also happens to have the approval of the International Telecommunication Union (ITU) , as the standards body announced yesterday. (See ITU Approves Home Standard.)

It all comes down to the idea of IPTV as a managed service, or an "assured" service, as Reed put it. Because IPTV targets a mainstream crowd, as opposed to early adopters who want to build home networks, AT&T and other carriers are starting to treat home networks as parts of their own networks. (See RBOCs Want Inside Your House.)

"We have to be very careful about selecting technologies that map themselves well into this assured services network," Reed said.

That's why Reed gives the thumbs-down to 802.11n wireless LANs as an option for U-verse. Wireless networks are subject to interference from uncontrollable sources, such as a neighbor's wireless network, Reed said. That unpredictability could translate into more service calls and more trips for technicians out to customers' homes -- the kinds of things any carrier is hoping to avoid.

"The feeling in the labs is, any wireless technology for moving video would have to be treated as an adjunct to the primary service. Dedicated wires are, for an 'assured service,' the mechanism of choice for the home network."

Still, AT&T evaluates every HomePNA competitor it's aware of, even the wireless ones, and Reed said he's also open to the idea of wannabe universal standards from the Digital Living Network Alliance, the Home Gateway Initiative (HGI) , and the ITU's G.HN effort. But any home-networking standard AT&T accepts would have to match the diagnostic capabilities of HomePNA, he said.

— Craig Matsumoto, West Coast Editor, Light Reading

<<   <   Page 2 / 2
paolo.franzoi 12/5/2012 | 3:13:12 PM
re: Why AT&T Likes HomePNA
Your coax is in your home and is therefore a home network. It is not a home IP network, which is your implication. If there is a serious amount of IP video in the home over the next couple of years, then there will be lots of need for high speed IP in-home networks. Verizon is already deploying one of these technologies (MOCA) for its home networking. AT&T is planning to (and perhaps starting to) deploy HPNA3 for this same function.

paolo.franzoi 12/5/2012 | 3:13:11 PM
re: Why AT&T Likes HomePNA
So, if I put 1 billion TVs in my home then my home video network is of value?

Or are you saying there is no value in the connection of the 100s of millions of TVs in the US alone?

Or are you simply trying to make the facts fit your perspective? Which is that the high bandwidth application of the Internet is not for wonderous new services, or great educational purposes, but for porn.


rjmcmahon 12/5/2012 | 3:13:11 PM
re: Why AT&T Likes HomePNA Your coax is in your home and is therefore a home network. It is not a home IP network, which is your implication.

No, that is not my implication. I'm not really an IP bigot.

I'll try again. Maybe you can understand my perspective but, so far, even after years of discussion and debate, I haven't had much luck.

The value of a network is not driven by some technology or protocol but rather by attributes such as connectivity and policies impacting access. So my position is more aligned with Metcalfe's law where value is driven by the number of users and devices able to communicate. Telco TV, DBS, Cable TV are all extremely limited with respect to their networking capabilites, not due to bandwidth, but due to lack of connectivity.


"Metcalfe's law states that the value of a telecommunications network is proportional to the square of the number of users of the system (n2). First formulated by Robert Metcalfe in regard to Ethernet, Metcalfe's law explains many of the network effects of communication technologies and networks such as the Internet and World Wide Web.

The law has often been illustrated using the example of fax machines: A single fax machine is useless, but the value of every fax machine increases with the total number of fax machines in the network, because the total number of people with whom each user may send and receive documents increases."

It's obvious to anyone really looking objectively at the issue that walled gardens are not what next generations need (or really want) but rather those who bring the unfettered ability to communicate, in any form, to the table are those who are adding value to society.

So, build what adds value and don't waste time on ideologues and their misguided demands, such as the current FCC who claim "market competition" and "facilities based competition" solves all problems.
rjmcmahon 12/5/2012 | 3:13:10 PM
re: Why AT&T Likes HomePNA So, if I put 1 billion TVs in my home then my home video network is of value?

No. TV sucks for lots of reasons. It's not a democratic medium and is getting worse. Read the following for some insights.


Or are you saying there is no value in the connection of the 100s of millions of TVs in the US alone?

Again, more all or nothing thinking. This is distorted thought it plays into the emotion. Let's try some reason for a change.

Of course there is value to broadcast TV. But that's not really the question, is it seven.

How do we progress from where we are. The telcos can invest our revenue streams towards modern, symmetric and democratic communication networks or they can become a fourth TV provider. The latter is a ver poor use of resources.

Or are you simply trying to make the facts fit your perspective?

No. I'm not stuck on winning an argument to save face. Satisfying my ego comes from other things, like doing the right thing when I know it's the right thing.

Which is that the high bandwidth application of the Internet is not for wonderous new services, or great educational purposes, but for porn.

Of course you have to throw porn in there. That's always been your fallback position. Give people freedom and they'll choose porn!!! Oh no!!! The kids will become a victim sexual explotation!! Just more fear tactics, seven. Get's a bit tiresome.

I believe you know the answer seven. You're stuck for some other reasons. I don't have the skills to help you. Sorry.
paolo.franzoi 12/5/2012 | 3:13:09 PM
re: Why AT&T Likes HomePNA

You still don't believe anybody has connected to FiOS and its all a lie yet right?

I provided a rather straightforward answer to a question on power line networking and you again went off. The fact that two other 100 Mb/s home wiring technologies have come into existance does not deter your commentary that they can't have a market because nobody needs them. The fact that one of the (MOCA) is already moving to a volume product does not deter you.

I have concluded that in fact you will delude yourself into anything. You simply choose to ignore facts that are not inside your model. Better broaden your mind and figure out that the reason your applications don't exist outside of entertainment is just that - they don't exist. Entertainment on the other hand is chewing up gobs of bandwidth and new applications are occuring here everyday. Porn, Gaming, Gambling, and other forms of recreation are the real deal. That is where the money is.

praxis7 12/5/2012 | 3:11:35 PM
re: Why AT&T Likes HomePNA The difference between the RF bandwidth from a cable/sat company and Ethernet/IP bandwidth is the issue. IPTV is inherently Ethernet - not directional RF and so all that cable/sat RF bandwidth means next to nothing to the IPTV company. Yes coax has a lot of bandwidth, but making bandwidth available for Ethernet in every room of the home using coax is not so simple. Neither HPNA nor MOCA claim 100% home coverage where RF cable does. It's apples and oranges.

Regarding 100Mbps from HPNA or MOCA, I don't think there is any danger of that for a while. Look at Corinex's HPNA 3 adapters and you'll note that in real-world tests, they average about 20Mbps to 40Mpbs. I suspect MOCA will test out similarly or maybe worse.

There are two issues: 1) The technologies of Ethernet over coax/phoneline (HPNA & MOCA); 2) The business problems associated with using those products.

Some 70% of US homes have coax. What do IPTV providers do with the other 30%? And of the 70%, what percentage have coax/splitters that will work with HPNA/MOCA right out of the box? I suspect it is far less than 70%. Big coax-based cable companies have to fix the coax plant in homes in 30% to 40% of their installations on average. That for regular cable that is not as sensitve to splitter and cable grade as Ethernet over coax.

The reality is that if the telcos want to compete with the coax based cable companies, they need a solution that is plug-and-play (and potentially subscriber self-installable) and does not require they train their installers on coax installation/repair, stock coax cable, splitters, and other supplies. IPTV providers are natively cat-5+, not coax. If they fix-up the coax plant in a home, when that home churns they may well have helped out their coax-based competitor.

These are the kinds of business issues telcos have to deal with. And it boils down to how fast and easily the IPTV company can install their services.
nmandel 12/5/2012 | 3:04:08 PM
re: Why AT&T Likes HomePNA Forgive me for the late reply. Allow me to add few comments to your analysis and other messages in the forum:
1.The VOD and IPTV applications requires a huge throughput capabilities from the home network (as well as from the access side) in order to support several HDTV channels simultaneously. This means that the current home network will be useless in the near future for the new applications. Allow me to remind you that currently broadband access in many countries is already 20 -30Mbps (and in few 100Mbps).
2.For multimedia application GǣstreamingGǥ and GǣQuality of Service (QoS) is the key word. More than that for multimedia application you are using UDP and not TCP protocol.
3.Currently we are facing 3 (4) high-speed technologies: HomePNA3 (& MOCA), HomePlugAV and Wi-Fi 802.11n (currently draft2). For streaming application the HomePNA3 (& MOCA) are the most stable and can support the VOD/IPTV requirements.
4.HomePNA3 is currently 128Mbps. By the end of the year you will have 160Mbps and by end of 2008 G 320Mbps. The current technology (128Mbps) supports about 50 Mbps for TCP protocol and above 90Mbps for UDP protocol. This is more than enough for the current VOD/IPTV services.
5.Standard Ethernet can provide higher speeds but requires cable installation which service providers are trying to eliminate.
6.Last argument is the overall system installation cost. At this stage the HomePNA3 is the cheapest one.
7.I think all the above is AT&T drive to choose HomePNA3.
<<   <   Page 2 / 2
Sign In