Home networking to get easier. And harder.
Two new home networking standards – G.hn and WHDI – are on the way. Each has its respective champions, who expect to not just earn a place beside MoCA, Wi-Fi, HPNA, Ethernet, HDMI, HomePlug, UWB and other extant communications standards, but to eventually supplant most of them.
The advantage most of those standards have is that they’re all here now and are pretty well proven in deployments.
What G.hn and WHDI have going for them is that each of those extant standards has drawbacks in home networks (HN). And even though some of those other standards have been around a while, home networking is still at a rudimentary stage; where/what/when-I-want is still in the future.
That opens the door for an advanced standard. The question is does it open the door for two of them? Or three? Or four … ?
The WHDI Consortium has just published its WHDI 1.0 specification. It is being delivered by some of the same people who brought the world HDMI, including Leslie Chard, the president of the WHDI Consortium’s limited partnership, WHDI LLC.
WHDI supports the delivery of equivalent video data rates up to 3 Gbps (including 1080p/60 Hz) in a 40 MHz channel in the 5 GHz unlicensed band, the consortium said. Range is beyond 100 feet, it penetrates through walls and latency is less than 1 millisecond.
The key point, explains Chard, is that it can transmit uncompressed HD, so it tends to be less susceptible to interference that can sometimes play havoc with compressed video on wireless networks, and better yet avoids the format problems associated with having multiple CE devices that incorporate different media players, codecs, DRM, etc.
WHDI’s backers include LG, Hitachi, Sharp, Sony, Samsung and Motorola – specifically Motorola’s set-top box group. The plan follows the map of the HDMI rollout; WHDI proponents expect to begin selling TVs, video cameras and other CE devices by Q3 2010, Chard said, and eventually it will percolate through to be incorporated in set-tops.
WHDI can be built directly into Wi-Fi chips, keeping the cost low.
Piggy-backing on Wi-Fi silicon is the same strategy espoused by WiMAX advocates. Chard said there’s no technological reason that all three (Wi-Fi, WiMAX, WHDI) couldn’t be integrated into a single chip, assuming a compelling market reason to do it.
WHDI devices would be able to communicate directly with each other. In other words, they would not have to connect through a WHDI router.
WHDI is designed to automatically create a network of WHDI-enabled products. Every home network would get its own 100-digit key. Whenever a new WHDI device is brought into proximity, the HN owner is given the option to approve the device – essentially authorizing it. This keeps the HN secure and provides a mechanism for rejecting any other WHDI devices – for example, those being used by a neighbor.
In terms of content security, WHDI relies on HDCP revision 2.0 to provide Hollywood-approved security and digital content protection (HDMI relies on HDCP).
Chard figures that since WHDI automatically builds secure home networks, it can be a vehicle that service providers can leverage to monetize the extension of their services to a new set of devices.
Meanwhile, G.hn, a wireline standard, is sailing toward ITU ratification. The spec is now stable enough that silicon designers can go ahead and produce chips.
Michael Weissman is vice president of marketing in North America for CopperGate Communications, and as such thinks in terms of unit sales. He said the market for HN chips is currently about 20 million to 30 million a year.
The reason it’s not bigger is because there are too many competing, incompatible standards, with multiple standards per wire type (coaxial, phone line, powerline).
G.hn would work over all three wire types, providing data rates in excess of 1 Gbps.
Where WHDI is being backed by CE companies, G.hn is being driven by the HomeGrid Forum, whose members feature prominent chip companies, including Intel, Infineon, Ikanos, Sigma Designs and CopperGate. Other members include CE-maker Panasonic (notably not currently a member of the WHDI Consortium), British Telecom and retailer Best Buy.
If the industry were to converge on a single, advanced, converged standard – G.hn – the market could quickly grow to 150 million chips a year and eventually – in five or six years, Weissman believes – to as many as a billion chips a year.
AT&T is going to go with G.hn (AT&T is an active participant in development of the standard), Weissman said, because it can save millions and millions of dollars by doing so.
HN can be integrated into a single chip. For set-top manufacturers, that could lower the STB’s bill of materials by 20 percent, by Weissman’s calculation.
With AT&T on board, that’s an automatic market of millions of units, plus millions more as others companies that typically follow AT&T’s technological lead embrace G.hn, as well.
Will the CE guys bite?
“The telco guys will do this anyway. They may not get Samsung and Sony to agree, but then Vizio will come in and kick their asses.”
Seems straightforward, except it’s not.
Of course, proponents of existing standards are not standing still. They are all improving the capabilities of those extant standards.
Oh, and if cable and telcos and satellite providers, and CE companies, and chip companies and PC makers wait a couple of years, the IEEE is working on yet more variants of the 802.11 Wi-Fi standards, with double letter designations (-aa, -ac, -ad, etc.), and it’s likely to have data rates in excess of 1 Gbps, with greater robustness, and including QoS management. …