In news that is huge for VoWi-Fi, the Wi-Fi Alliance announced on June 30th a new certification program, “Voice-Personal.” Eight devices have already been certified under this program, including enterprise access points from Cisco and Meru, a residential access point from Broadcom, and client adapters from Intel and Redpine Signals.

Why is this huge news? Well, as the press release points out, by 2011 annual shipments of cell phones with Wi-Fi will be running at roughly 300 million units. The Wi-Fi in these phones will be used for Internet browsing, for syncing photos and music with PCs, and for cheap or free voice calls.

The certification requirements for Voice-Personal are not aggressive: only four simultaneous voice calls in the presence of data traffic, with a latency of less than 50 milliseconds and a maximum jitter of less than 50 milliseconds. These numbers will produce an acceptable call under most conditions, but a network round-trip delay of 300 ms is generally considered to approach the limit of acceptability, and with a Wi-Fi hop at each end running at the limit of these specifications there would be no room in the latency budget for any additional delays in the voice path. The packet loss requirement, 1% with no burst losses, is a very good number considering that modern voice codecs from companies like GIPS can yield excellent sound quality in the presence of much higher packet loss. This number is hard to achieve in the real world, as phones encounter microwave ovens, move through spots of poor coverage and transition between access points.

Since this certification is termed “Voice-Personal,” four active calls per access point is acceptable; a residence is unlikely to need more than that. Three of the four access points submitted for this certification are enterprise access points. They should be able to handle many more calls, and probably can. The Wi-Fi Alliance is planning a “Voice-Enterprise” certification for 2009.

There are several things that are good about this certification. First, the WFA has seen fit to highlight voice as a primary use for Wi-Fi, and has set a performance baseline. Second, this certification requires some other certifications as well, like WMM power save and WMM QoS. So far in 2008, of 99 residential access points certified only 6 support WMM power save, and of 52 enterprise access points only 13 support WMM power save. One of the biggest criticisms of Wi-Fi in handsets is that it draws too much power. WMM power save yields radical improvements in battery life - better than doubling talk time and increasing standby time by over 30%, according to numbers in the WFA promotional materials.

The forthcoming transition to digital TV transmissions will free up about half the spectrum currently allocated to TV broadcasters. This freed-up spectrum was the subject of the FCC’s just-concluded 700MHz Auction, which yielded about $20 billion in license fees to the government. The fate of the other half of the TV spectrum, the part that will remain assigned to TV broadcasts after the digital transition, remains in contention.

This spectrum will be shared by licensed TV broadcast channels and wireless microphones, but even so much of it will remain mostly unused. These chunks of spectrum left idle by their licensees are called “White Spaces.” The advent of “spectrum sensing” radio technology means that it is now theoretically possible for transmitters to identify and use White Spaces without interfering with the licensed use of the spectrum.

The FCC has issued a Notice of Proposed Rulemaking and a First Report and Order to explore whether this is a good idea, and if so, how to handle it.

The potential users of the White Spaces have formed roughly two camps, those who see it best suited for fixed broadband access (similar to the first version of WiMAX), and those who see it as also suited for “personal/portable” applications (similar to Wi-Fi).

Google, along with Microsoft and some other computer industry companies, advocates the personal/portable use. The FCC’s Office of Engineering and Technology (OET) is currently lab-testing some devices from Microsoft and others to see if their spectrum-sensing capabilities are adequate to address the concerns of the broadcast industry, which fears that personal/portable use will cause interference.

Google filed an ex-parte letter with the FCC on March 24th, weighing in on the White Spaces issue. The letter is well worth reading. It concedes that in the introductory phases it makes sense to supplement spectrum sensing with other technologies, like geo-location databases and beacons. The letter asserts that these additional measures render moot the current squabble over a malfunction in the devices in the first round of FCC testing, and that the real-world data gathered in this introductory phase would give the FCC confidence ultimately to repeal the supplemental measures, and perhaps to extend open spectrum-sensing uses to the entire radio spectrum, leading to a nirvana of effectively unlimited bandwidth for everybody.

The kicker is in the penultimate paragraph, where Google recycles an earlier proposal it made for the 700MHz spectrum auction, suggesting a real-time ongoing “dynamic auction” of bandwidth. Google now suggests applying this dynamic auction idea to the white spaces:

For each available spectrum band, the licensee could bestow the right to transmit an amount of power for a unit of time, with the total amount of power in any location being limited to a specified cap. This cap would be enforced by measurements made by the communications devices. For channel capacity efficiency reasons, bands should be allocated in as large chunks as possible. The airwaves auction would be managed via the Internet by a central clearinghouse.

Current expectations are for spectrum-sensing use of the whites spaces to be unlicensed (free, like Wi-Fi). Now Google appears to be proposing “sub-licensed” rather than unlicensed spectrum use. The word “auction” implies that this could be a revenue producer for TV broadcast licensees, who received their licenses free from the government. This is a very different situation than the original context of the dynamic auction proposal, which applied to spectrum for which licensees paid $20 billion. Depending how it is implemented, it could fulfill the telcos’ dream of directly charging content providers for bandwidth on a consumer’s Internet access link, a dream that Google has opposed in the network neutrality wars. Google may ultimately regret opening the door to this one, even though it presumably sees itself cashing in as the ideal candidate to operate the “central clearinghouse.”

Update April 10th: Interesting related posts from Michael Marcus and Sascha Meinrath.

It took a while, and 261 rounds of bidding, but its over. Click here for the write-up from Wired.

The attractive thing about the 700 MHz spectrum that was freed up by the move to digital TV broadcasting is that transmissions at these frequencies pass through walls. The unusual thing about the “C Block” of this spectrum is that the FCC attached “open access” conditions to the license. This was at the behest of the computer industry, spearheaded by Google, who may even have made a bid on this block. But as the Wired story points out, Google had already won their victory with the imposition of the open access rules - winning the spectrum would have been more of a headache for them than losing it.

Don’t confuse the spectrum licensed in this auction with the White Spaces spectrum. The White Spaces spectrum is the spectrum that the TV broadcasters retained for their transition to digital transmissions in February 2009. The White Spaces issue is still unresolved by the FCC. The FCC is deliberating over whether to allow unlicensed use of the digital TV spectrum when it is not being used by a TV broadcast (hence “White Spaces.”) This use depends on effective functioning of “cognitive radio,” which lets transmitters sense by listening (and other means) when spectrum is available for use. If the FCC allows it, they still have to decide whether to allow only fixed broadband replacement like 802.22, or to allow “Personal and Portable” use as well.

Intel published a white paper last year about a trial deployment of 802.11a as a replacement for wired Ethernet at a 5,000 person campus. The results were lower costs and happier workers. This was just for PC connectivity. The dual-mode phone phase of the deployment is still to come.

There are several interesting findings in the white paper. First, while the latency of the network increased somewhat, the difference was imperceptible to the users. Second, Intel chose to abandon the VPN, relying on 802.11i for security. This made joining the network faster and easier.

The decision to use 802.11a was presumably for the greater capacity (more non-interfering channels than 11g), and for the cleaner spectrum. 802.11n is superior to 802.11a in capacity and rate at range. This means that what was doable with 11a will be even easier with 11n.

There’s a great lead article in the Wall Street Journal this morning, explaining how cell phone manufacturers and cellular service providers have opposing interests concerning phone features and how they are delivered.

The article pitches the fight as one over which of these two will end up controlling the data services on the phone. The interests of consumers would be best served by neither of these outcomes. The essence of the Internet is its openness. Nobody controls the services. Ideally, the mobile model of Internet access will follow the wireline model, where neither the device manufacturer nor the Internet access provider exerts any control over the content of the network data packets.

The estimable Brough Turner has written at length on the topic of Net Neutrality.
The first thing to read on this topic is his blog entry “Why there’s no Internet QoS and likely never will be”. In this article he makes the point that the only place where QoS measures make a difference is in the access link, and that the best way to ensure access link QoS is by putting a traffic shaping device behind your broadband modem. So no need for anything from your ISP beyond what you already expect, namely bandwidth and uptime.

In this article, Brough advocates a long view, focusing on increasing last-mile bandwidth, pointing out the danger of unintended consequences of regulation. He makes the point that fiber is not a natural monopoly, in the sense that there is adequate revenue per square mile in moderately densely populated cities to sustain multiple runs of fiber to each home. He identifies rights of way restrictions as the real barrier to last mile competition. In a similar spirit, he advocates opening up spectrum for license-exempt operation for last mile access. This article has similar arguments.

In this later article, Brough backs off a little to what seems to me to be a better position, namely regulating dark fiber, and fostering competition above it.

By 2007, Brough had nailed his colors to the mast. In this review of Susan Crawford’s paper, “The Internet and the Project of Communications Law,” he says:

I’d really like to see a national strategy to get as much of the population on dark fiber as possible.

And just a couple of months ago, he proposed a way to do it:

…unfettered municipal experimentation by any of the 22,000 municipalities in the US and/or interested community groups.

So there you have it. A relatively simple, seemingly doable solution to Net Neutrality, implementable by the grass roots, thought through by a smart guy who knows the subject inside and out.

I’m aboard!

Cisco has two main customer constituencies: network service providers and business IT departments. One of WebEx’s crown jewels is its MediaTone network. This is a global private network, with dedicated fiberoptic links and multiple peering points to the Internet. If Cisco doesn’t sell this off, they will be competing with their customers in one of their primary markets. Unlikely to fly, though Cisco sometimes doesn’t seem to mind treading on toes.

This leaves the remainder of WebEx, the application (SaaS) side. It’s a natural complement to two of Cisco’s current business lines, filling a gap between their Unified Communications Manager (VoIP PBX) products and their high-end telepresence offerings.

As Cisco gets into more and more of the software services that run over IP networks, they end up competing more and more with Microsoft among others, and in an odd way for an Internet company.

Cisco rode the Internet Protocol to the stars. An article of faith amongst the IP illuminati is that the network must be stupid. This means that we conceive of the Internet as an amorphous connectivity cloud with computers on its periphery. Some of them are clients and some of them are servers. The Internet doesn’t care which is which. This is very powerful, because anybody with an IP address can set up a web site (a.k.a network service). This is anathema to the traditional network service providers who want to provide value (get revenue) in the network beyond mere connectivity. The Internet world (like Google) and the PC world (like Microsoft and Intel) love the stupid network model because it lets them innovate. The owners of the wires hate it because it forces them into the role of mere connectivity providers, since they are incapable of innovation at the service level.

But Cisco’s bread and butter is network equipment. Cisco doesn’t sell servers. So every service that migrates from the stupid network model to the intelligent network model increases Cisco’s potential market. Cisco hasn’t yet apostatized, but these actions are building gravitation in that direction. They have already ported their IP PBX to IOS, and they are allegedly even warming up to IMS!

Background
Forbes article on the acquisition.
CNET interview with WebEx CEO Subrah Iyar.

Network neutrality is a contentious topic, particularly because billions of dollars of revenue hinge on the way it pans out. Because of these high stakes, partisans are motivated to use all the tools of rhetoric to argue for their positions. One way that the debate is commonly misrepresented is as one between regulation and the free market. It is actually a debate between more or less regulation. A free market is one where competition leads to abundant choice, and where consumers have the option to select one service over another. A monopoly is one where there is no choice, and consumers must take or leave what they are offered. Internet access is not a monopoly, but it is close, because of the limited number of suppliers in any particular location. The supposed purpose of network neutrality regulation is to preserve the current vibrant free market of internet services (services provided via the internet, as opposed to internet access service.)

There are several arguments against network neutrality. One is that it is impossible to enforce, that the Internet access providers will subtly strangle third party services by intentional incompetency if the regulators try to force them to stay open. This seems to be one of the ways that the CLECs were held off until the demise of UNE-P. A second one is that there is no need for regulation because there is no threat to third party services. This argument points out that the commercial Internet has been going strong for over a decade, so why fiddle with it when it is working so well? The rebuttal to this argument is that Ed Whitacre, CEO of AT&T has famously said that they intend to double charge providers of services over the network like Google: once for their own internet access like now, and a second time for their customer’s network access, additional to the one that customers are already paying for their own internet access. If the access providers succeed in this it will become much harder for small companies to offer new services on the Internet.

Another anti-network neutrality regulation argument is that network neutrality is good, that we have it now, and that any Internet Access Provider that tries to mess with it will get savaged by customer opinion and lose customers to competitors who keep their network open. This argument assumes that there is a free market in Internet access, and that a move by the telcos to double-charge service providers will not be immediately echoed by the MSOs. Or in a slightly modified version, it assumes that if the telcos and MSOs jump on this together, public outcry will force immediate regulatory reversal, so why try to fix something that ain’t broke yet?

Putting on the other hat, the access network operators have poured and continue to pour billions of dollars a year into upgrading their networks. They certainly deserve a fair return on their investments. The problem is that competition is so vigorous that neither the telcos nor the MSOs can afford to raise their rates to pay for all this build-out. But service providers (Google) have cash coming out of their ears, and they would happily share some of it in exchange for guaranteed priority of their content on congested networks.

OK, let’s take that hat off again. While Google does indeed have money coming out of its ears, and probably could afford to send some of it the way of the access providers, what about service providers with no spare cash, in other words, startups? The Internet has been a hothouse of innovation because of its spectacularly low barriers to entry. It costs virtually nothing to set up a new Internet service. Loss of network neutrality would shut down this innovation, because the telcos and MSOs would get to decide which of these services survived. Their track record at this stinks. If they had an inkling how to do it there would be no Google, UTube, MySpace or eBay, because the telcos would have put services like this into place 20 years ago. But they didn’t. Large companies are structurally incapable of this kind of radical innovation, while a vibrant free market does it naturally. This is because evolution by natural selection is a far more potent force than executive edict, and “let a thousand flowers bloom” is a vastly more fertile approach than “let’s not do anything risky.” Looked at this way, the telcos’ presentation of tiered service as “free market” rings hollow.