The Bluetooth SIG put out a press release the other day saying that by mid-2009 a specification that includes Wi-Fi as an “alternate MAC/PHY” will be released.

This is the relevant part of the release:

What we’re doing is taking classic Bluetooth connections – using Bluetooth protocols, profiles, security and other architectural elements – and allowing it to jump on top of the already present 802.11 radio, when necessary, to send bulky entertainment data, faster. When the speed of 802.11 is overkill, the connection returns to normal operation on a Bluetooth radio for optimal power management and performance.

So this innovation specifies an interface above the Bluetooth MAC layer that enables Bluetooth session data to flow through a Wi-Fi radio instead of a Bluetooth one.

The press release says that the Alternate MAC/PHY will be used to do things like:

Wirelessly bulk synchronize music libraries between PC and MP3 player.
Bulk download photos to a printer or PC.
Send video files from camera or phone to computer or television.

But the new specification will only be useful in devices that already have both Wi-Fi and Bluetooth, and Wi-Fi can do these things without any assistance from Bluetooth. So the increment in value may be small; but Bluetooth can still bring something to the table. According to Wikipedia,

The Bluetooth Radio will still be used for device discovery, initial connection and profile configuration, however when lots of data needs to be sent, the high speed alternate MAC PHY’s will be used to transport the data.

It seems oddly limiting not to go the whole way, and enable Bluetooth applications to run on top of Wi-Fi even when there is no Bluetooth radio.

George Ou of ZDNet reports that the 802.1X authentication techniques used on some Wi-Fi handsets may be vulnerable. The problem is that these handsets may not validate the certificate from the authentication server. This design choice speeds up roaming, but means that the handset could disclose user login credentials to a sophisticated, determined attacker. Ou suggests using WPA-PSK with a long password instead of 802.1X with these handsets.

Vocera’s documentation, which Ou references, has more depth on the performance trade-offs of various Wi-Fi security options.

Redpine Signals has announced that it is sampling a low power 802.11n chip suitable for cell phones. A reference design was certified in January, making it the first handset-grade 802.11n chip to market.

One of the major benefits of 802.11n is MIMO, so you might think that since a handset is unlikely to have multiple antennas, 802.11n isn’t going to help much. Actually, it will make an enormous difference in reliability and range, and consequently throughput. I wrote before about the array of improvements incorporated in 11n. The one of greatest interest in this context is Space-Time Block Coding (STBC).

The WFA website shows 90 Access Points (APs) certified for 802.11n, but STBC is optional in 11n, not mandatory, and not all the AP chipsets support it. The main makers of AP chipsets are Atheros, Broadcom and Marvell. None of these have mentioned STBC until recently. But now Broadcom says it is in the BCM4322, which is set to ship in the first quarter of 2008, and Marvell says it is in the TopDog 11n-450, which is scheduled to ship in 2Q 2008.

This Techworld article has a good discussion of the current state of enterprise 11n access points, noting that multi-radio APs are currently too power-hungry to be powered over Ethernet (PoE).

In May 2007 I showed a chart of dual-mode phone certifications by time. Certifications have continued to grow since then, as the updated graph below shows. These numbers are pretty raw, for example six certifications in November 2006 were for variations on a Motorola phone first certified in October. If you go back and look at the previous chart you will also notice discrepancies in the number of certifications for any particular month. These are presumably because of revisions at the WFA website.

From 2006 to 2007 smartphone certifications were essentially flat, going from 33 to 36, while feature phone certifications went from 11 to 21. These add up to 44 dual mode phone certifications in 2006 and 57 in 2007.

I have previously written about OpenMoko. It seems now that it was the drop before the deluge. Google’s Android appears to have gained good traction with Sprint and T-Mobile joining the Open Handset Alliance, with Dell rumored (update) to be planning an Android-based phone, and with Verizon expressing lukewarm support. Nokia has for some time sponsored open source handset software through Maemo.org, but this week it upped the ante with its acquisition of TrollTech. Trolltech is responsible for Qtopia, a semi-open source platform used in Linux-based phones. That makes four credible Linux-based mobile phone software platforms. Update: Make that five - the LiMo Foundation is a consortium of carriers (including NTT DoCoMo and Vodafone), phone makers (including Samsung, Motorola and LG) and others “dedicated to creating the first truly open, hardware-independent, Linux-based operating system for mobile devices.”

But a phone doesn’t have to be open-source to be an open application platform, and this category is just as vigorous, but better established. Nokia’s Symbian phones have always been open to an extent - there are over 2 million developers registered in Nokia’s developer organization, Forum Nokia. Then we have Microsoft. Microsoft claims that sales of Windows Mobile phones are set to double year-on-year, to 20 million units. Windows Mobile provides a sufficiently open application environment that applications like Skype run on it. The iPhone is not yet officially an open application environment, but there is still a healthy slate of applications from third parties for those with the stomach to take the unofficial route. This is scheduled to change in February when the open-ness goes official with the release of Apple’s SDK for the iPhone. So that’s three major open application environments for smart phones.

2008 is also the year that Wi-Fi phones will come into their own. The dam broke with the iPhone. Wi-Fi on the iPhone raises the bar for all the other smart phones, making Wi-Fi a baseline checklist item for the next generation of smart phones. Previously mobile network operators were fearful that Wi-Fi in a phone would divert traffic from their data networks. This fear led, for example, to AT&T’s removal of Wi-Fi from their version of the Nokia E61. But there is now new evidence. At last week’s IT Expo East I heard an unsubstantiated report that 60% of wireless data usage in December was by 2% of the phones: iPhones. If this is even partly true, it would demonstrate that a web-friendly phone will drive traffic on the cellular data network even when it has Wi-Fi.

Sony has added Skype to the PSP. The Sony CES website says:

Call friends, talk trash to fellow gamers or catch up with acquaintances via Skype for PSP system.

PC Magazine says that the PSP supports both SkypeIn and SkypeOut, so the PSP can substitute for a phone when you are at home or somewhere else where you have Wi-Fi access.

This isn’t a breakthrough, just another feather in the scale, tipping us towards a world where just about any connected device can make an internet phone call. The speed of this evolution is governed by the enormous legacy public telephone network; because of Metcalfe’s law anything that aspires to be a useful substitute for the phone system must first interoperate with it.

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.

802.11n incorporates all earlier amendments to 802.11, including the MAC enhancements in 802.11e for QoS and power savings.

The design goal of the 802.11n amendment is “HT” for High Throughput. The throughput it claims is high indeed: up to 600 Mbps in raw bit-rate. Let’s start with the maximum throughput of 802.11g (54 Mbps), and see what techniques 802.11n applies to boost it to 600 Mbps:

1. More subcarriers: 802.11g has 48 OFDM data subcarriers. 802.11n increases this number to 52, thereby boosting throughput from 54Mbps to 58.5 Mbps.

2. FEC: 802.11g has a maximum FEC (Forward Error Correction) coding rate of 3/4. 802.11n squeezes some redundancy out of this with a 5/6 coding rate, boosting the link rate from 58.5 Mbps to 65 Mbps.

3. Guard Interval: 802.11a has Guard Interval between transmissions of 800ns. 802.11n has an option to reduce this to 400ns, which boosts the throughput from 65 Mbps to 72.2 Mbps.

4. MIMO: thanks to the magical effect of spatial multiplexing, provided there are sufficient multi-path reflections, the throughput of a system goes up linearly with each extra antenna at both ends. Two antennas at each end double the throughput, three antennas at each end triple it, and four quadruple it. The maximum number of antennas in the receive and transmit arrays specified by 802.11n is four. This allows four simultaneous 72.2 Mbps streams, yielding a total throughput of 288.9 Mbps.

5. 40 MHz channels: all previous versions of 802.11 have a channel bandwidth of 20MHz. 802.11n has an optional mode (controversial and not usable in many circumstances) where the channel bandwidth is 40 MHz. While the channel bandwidth is doubled, the number of data subcarriers is slightly more than doubled, going from 52 to 108. This yields a total channel throughput of 150 Mbps. So again combining four channels with MIMO, we get 600 Mbps.

Lower MAC overhead
But raw throughput is not a very informative number.

The 11a/g link rate is 56 Mbps, but the higher layer throughput is only 26 Mbps; the MAC overhead is 54%! In 11n when the link rate is 65 Mbps, the higher layer throughput is about 50 Mbps; the MAC overhead is down to 25%.

Bear mind that these numbers are the absolute top speed you can get out of the system. 802.11n has numerous modulation schemes to fall back to when the conditions are less than perfect, which is most of the time.

But to minimize these fall-backs, 11n contains additional improvements to make the effective throughput as high as possible under all circumstances. These improvements are described in the following paragraphs.

Fast MCS feedback - rate selection.
Existing equipment finds it hard to track rapid changes in the channel. Say you walk through the shadow of a pole in the building. The rate may go from 50 to 6 to 50 mbps in one step. It’s hard for conventional systems to track this, because they adapt based on transmit errors. With delay sensitive data like voice you have to be very conservative, so adapting up is much slower than down. 11n adds explicit per-packet feedback, recommending the transmission speed for the next packet. This is called Fast MCS (Modulation and Coding Scheme) Feedback.

LDPC (Low Density Partity Check) coding
LDPC is a super duper Forward Error Correction mechanism. Although it is almost 50 years old, it is the most effective error correcting code developed to date; it nears the theoretical limit of efficiency. It was little used until recently because of its high compute requirement. An interesting by-product of its antiquity is that it is relatively free of patent issues.

Transmit beam-forming
The term beam-forming conjures up images of a laser-like beam of radio waves pointing exactly at the client device, but it doesn’t really work like that. If you look at a fine-resolution map of signal intensity in a room covered by a Wi-Fi access point, it looks like the surface of a pond disturbed by a gust of wind – it is a patchwork of bumps and dips in signal intensity, some as small as a few cubic inches in volume. Transmit beam-forming adjusts the phase and transmit power at the various antennas to move one of the maxima of signal intensity to where the client device is.

STBC
In a phone the chances are that there will only be one Wi-Fi antenna, so there will be only one spatial channel. Even so, the MIMO technique of STBC (Space-Time Block Coding) enables the handset to take advantage of the multiple antennas on the Access Point to improve range, both rate-at-range and limiting range.

Incidentally, to receive 802.11n certification by the Wi-Fi Alliance, all devices must have two or more antennas except handsets which can optionally have a single antenna. Several considerations went into allowing this concession to handsets, mainly size and power constraints. STBC is particularly useful to handsets. It yields the robustness of MIMO without a second radio, which saves all the power the second radio would burn. This power saving is compounded with another: because of the greater rate-at-range the radio is on for less time while transmitting a given quantity of data. STBC is optional in 802.11n, though it should always be implemented for systems that support 802.11n handsets.

Hardware assistance
Many of these features impose a considerable compute load. LDPC and STBC fall into this category. This is an issue for handsets, since computation costs battery life. Fortunately these features are amenable to hardware implementation. With dedicated hardware the computation happens rapidly and with little cost in power.

Wired has a story about the struggles of several municipal Wi-Fi deployments. Turns out that good coverage requires double the number of access points than planned, and that consumer subscriptons were an order of magnitude less than anticipated (1-2% instead of 10-25%).

The article holds out a few rays of hope: 802.11s for mesh deployments in residential areas, concentrate on high-traffic areas rather than trying to replace fixed broadband access, get the cities to become “anchor tenants.”

On the other hand, there still seem to be plenty of lower profile metro Wi-Fi deployments that are doing OK.

There’s an interesting article on femtocells in EETimes. It mentions the Femto Forum. It is a thoughtful look at the prospects for femtocells, a welcome counterbalance to the hype. The most telling quote is from the CTO of Ubiquisys:

We—that is, the femtocell ecosystem—probably have a two-year window to make our mark, ensure we come up with standard interfaces, and, above all, avoid fragmentation.

The two year comment is about beating Wi-Fi dual mode phones to the punch. But currently the primary driver for Wi-Fi in cell phones is feature inflation in high-end handsets, not FMC. In other words, there are really two dynamics driving Wi-Fi into handsets, FMC is the minor one and feature inflation is the major one; femtocells don’t affect the latter.

So if femtocells overcome their numerous challenges, FMC services for consumers will come mainly through femtocells. Femtocells will not impact Wi-Fi attach rate much, since Wi-Fi is becoming a checklist feature on high end phones. How useful the Wi-Fi in these handsets will be depends on how successful the phone makers are at keeping them open.