Receive new posts as email.
This site operates as an independent editorial operation. Advertising, sponsorships, and other non-editorial materials represent the opinions and messages of their respective origins, and not of the site operator. Part of the FM Tech advertising network.
Entire site and all contents except otherwise noted © Copyright 2001-2010 by Glenn Fleishman. Some images ©2006 Jupiterimages Corporation. All rights reserved. Please contact us for reprint rights. Linking is, of course, free and encouraged.
I'm a bit dubious about the vast amount of overhype pouring out about white-space spectrum after the FCC's new rules were set (PDF file): I don't see how what's postulated is possible. The TV channels in question are 6 MHz wide. Shannon's Theorem always wins. Channel capacity is a function of bandwidth mitigated by the level and ratio of signal to noise.
Wi-Fi can use 20 to 40 MHz channels in 2.4 and 5 GHz, and likely 80 MHz or more in future 5 GHz iterations. Without multiple radio receivers, encoding improvements in 802.11n over 802.11g bumped the raw rate from 54 Mbps to about 65 Mbps. Take two radios and 40 MHz, and your raw rate approaches 300 Mbps. Three and four radios and 450 Mbps to 600 Mbps.
White-space spectrum can only be used in 6 MHz blocks. Even with an extremely efficient encoding, I don't see how one can get more than 15 to 20 Mbps out of a channel. I've seen several statements that white-space networks will hit 400 to 800 Mbps.
The high power that's allowed--4 watts EIRP, the effective power after antennas--is pretty remarkable. Wi-Fi is limited to 1w EIRP, and in the nature of radio waves a 4fold increase in EIRP means more than 4fold improvement in distant reception. Correction: Wi-Fi is limited to 1W of transmitter power, but 4W of EIRP. The greater range of white-space devices will come from much, much lower frequencies, which carry further and penetrate better.
However, my understanding is that by the same token, MIMO is ineffective because MIMO doesn't work over long distances. It requires reflection over short spaces to provide the multiple spatial paths that boost speed. So by going long, you lose MIMO, and encode with a single radio.
Also by going high power, you lose the advantage of cellular infrastructure, whether for Wi-Fi or 2G/3G/4G mobile networking. The greater area you cover, the more your shared medium is split among users, even in a contention-free scheduled environment, which will likely not be what happens. As an unlicensed band technology, you could be contending with interferers of all kinds the higher power you use and greater area you cover.
Now perhaps the 400 to 800 Mbps figure is if you took all the white-space in a given market and bonded it together with a transceiver that could handle multiple separate bands at once. Or it's 400 to 800 Mbps of aggregated additional capacity, not for one device. (I can't run down the source of the number, only uses of it without reference.) By that token, Wi-Fi in 2.4 GHz and 5 GHz would add up to several Gbps.
I also haven't run through channel maps in given markets under consideration. How many channels are free in urban areas where a dense deployment would make sense? One colleague wrote to say he believes only a couple may be available for unfettered use.
I'm not even getting into the issue of competing licensed uses, the set aside by the rules of two channels in each market for wireless mics, and the ability for special-event permits and special-use mic permits (limited in area) that would trump pure unlicensed networking purposes, too.
Further, there's a canard circulating about how Microsoft has "covered its campus" with two white-space transmitters. That's true--that's not the canard. No, the problem is that Microsoft can serve the space but not the user base with two transmitters, even if the transmitters could handle the mythical 400 to 800 Mbps of raw throughput. (I should note that Microsoft has nothing to do with spreading this notion; Microsoft Research has a been a very reasonable driver, promoter, and engineer on this spectrum. Visit the Networking over White Spaces site for more information.)
Microsoft installed thousands of Aruba Wi-Fi access points across its campus a few years ago not just to provide coverage but also to provide bandwidth. WiMax has been hyped in the same way. You can have distance or speed but not both: the more area you cover, the more users you cover, the more you have contention for air space or time slots, and the less bandwidth available to each user.
White-space spectrum will spawn a lot of interesting devices, and I could see companies and buildings migrating to it for particular purposes. But replace a cellular network or Wi-Fi? I'm not seeing it yet. I welcome more insight in the comments.
The FCC's rules on white-space spectrum seem rather clever to me: The rules adopted today in a unanimous vote by FCC commissioners--a rarity on major policy issues--should be good for all parties. That's hard to achieve. The full rulemaking hasn't been posted yet; an FCC spokesperson told me via email it would posted later today.
White-space devices will have to consult a geolocation database that's regularly updated to avoid stepping on the toes of television broadcasters and other users, notably churches, sports venues, and performing spaces that rely on wireless mics.
To help preserve the use of wireless mics without interference, the FCC will require two channels in the former 7 to 51 (VHF up to UHF) range be reserved in each market for such transmissions. Wireless mic users can petition for additional space, apparently for special events, which means white-space transceivers will have to consult the database on a regular basis.
It's unclear at the moment how devices will grab database info. I could imagine a narrowband repeating transmission on a dedicated otherwise unused channel that would simply dump the local database. White-space devices will certainly require GPS receivers, and computation power and software to figure out the area in which they operate as a distance from other points that have to be offset from use.
The Wi-Fi Alliance put out a press release immediately after, noting that 802.11af is already in progress for adapting WLAN IEEE rules for white-space spectrum and options, and that the alliance already has a plan under way to set a certification programs for such devices.
White-space isn't "Wi-Fi on steroids," but it could be a great enhancement for particular purposes in which Wi-Fi doesn't reach far enough, and a cellular network restricts uses while being overkill and too slow.
There's a potential for competitive wireless networks to emerge over white-space spectrum, but the real-estate issue still intrudes. You might need 1/3 or fewer transmitters per square mile to build a Wi-Space network instead of a Wi-Fi one, but you still have to secure the right to mount gear.
JetBlue has signed a memo with ViaSat for next-generation satellite-backed Internet service for its current 160-aircraft fleet: JetBlue is opting for Ka-band satellite rather than ground-to-air (Aircell) or Ku-band (Row 44, Panasonic). As I understand it, the Ka-band satellites, which use higher frequencies, are designed for greater capacity, but that may be a matter of marketing rather than technology.
There are a few things that are odd about this deal. First, it's a memo, not a contract, which means this isn't a done deal, it's just an intended deal. Second, mid-2012 is a long way away; this is vapor service.
Third, JetBlue owns a sliver of valuable ground-to-air spectrum (1 MHz) that it acquired at the same auction at which Aircell bought the 3 MHz it uses for Gogo Inflight Internet. JetBlue's LiveTV division technically bought the spectrum, and so far it hasn't done anything interesting with it (at least in public).
I understand the site at which this article may be found is called FlightGlobal, but it's peculiar to not mention Aircell or Gogo, given that Aircell has equipped over 1,000 aircraft that fly over the US with Internet service, with about 1,000 more committed by the end of 2011.
Virgin Mobile's unlimited, no-contract data plan seems to have rattled AT&T's cage: Virgin back on 23 August announced a change in its no-contract plan options. Instead of four tiered plans, the highest offering up to 5 GB used within 30 days (on Sprint's network) for $60, there would be two: a $10/10-day/100 MB option and unlimited 30-day usage for $40.
That so undercut the rest of the market, I was wondering if there would be any response. Verizon has long offered a one-day $15 data pass, which always seemed overpriced to me since the market it was trying to reach were those with otherwise inactive 3G cards or MiFis.
AT&T's response appears to be a modest rejoinder. Three tiers: $15 for 100 MB used within a day, $30 for 300 MB used within a week, and $50 for 1 GB used within a month.
What AT&T doesn't seem to still realize is that Virgin Mobile's deal can be paired with a $100 MiFi (no contract), meaning that a few months of AT&T-priced usage would be outweighed by cost savings and flexibility. AT&T doesn't offer a MiFi-like device, and thus service is limited to laptop cards and notebooks.
It's a step in the right direction, as was AT&T's change to metered 3G broadband with reasonable overage charges for heavier users.
Bloomberg reports Skyhook Wireless has sued Google over two separate matters: It's no surprise to me that Skyhook might maintain it has patents that Google was violating for deriving location from Wi-Fi signals. Skyhook goes way back, when Google wasn't even showing ads on its search results, and Skyhook was still developing its initial database. The suit reportedly alleges four patents were violated.
However, the other charge in the suit is more surprising. Skyhook says that Google threatened Android handset makers Motorola and Samsung in a way that I didn't think was even possible.
Android is an "open" operating system in name only. Sure, you can get the source code and mess around with it, but there are no mainstream generic Android phones that work on any carrier, and no carrier-sold phones are simple to crack open and do what you will.
"Open" refers to a carrier's ability to modify the phone's software to its will, not the consumer or developers'. In fact, many Android phones come with garbageware installed on the phones' home screen, with no way to remove it.
Skyhook alleges that Google's Android chief, Andy Rubin, specifically pressured Motorola by stating that with Skyhook's technology on board, Motorola phones would be in violation of "Android licensing terms." Strange, for an open system. Samsung apparently also was pressured to remove Skyhook's software.
Update: I've read the lawsuit about Google interfering with business partners, and the specific issue at stake for Motorola and (ostensibly) Samsung was the use of the "Android Compatible" brand and program; without this certification, a vendor can't participate in the Android Marketplace, among other things.
Apple recently removed Skyhook Wireless technology from new versions of its iOS operating system, and is gathering location information itself. But no threats were alleged.
Very interesting story out of New York City: Cablevision and Time Warner Cable agreed to spend $10m to build out Wi-Fi in 32 city parks as part of the requirements for renewing cable franchises in the city. The country is divided into thousands of cable franchise zones, in which local bodies negotiate with cable firms to allow monopoly or limited competitive access to rights of way and other resources in exchange for typically a gross-revenue fee, public-access and government channels with budgets and facilities, and other add-ons.
While franchise boards are prohibited by law, regulation, and court decision from considering broadband and VoIP service as a condition of renewal--only the FCC can regulate broadband, and voice is a separate state regulatory domain--this is a neat twist. The NY negotiators figured out that they can require broadband to be offered.
The New York Daily News (a competitor to Cablevision-owned Newsday) reports that the service will be available for 30 minutes free each day to users, and then charged at a rate of 99 cents per day. Correction: My brain apparently couldn't cope with the fact that it's 30 minutes per month! In three 10-minute sessions, no less. That's fairly ridiculous.
Many New York parks have free Wi-Fi through various business districts and other sponsorship, such as Bryant Park.
WiFi Salon at one point had the contract to provide service in several parks, and had planned to use sponsorship as the driver. That deal with city parks ended in late 2008.
I've always wanted to put the country-music sweetheart into a headline: Dolly Parton, megachurch pastors, and theatrical promoters object to white-space spectrum rules proposed by the FCC in 2008, that would allow unused television frequencies in any market to be employed for Wi-Fi-like networking with far higher signal strength. The low-frequency spectrum can also penetrate walls and obstacles far better than the 2.4 and 5 GHz ranges used for unlicensed Wi-Fi.
The opposition from that group was related to wireless mikes that rely on low-power use of frequencies that could be affected by new white-space gear. Other opponents to white-space rules included broadcasters concerned about interference, and owners of expensive licensed frequencies.
The FCC's new rulemaking, due out next week, will apparently address these concerns, while also removing some cost obstacles for producing the gear.
Broadcom has released its Linux 802.11n drivers as open source code: Drivers for several current chips are part of the full-source code release. This may seem obscure for wide consumption: why should you care what the relatively small market of people using Linux on a laptop do?
It's actually a big market. Versions of Linux are used on hundreds of millions of digital appliances, in which an "embedded" form of the operating system is used (one designed to be limited and robust to carry out a specific set of tasks, like driving digital video recorder).
Atheros went open source on a set of its drivers in 2008; Broadcom, for competitive reasons, may have needed to join them in 2010.
Excellent report on the state of Wi-Fi data collection and how it's continuing to expand: Google doesn't collect Wi-Fi from Street View (at least in some countries) following its data acquisition debacle, but Android does. And iPhones. And trucks driven by Skyhook Wireless. And other sources.
Bob McMillan at IDG News Service runs through how it works, the current efforts, and where privacy concerns lie. In general, publicly broadcast information is hard to contain, but McMillan examines the connection between collecting millions of SSID and MAC association by location and making this available in easily retrievable form, as Google does.
The National Mall in DC gains free Wi-Fi: The AP says it was a joint effort between the US Department of Agriculture, the Smithsonian Institution, and the DC city government. Cisco donated hardware; Level 3 backhaul.
The in-flight broadband firm is yet to launch commercial U.S. service with a carrier, but just took in $37m more in funds: Row 44 launched itself several years ago as a Ku-band in-flight Internet provider for airlines that would make use of the many gains in antenna and receiver technology since Boeing's failed Connexion effort was designed, launched, and cancelled.
In the intervening years, the company signed up Southwest Airlines, but has yet to launch service commercially; it snagged Alaska Airlines, only to lose it to Aircell's Gogo Inflight Internet. Alaska made perfect sense for Row 44, given that Aircell cannot provide over-water coverage for Alaska and Mexico routes offered by the airline.
However, Aircell agreed to add many sites in the state of Alaska, and will at some point deploy in Canada via a partner as well.
Row 44 had a dilemma, from what I heard. It lacked the cash on hand to build out Southwest's service, and Southwest wasn't inclined to be the backer. With the additional money, I expect we start seeing Southwest Wi-Fi in the near future.
Aircell recently crossed 1,000 planes outfitted and in the air, and 2,000 is likely by the end of 2011. Row 44 has catching up to do, but it still has unique advantages for over-water and international flights.
OnAir was once a significant competitor in this market, but frittered away its huge lead time, and has only a handful of operator networks focused on mobile service (texting and email) rather than Internet access.
(Row 44 needs to spend a couple of dollars on its Web site. The press release and news sections are out of date.)