Slashdot Mirror
2.4 Megabit Cellular Modem
lew writes: "Ars has a review of a cellular modem that provides 2.4 megabits / second downsteam and 153 kilobits / second upsteam... and it works! Check it out" How much for unmetered service on such a system? :)
← Back to Stories (view on slashdot.org)
I can use my Voicestream unlimited weekend minutes to trade Jamie Curtis movies and pictures, now...
Get off my launchpad!
What that article doesn't mention, and what people usually don't know when discussing 3G mobile is that the data rates quoted are PER CELL not PER USER (unless only one user per cell is active at a given moment).
This is the big lie of 3G mobile. In cities, it will never support the data rates they keep talking about because of the duty cycle: the number of users per cell at any one moment.
How long before there are enough Quake and Home Porn/Warez servers filling the airwaves with frags and grainy shots of Britney Spears to make it as slow as a 56k connection on a bad phone line.
Eww, just think, I could have a pirated copy of Windows XP wisping its way through my body in the form of radio waves. That alone might add credence to the celluar gives cancer argument.
--Won't that be grand? Computers and the programs will start thinking and the people will stop. - Dr. Walter Gibbs
Picture this: you're sitting on the beach sipping something cold and sweet while browsing your favorite website, listening to some streaming audio, and communicating with a friend or co-worker. You have untethered bandwidth at your fingertips. Pipe dream? No, it's 3G.
I suppose this is an unpopular opinion, but isn't the purpose of 'getting away' actually to avoiding talking to a co-worker? I mean I would love the bandwidth they talk about at home but it's just not here yet.
The last thing I want on the beach is some dweeb cellphone going off 'cos his download of the latest Britney video is done. Just enjoy your vacations and leave the office crap at home.
--
Some weasel took the cork out of my lunch.
Monet Mobile Networks provides wireless broadband in rural areas using cdma2000 1xRTT (144kbps), and is upgrading to 1xEV-DO which provides 2.4mbps downstream and 144kbps up. The already have 1xEV-DO trial network Manhattan, Kansas. Their service fee is a flat 49.95 a month, unlimited usages. They also have 1xRTT service up in Fargo, N.D., and Sioux Falls, S.D.
Here is more info on the 1xEV-DO network.
I miss Ricochet. I ended up moving into an area where they offered service -- 6 months too late. (Dammit.) They were the only ones offering flat-rate service, although only at 128-256 Kbit. Yes, I know they're trying to re-light the network, but that's not happening up here -- at the last I'd heard.
1. THIS WAS A SERVICE TEST. They set up a few cell towers just for this engineering test.
2. Fat chance any cell provider will give you an all-you-can-eat plan! That's for businesses, you don't need that! You're just a consumer so take our advertising and consume!
Feh.
I've become so cynical regarding cellphone companies and their greed that I can easily see them crippling this service to the point where it's no fun for any of us. I expect:
- Throttled service levels (want more speed? PAY!)
- Outrageous fees per kilobyte (want a discount to buy blocks of bytes? forget it...)
- and "service" plans that sell you a dozen features you don't want, just to get the features you do.
We've become so used to "paying for minutes" that the cellphone companies aren't going to let that go without (1) a lot of money, or (2) a fight. I know people that pay "only" $40/month for cell service, yet barely use a quarter of their 'allotment' -- the rest of their money is wasted! It amazes me that people continue to accept this... I guess it shouldn't."...America's great minds of today, teaching America's great minds of tomorrow. Poor bastards." -- A Beautiful Min
That's the optimal, best-case, never-gonna-see-it-in-real-life (unless you're testing the system before it's released to the public) speed. In real life use you'll be sharing with everyone else on the cell, just like a neighborhood of cable modems.
From the article: Which brings us to the next point: that 2.4 Mbps is shared among all users on a cell sector, just like cable bandwidth is shared by everyone in a neighborhood. What's a sector, then? Cell sites are generally divided into three sectors that each cover different parts of the surrounding area, so each site can have up to 7.2 Mbps of bandwidth to play with.
And FWIW, latency: Round trip times were in the 110-120 ms range on average, with the minimum I recorded coming in a bit under 80 ms.
Have fun: Join D.N.A. (National Dyslexics Association)
Now perhaps I've been reading too much Neal Stephenson lately, but...
... no wait, those are just real girls getting a suntan.)
Do I foresee roving bands of samurai-warrior-programmers with laptops and wearable interfaces to the 'metaverse'?
Yes I do!
Well, maybe not Samurai Warriors, but pale geeks, surely! (And soon, perhaps not so pale?)
Could the stereotypical geek image change from pasty-faced teens languishing in a darkened computer room to well-traveled, olive-skinned men on beaches with laptops? (Oooh, look, live porn!
How much will civilization change when high-tech commuters can work from anywhere -- literally?
"And like that
You're entirely right. Cell phone pricing is silly, and I'm sure the vaunted 3G wireless will be underpowered and overpriced.
But changing that starts at the bottom of the communications industry, not the top. Why do cell phones have minute-based plans? Because land-line long distance does. They cost more because the consumer perceives greater value in the cell phone service (which is accurate), and therefore not only is willing, but demands to pay more. It's no secret that most people equate "more expensive" with "better."
Why does long distance charge per minute? Because local calls are flat-fee. Again, greater perceived value requires higher cost.
The same will be true of 3G connectivity. The only way to change that is to start at the bottom--why aren't local calls included gratis with the cost to have a phone line to a building?Why aren't long-distance calls flat-rate?
If that changed, everything above it would shift downwards. Either that, or someone has to hammer home to the public at large that cost and value don't necessarily have anything to do with each other.
Of course, if Windows hasn't done that already, I don't know that there's much hope...
Reality has a conservative bias: it conserves mass, energy, momentum...
Why do all the new broadband technologies limit the upload to a very slow speed? 2.4Mbps is nice and all, but for it to be useful beyond surfing the web 153Kbps doesn't leave for much of anything else.
Collisions. Same reason your upstream is often capped on a cable modem. On shared media you will get a lot of collisions from the individuals on the network as they choose to transmit at random times.
From the downstream perspective this is simple to control; you have one broadcast point, you simply queue things to be sent, and there are no collisions. On the upstream side, you need to know when someone else will be transmitting, and this is harder.
I imagine one way of doing this is to assign time slices to groups of people; you do not transmit unless it is your turn, and you compete with far fewer people (the others in your group). If you have 2.4Mbps available and you, say, divide this by 16 groups, you get a ~153Kbps window to transmit in (plus 9.6Kbps left over on the spectrum possibly for out of band housekeeping duties).
This is what is probably happening here.
Another options (and a long shot), but perhaps they are just plain mean (or not confident in their ability to control who uses their service) and want to discourage people from using the system to host anything. "Hey, our security is lousy, we know people will start stealing our wireless service to host copyrighted material/launch dos attacks from, maybe if we lock the bandwidth down at the tower this will not be attractive and the phreaks will go elsewhere".
this is the meat of the article. the pictures aren't really worth too much looking at. get a dvd case out and two black pens. there you go. here's the article:
The System
The particular 3G technology under examination in this review is called 1xEV-DO, which is a CDMA (Code Division Multiple Access) technology developed by Qualcomm. Picking apart the acronym is instructive. If you ask an engineer, the "1x" stands for "single carrier," which means it operates in a single 1.25 MHz frequency band just like existing CDMA cellular systems. If you ask a marketing rep, "1x" means the "first phase" of the third-generation wireless systems, implying more good things to come. The "EV" is for "Evolution," meaning the technology is an outgrowth of the base 1x standard, functioning as an interim solution for high-speed data while waiting for the "3x" multi-carrier systems being standardized by the ITU. "DO" stands for Data Only (the marketing guy would say "Data Optimized"), meaning that the entire 1.25 MHz channel is dedicated to data traffic and not shared with voice calls. So the present system implements the data-only variety of the evolution of the first phase of the third generation of wireless cellular technology. Got it?
If acronym soup isn't your bag, simply "fast" will do. 1xEV transmits in the same frequency bands as existing cellular systems and uses similar radio-frequency transmission equipment (the cell sites you see popping up everywhere), but employs packet-switched connections and a new radio link protocol optimized for high data throughput. The maximum speed of 1xEV -- no drooling now -- is 2.4 Megabits per second on the download link and 153.6 kilobits per second on the upload link. As you're probably thinking, that kind of bandwidth is on par with broadband wired connections like cable or DSL -- and the system delivers.
I was given the opportunity to test out an engineering prototype of a 1xEV-DO wireless cellular modem called the HDR Hornet, developed by Qualcomm as a reference design for their 3G chipsets. HDR is short for High Data Rate, Qualcomm's internal name for 1xEV. Qualcomm just makes the chips and does not sell retail devices, so you will not see this modem on the market. What you will see is a plethora of devices incorporating Qualcomm chips, from cell phones to PDAs to PC Cards to notebooks and devices that have yet to be conceived. Of course, any cellular technology without an appropriate infrastructure is about as useful as a frozen brick; Qualcomm also develops chips and software for cellular base stations, and the HDR modem under review was provided as part of a small over-the-air field trial conducted by Qualcomm in conjunction with the University of California, San Diego. There were three 1xEV cell sites set up on top of Qualcomm and UCSD buildings in the La Jolla, California area for the purpose of stress-testing the system in real-world conditions. Free bandwidth, in range of the beach? One stress test coming up!
The Setup
The unit I was supplied with came in a plain white box and a static-proof bag, along with an AC adapter, a dongle to connect the modem to an Ethernet jack, a two-page quick-start guide, and four Velcro stickies to attach it to a laptop. The Hornet itself is something between the size of a DVD movie box and a VHS cassette, measuring 7 1/8" x 4 1/8" x 3/4" HWD (18 cm x 10.5 cm x 1.9 cm) and weighing about 3/4 lbs. (0.35 kg). As you can see, the unit has two 5 3/4" (14.6 cm) antennae that independently swivel up about 200 from alongside the unit, enabling diversity reception for a stronger signal. Keep in mind that this is an engineering prototype; you will probably not see retail devices with this form factor. PC Cards and PDA modules with the same chips inside will likely be the most popular paths to 3G in the near future.
The first thing that struck me about the Hornet is that it looks pretty darn smooth for an engineering reference design, no frills, but all the essentials: AC adapter plug, on/off switch, USB port on the bottom, Ethernet dongle on the right, and four status LEDs on top that wrap around to the back so as to be visible while the unit is stuck to your laptop lid.
Installation and set-up can't be any easier. Taking a cue from the quick-start guide, the process goes something like this:
1. Plug it in.
2. Turn it on.
3. You're good to go.
The unit I was supplied with interfaced via TCP/IP over standard 10 Mbit Ethernet. The Hornet has a built-in DHCP server that automatically serves up the correct TCP/IP settings to your laptop and acts as your default gateway to the network. The connection is "always on" and there is no special dial-up or logon procedure. Having connected this thing to a dozen different computers, I can say that setup was simply a non-issue and took at most two minutes.
USB connectivity was not implemented on the test unit I received, but I can't imagine it being any easier to use than the Ethernet connection. USB will probably be the interface found in most external devices for laptops; unfortunately, this means you are at the mercy of the manufacturer for driver support and you will probably have to install a CD full of video-mail-grandma-with-one-click software to make it work. On the other hand, TCP/IP over Ethernet is standard, well-understood, supported out of the box by every operating system, and already used for Internet connectivity by most laptops. An Ethernet-enabled wireless modem would be a drop-in replacement for a huge installed base of users, but USB + Plug-and-Pray is perceived as being easier for consumers. Go figure. I tested the Hornet through its Ethernet interface with desktops and laptops using a variety of Ethernet cards under Windows 95, 98, and Me, Windows NT4, 2000, and XP, MacOS 8, 9, and X, and Mandrake Linux 7.1 (kernel 2.2.17). All worked flawlessly. A big nod goes to Qualcomm for sticking with open systems and standards. We can only hope retail products will do the same.
Once the unit is connected up and turned on, it takes about five seconds to initialize and then begins searching for a connection. If you're in a covered area, the service light goes green and the receive and transmit lights flash as the fire-breathing modem awakes and stretches its muscles. After living with this unit for a while, the sight of those lights when service comes up is like the geek's version of a well-tuned big-bore Harley's guttural rumble.
moox. for a new generation.
Hang on...not everything should be flat-rate.
When my girlfriend and I lived in separate states, our long-distance bill was huge...but we expected that. We were able to minimize it by using calling cards and talking in the evening.
Now my girlfriend and I live together...and our long-distance bill is small. If there was a flat rate for long-distance, it would certainly be higher than I'm paying now. All that would do is anger the 80% of people who use a less than average amount of long distance. (Yes, my math is right - the top 20% of long-distance callers talk five times as long.)
I would actually be willing to pay for cable/DSL by the megabyte. Why? Because that would encourage adoption...my grandma would be able to get DSL for $3 a month because she just checks email. I'd pay $60 a month, but I'd be getting my money's worth. And when I go out of town for two weeks, my bill would reflect it.
Having the option of a flat-rate plan is fine, but I think that it's not best for most people.
As much as rolling this out will cost, it's still going to be less than rolling out high speed land lines. In places where local conditions (terrain, politics, the whims of Time Warner) make DSL or cable unavailable, this may be the way brodband finally comes to the consumer market in big numbers.
The most compelling reason to suspect this may happen is that you can do an incremental buildout. Put up a few cell towers in an area and sell service. As enough people sign up to demand more bandwidth, you can add towers. You can't do that with land lines.
Nope, no sig
Many people have brought up ther valid point that if this (seemingly large) 2.4Mbit bandwidth is spread amongstwhoever is using the cell, then some people will hog everything, and others will get almost none, thereby creating a really terrible situation for the great majority of users.
:)
The point has also been brought up that paying by the kilobyte sucks for those who want high bandwidth...
My point is that the two effects would tend to cancel each other out, or, more specifically, that The people hogging the bandwidth would have to pay more, thereby eliminating the use of a cell phone for downloading warez or such.
Okay, so its not so nice... but it works. People will end up using the system for IMing and light web page browsing, which is what it is designed for. No-one intended cellphones to be used as hotline servers.
Now, it would be really nice if 3G meant more bandwidth than you could shake a nokia at, but its just not feasible. And who really wants to host a quake 3 server on a laptop. Most laptops can't even PLAY quake 3 with decent FPS (note I said most). And the payment scheme, though I am sure it will exact several orders of magnitude more dollars than are needed, making you pay the jerks through the nose for some crappy junk, works. Don't be surprised. We live in a real world
my grandma would be able to get DSL for $3 a month because she just checks email.
And she'd be happy until the first month she gets a screenfull of animated adds, a mailbox full of spam, and a $750 bill for the privilege.
Current internet technology evolved in an unmetered, bandwidth-limit-only enviornment. The content of the web and email - or the intelligence of the browsers and delivery agents - will require major revision before metered broadband internet service becomes practical.
Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way
Also, that downstream transmitter can push more watts, hence has better signal-to-noise, hence can use more complex modulation techniques and get more bits per Hz of bandwidth. Given 1 MHz of bandwidth for each direction, a base station using 256-QAM modulation has a raw bit rate of 8 Mbps (then subtract out a bunch for forward error correction, framing, etc). The low-powered upstream transmitters may only be able to code at two bits per Hz, for a 2 Mbps total.
Why do all the new broadband technologies limit the upload to a very slow speed? 2.4Mbps is nice and all, but for it to be useful beyond surfing the web 153Kbps doesn't leave for much of anything else.
Collisions. Same reason your upstream is often capped on a cable modem. On shared media you will get a lot of collisions from the individuals on the network as they choose to transmit at random times.
Collisions can be managed by assigning time slots for the inbound direction. There'd be some reduction due to variation in turnaround time among customers sharing the bandwidth, but nothing like a 20:1 degradation. (And it can also be managed by smarter schedulers.) You have to do some of this anyhow.
But the upstream doesn't (or doesn't HAVE to) apply to non-shared services like DSL. There the bandwidth is divided between the upstream and downstream link - currently with a fixed ratio though in principle the modems COULD have dynamically adjusted it.
No, I believe the issue is that the network designers just built networks on the assumption that the customers were mainly browsing the web or pulling down content, rather than serving others. For such a content consumer you want the downstream to be as fat as you can afford, and the upstream to be adequate for TCP ACKs URL references, and keystrokes. Then they massively oversubscribed the (symmetrical) network link feeding the local node (DSLAM, cell, what-have-you) and let the users stat-mux themselves. When they have little competition (which, the carrier hopes, is most of the time) they can fill their fat personal downlink pipe to its capacity. They don't lose packets on the uplink (which would break them badly) beacuse they're throttled back so far that the network link doesn't saturate.
Users running servers break that model. They cost the ISP more to support because he can't oversubscribe the network link to such an extreme - or much at all - without degrading their service. Even with the throttle, a few users hosting servers on a DSLAM can start causing other users to lose upbound packets and see download degradation.
Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way
To really be deployable in an un-metered fashion with a reasonable business model, you need something much more spectrally efficient, like ArrayComm's i-BURST for high-speed data service. A recent demo in South Korea shows it working at 1 Mbit/s. Two of South Korea's big telcos, Hanaro and KT, are planning to roll it out next year some time. Remember that Korea is where CDMA got its start.
ArrayComm licensed some spectrum in Australia, where they plan to roll out a wireless broadband service in the major cities in just 5 MHz of TDD spectrum. It looks like recent FCC rule changes have made some national TDD spectrum licenses available in the U.S. as well
It uses IntelliCell spatial processing and spatial channels to get multiple users on the same spectrum, at the same time. I've been lucky enough to see the i-BURST system in action, and it looks pretty cool, is real, and actually works. There are other smart antenna companies as well that are working on broadband data products, but I don't think any of them are as far along as ArrayComm.
Yesterday it worked; today it is not working; Windows is like that...
That assumes that the flat-rate amount multiplied by the number of customers would have to equal what the long distance companies currently earn.
The parent post was correct. Things are expensive because LD is still (comparatively) expensive.
LD used to be expensive because the COSTS were high to provide it. Laying the lines, relatively low number of users, etc. Now, telephones are virtually everywhere. Local calls are unmetered, but long distance is still relatively expensive mostly because people got used to paying for it. They value the service monetarily because they are used to paying for it.
LD no longer is as expensive as it used to be to provide. In fact, technically, it can be provided almost free. Most of the actual telecomm costs are in "the last mile" (read: the local telephone service that you already pay a monthly bill for).
Believe me, in 10, maybe 15 or 20 years max, there will be no "long distance charge" per-minute nor per-call and the companies providing them will either be much smaller and paid some monthly amount by local providers paying for international connectivity (like ISP access to the backbone).
Why? Because the price we pay for long distance is a perceived cost based on habit, not based on the actual real value or cost of the service. The price is, thus, unnaturally high. It may take time, but the free market will ensure that an unnaturally high price comes down. And it will.
While VoIP seems to have lost it's dazzle (with the dot com boom), I think VoIP is really what's going to eventually lead to free long distance. VoIP is in its infancy. When there is more infrastructure VoIP will be able to charge less than long distance companies. To compete, the long distance companies themselves will have to resort to VoIP. And, at some point, the local telephone company will end up simply being the local POP for the VoIP network... and the long distance companies will no longer exist.
That's my guess, anyway.