Intel Pushes 802.16a Wireless MAN Standard

An anonymous reader writes "The 802.16a standard, approved in January of this year, is a wireless metropolitan area network technology that will connect 802.11 hot spots to the Internet and provide a wireless extension to cable and DSL for last mile broadband access. It provides up to 50-kilometers of range and allows users to get broadband connectivity without needing a direct line of sight with the base station. The wireless broadband technology also provides shared data rates up to 70-Mbit/s."

Comments

Martin Cooper on WiFi

[zeoslap]

Martin Cooper, the inventor of the cell phone had this in a recent interview http://news.com.com/2008-1082-995667.html

"Wi-Fi is wonderful. It is a superb local area network--what it was designed to do--and it does that very well. When you try to make Wi-Fi cover a wide area, it's absolutely the worst way to do it. Think about it. In order to cover a city, you need a million sites; we actually did an analysis of that. And every one of them has got to have backhaul. So it turns out it's neither economical nor practical."

I realise this is WiMax but I wonder what they are doing to move beyond the limitations these guys found.

Re:Martin Cooper on WiFi

[robslimo]

Go to wimaxforum for technical info.

Re:Martin Cooper on WiFi

[robslimo]

To clarify,

And every one of them has got to have backhaul

WiMAX isn't expected to be what you use to hit the 'hotspots' with your notebook. It is expected to feed the hotspots... it *is* the backhaul. Naturally it must have it's own, land-based backhaul, but that's no sweat for guys who'll be rolling this out.

The idea of 'free' zones will largely pass when the people with the money to make wireless internet work finally get the tech and the business model worked out. Yes, I said *business*. Sure, there will be people, organizations and towns who'll foot the bill for small hotspots, but to make it work, to make it ubiquitous such that you *expect* it to work, will be require a commercial model. 802.16a is the first major technological step toward this model's feasibility.

Re:Martin Cooper on WiFi

[zeoslap]

Hmm Ok, so am I right in saying that you have WiMax providing a backbone of sorts and then the WiFi hotspots pulling their bandwidth from this as opposed to landbased solutions.

Is there really a problem right now with regards getting bandwidth to hotspots ? Is this solution targerted more towards rural communities than cities or am I still missing the point ?

Re:Martin Cooper on WiFi

[robslimo]

Unfortunately, I see rural areas being below the radar of ISP's for a long time to come. Unless it's government mandated (and 'ratepayer' funded in the utility model) there's just not enough money in it.

No, I think the target is precisely metro areas. Just as cellular telephony started in cities, then expanded to the countryside, the rollout cost of blanket wireless networking must be paid by a large initial audience before it will succeed. The backers seem to think that 802.16a is the solution... it's impractical in terms of infrastructure to feed relatively low coverage, low bandwidth areas by copper or fiber (and associated equipment), so enter WiMAX.

Before you try to draw an analogy to cellular telephony and related data services, recall that the bandwidth requirements of modern cell phones is modest compared to what the public (and me) expect from wireless Internet. Hardwiring every 802.11x hotspot back to the net is too expensive. Hardwiring a *few* 802.16x's (yeah, I bet it don't stop there) the net ain't so bad

Re:Martin Cooper on WiFi

[petecarlson]

Yes, there is a problem getting bandwidth to hotspots.
I run a wireless "hotspot" in Baltimore which serves a two block area. If I were to bump to a T1 I would need about 50 monthly subscribers to cover costs and a small profit. In order to do this I need to expand my range which means I need to set up additional acess points. The problem is that where the acess points need to be is not line of site to my base station so I would have to have a wired conection to each point or have a series of repeaters. This isn't practicle or cost effective.
If I could set up an 802.16 base pushing bandwidth to five or six 802.11b acess points then I could run them all off of one T1 line and put them in locations where they need to be.

Re:Martin Cooper on WiFi

[cnkeller]

I realise this is WiMax but I wonder what they are doing to move beyond the limitations these guys found.

We've been trialing on east coast for most of the last year.

Re:Martin Cooper on WiFi

[Wesley Felter]

Hotspots are a red herring; the article only mentions them because that's the latest hype wave.

802.16 is a wireless competitor to DSL and cable modems.

Re:Martin Cooper on WiFi

[MobileDude]

>>In order to cover a city, you need a million
>>sites; we actually did an analysis of that. And
>>every one of them has got to have backhaul.

Not necessarily true. Higher end Wifi vendors have dual radio APs. One radio handles data traffic, the other "hops" wirelessly to other APs eventually back to a wire. All you need is AC power. Imagine a network of these wireless APs with one (or two) root APs that have both 802.11b for data hops and 802.16a for connecting back to the wire.

Re:Martin Cooper on WiFi

[the_2nd_coming]

here is the comercial modle. you walk into a zone owned by some one and your Palm/PocketPC beeps and displays a message

"you have entered a hot zone owned by such and such, would you like to make a connection?"

yes no

you can pay 20 dollors for the day or 50 cents per min. or if you like, you can get a year subscription for $400.

what would you like to do?

day
min
year .....

that is going to be tyhe business modle.

sure eventualy you will have consolidation so you end up with companies with very large ranges so you will become more apt to get a yearly subscription and it will be cheaper than 400 bucks at that point. but at the start that is what you will get.

Re:Martin Cooper on WiFi

[Anonymous Coward]

Mod that one up. It's funny.
Welcome to Fantasy Island. . .

Re:Martin Cooper on WiFi

[tchuladdiass]

Actually, I've got an alternative commercial model:
An area (or national) ISP offers (and bills) it's customers hotspot access. Anyone who owns a hotspot can sign a deal with the ISP (or multiple ISPs). When the ISP signs on and authenticates via a participating hotspot, the hotspot providor then bills the ISP for the access.
The authentication and accounting aspects can be handled with existing protocols, but it may be easier to impliment with IPv6 (roaming IP's without tunnels, encryption at the link level, etc...)

Re:Martin Cooper on WiFi

[jdray]

Now THAT makes sense. I was trying to figure out how having some wireless backbone technology to connect WiFi hotspots together was going to do anything. I don't know how it is in other metro areas, but here in Portland, most of the city has either DSL or cable Internet available. Most places have a choice. Both are expensive, but worth it.

But if (as some previous poster said) it takes a million WiFi hotspots to cover a city, then I don't see how changing the way they connect to the Internet changes that. Each hotspot still covers the same footprint. :-\

But if 802.16 is a competitor to DSL/Cable, then that makes a lot more sense. I wonder what it would take to be a provider...

Re:Martin Cooper on WiFi

[Anonymous Coward]

Both the websites that were referred don't have any information on spectrum required or what frequency band this magic technology is going to operate in.

Is it 10 Mhz or 1Ghz of spectrum that is required? With a balloon in the sky, even CDMA can provide 200 Mbps using 125Mhz of bandwidth. Sprint/Verizon are providing/planning to provide 2Mbps in 1.25Mhz of bandwidth.

Both Intel (and its subsidiary's predecessors) and Nokia having miserably repeatedly failed over 10 years to master CDMA keep issuing attention grabbing press releases about supposedly 3G-trouncing vaporware.

There are better & newer technologies out there, but Intel, Nokia and Proxim have no involvement in them.

Security!!!

[Slack0ff]

Hey I never got much of a chance to read up on this but with the advertised range what is the security like? Dont tell me its like that pushover excuse for protection known as WEP on 802.11b. My big concern is that with all this range it will be hard to pinpoint where the guy with a card and a laptop is tryign to get your stuff. Or steal connection from an ISP? Anyone got any thoughts or know the security specifics?

Re:Security!!!

[Anonymous Coward]

Security with WiFi is no less secure than hard wired networks. The fact that anyone even suggests this at all is extremely frustrating. Its not unlike the claims made by mainstream reporters claiming that web cookies are a way to spy on you.

Check out the following oscast editorial for more info on the subject: No need to feel insecure about Zeroconf / Rendezvous security - February 27, 2003

Re:Security!!!

[ackthpt]

Hey I never got much of a chance to read up on this but with the advertised range what is the security like? Dont tell me its like that pushover excuse for protection known as WEP on 802.11b. My big concern is that with all this range it will be hard to pinpoint where the guy with a card and a laptop is tryign [sic] to get your stuff. Or steal connection from an ISP? Anyone got any thoughts or know the security specifics?

Right on the heels of this article, I'm more worried about War Cooking... gangs of nerdish thugs driving around cities, looking for open access to my microwave.

07:10 AM Cook for 10 minutes
07:20 AM Done
07:22 AM Cook for 15 minutes
07:37 AM Done
07:48 AM Cook for 5 minutes
07:53 AM Done
08:04 AM Cook for 3 minutes
08:07 AM Done
08:14 AM Cook for 25 minutes
Smoke alarm goes off, firemen arrive, haul smoking carcass of microwave out into street.

Re:Security!!!

[malfunct]

I think where people get mixed up is in the fact that with a wired network you have to find the wire to tap the network where as with wireless the signal is just out there to grab and look at.

So as far as things go the wire protocols are probably less secure but far more obscure.

Re:Security!!!

[WolfWithoutAClause]

You can probably get on to the WWAN quite easily. Whether you would be able to get on to the internet from there is quite a different matter. The ISP may deploy a VPN over the WWAN that you have to log in to in order to allow access to the internet to people have paid the fees.

Re:Security!!!

[Wesley Felter]

802.16 has good security. IIRC all endpoints are authenticated and all traffic is (correctly) encrypted.

Re:Security!!!

[twiztidlojik]

I doubt it.

Home users would rather have a router with (gasp) PPPoE or the like than a VPN. I certainly don't want to have to set up a VPN client to get online. Where was the last sub-$150 router that supported VPN access by the router?

Re:Security!!!

[WolfWithoutAClause]

The home users wouldn't have any choice in the matter. If the ISP only supports VPN access then that's it.

I'm sure any WWAN router would handle VPN. I've not found that setting up a VPN is any more difficult than PPPOE.

Re:Security!!!

[twiztidlojik]

...but the ISP would have a choice.

Any WWAN router would be merely an 802.11(x) router with a WWAN card slapped in. I know of no routers or firewalls that directly support VPN, and I believe that this would be a poor choice on the part of the ISP. A widely used standard (PPPoE) should be used for fast adoption of the service. A special client/service rings much too close to AOL-like.

Re:Security!!!

[WolfWithoutAClause]

Any WWAN router would be merely an 802.11(x) router with a WWAN card slapped in.

Most routers I'm familiar with don't have separate cards. In any case we're talking about 802.16 here, remember?

I know of no routers or firewalls that directly support VPN,

I'm using two at the moment. Contivity firewalls support VPN among others. You'd need a new router for 802.16 anyway,and the newer routers tend to support it out of the box.

A widely used standard (PPPoE) should be used for fast adoption of the service.

Could be. Stunningly bad idea though. As I understand it, PPPoE has absolutely no security, connections can be hijacked one way or another, and all kinds of evilness can be perpetrated; most of which represent a loss of bandwidth to the users, and a reduction in confidence, leading to a loss of customers for the ISP.

A special client/service rings much too close to AOL-like.

Must be your provider huh?

Re:Security!!!

[PapaZit]

You'd be correct if there weren't multiple bugs in both the WEP protocol and in most common implementations. As it stands, most implementations (Cisco's is a notable exception) use the same keys for all traffic. That means that you can park your wireless notebook near an AP and deduce the keys in a few hours. If your wired office is next to mine, there's no way for me to sniff your traffic. If your wireless office is next to mine, I can crack your keys and see your traffic.

There are also fundamental key distribution problems in large organizations. I work at a university. We don't use WEP because we can't come up with a way to distribute the keys that doesn't also make it trivial for non-authorized users to use the network. We can use physical security to keep our wiring closets and network jacks safe. We can't put faraday cages around our buildings to keep the wireless network safe.

Re:Security!!!

[twiztidlojik]

A special client/service rings much too close to AOL-like.

Must be your provider huh?

Nice 'bait.

Let me clarify myself:
I meant sub-$150 routers.

Do you know of any routers for under $150 that could add VPN support for less than an additional $5 or so, or if these exist, then can you link instead of flaming me?

kthxbye

Re:Security!!!

[WolfWithoutAClause]

Do you know of any routers for under $150 that could add VPN support for less than an additional $5 or so, or if these exist,

I have a Netgear MR-314 which is a 802.11b wireless router with 5 ethernet ports; you should be able to get that for much less than that; and it has VPN passthrough, which is what you need. I've also got a D-LINK DSL-504 adsl router which also has VPN passthrough, both built-in, no extra cost. The former is much less than $150 right no, the latter goes for about that much.

then can you link instead of flaming me?

You cheeky bastard; you're the one that started flaming me!

Re:Security!!!

[twiztidlojik]

Cheeky, eh? Long time since I heard that one. ;)

First Post...

[Anonymous Coward]

This is the first post originating from at least 50km away from the nearest communications cable.

Please let it be 802.11 "G"

[oscast]

Will it be based on 802.11a 802.11b or 802.11g? Lets hope "G" is used so that all our "B" hardware can be maintained. Knowing Intel though, they'll probably back "A" just to spite everybody.

Re:Please let it be 802.11 "G"

[marcgul]

It won't be compatible with A/B/G, 16a is a backhaul standard, and you (probably) won't ever have to talk to a 16a radio with your laptop.

The idea is to use this to supply bandwidth to hotspots.

Some acronyms for ya

[cmburns69]

LAN = Local Area Network
WAN = Wide Area Network
MAN = Metropoliton Area Network
WOMAN = Wide Open Metropolitan Area Network, which is what most of those 802.11 networks will be...

An online Starcraft RPG? Only at

Re:Some acronyms for ya

[MarkGriz]

"WOMAN = Wide Open Metropolitan Area Network, which is what most of those 802.11 networks will be..."

Presumably these will be equipped with an 802.11g-spot?

Re:Some acronyms for ya

[swb]

Presumably these will be equipped with an 802.11g-spot?

Yes, except you won't know where to find it, and the equipment will always fake a link light, so even if you think you've found it, you can't be sure.

Range for the big boys

[Capt'n Hector]

50 km... so you could get on your home network from the next city over? What about interference or hacking? If you have 1 million people using this standard, each overlapping, wouldn't that create massive headaches? Or, would that 50 km range be only applied by "select" companies that pay cash to the city(s) and maker of the hardware? It has the potential to become a disaster, but if done right... imagine surfing from home connected to work with a T1 line.

Re:Range for the big boys

[Anonymous Coward]

"What about interference or hacking?"

Security with WiFi is no less secure than hard wired networks. The fact that anyone even suggests this at all is extremely frustrating. Its not unlike the claims made by mainstream reporters claiming that web cookies are a way to spy on you.

Check out the following oscast editorial for more info on the subject: No need to feel insecure about Zeroconf / Rendezvous security - February 27, 2003

Re:Range for the big boys

[Capt'n Hector]

"Security with WiFi is no less secure than hard wired networks."

I beg to differ. With WiFi, you can leech bandwidth off your neighbor (ie the whole city with 16a.) with wired, it's a wee bit harder.

And I'm sorry I frustrate you, I'm doing all I can!

Of course they're doing this

[Surlyboi]

How else would they get people to rationalize buying
all their new Centrino crap?

Re:Of course they're doing this

[LordMyren]

Sorry kids, this aint geared towards mobile usage.

802.11a/b/g already has enough problems with power consumption. Somehow i dont think this is going to improve the situation.

Re:Of course they're doing this

[Anonymous Coward]

"802.11a/b/g already has enough problems with power consumption" Huh? 802.11x has almost no influence on power consumption.

Re:Of course they're doing this

[Anonymous Coward]

Bullshit. Fucking Wi-Fi card cuts my laptop's battery life in less than half.

How about those WiFi startups?

[robslimo]

You reckon those (possibly overabundant and overfunded) WiFi startups are in on this?

It would be par for the course for a newer technology to lay waste to grand entrepeneurial visions... but since this standard was approved in January, hopefully some of those startups have it 802.16a in their sights.

Standards...

[sgtsanity]

They had better make a standard for the naming of all these standards, or else my head is going to start spinning.

Re:Standards...

[Hognoxious]

my head is going to start spinning.

With all them there wavesflying about, be grateful it doesn't start fizzing.
Protect yourself!

sounds fun.

[Anonymous Coward]

being able to connect multiple wireless pops together, to form a cohesive network, relying on very few wires, and no outside service providers.

only problem is, with the current government backlash against wireless networks, and their potential use as "terrorist devices", how long will it take until regulation is proposed?

No thanks

[Jaguar777]

I don't think I want to share 70-Mbit/s with everyone using the service within a 50 kilometer range.

Re:No thanks

[robslimo]

So don't share, sell. See if you can work out a bandwidth reselling agreement from an ISP in your area, get a fat pipe, go to town.

Obviously it's not just that easy, but if you can work up a business plan, get a willing ISP partner, and 802.11x partners around town, you could sign people up for wireless... Which is what 802.16a is for anyway.

Re:No thanks

[malfunct]

Yeah thats always been my big argument against wide area networks. You end up with a fixed amount of bandwidth over a given area. With wired networks if you need more bandwidth there is usually room for another wire, but rarely is there more room on the air for more data.

I think the key (and they might have this as part of the protocol given that I haven't read the article) is to have a large number of very small area connections that are wired together. That way you split the 70mbit/s connection with far fewer people. That said I think it would totally awesome if we could deploy hotspots to a neighbor hood that hit maybe 100 people and then have one fiber connecting to it on the backend. I don't know that it would be cheaper than putting a DSL concentrator or cable loop in that same neighborhood though.

Potential?

[bombkit]

Mabey this will make wireless networking less of a novelty and more useful. As far as Home / SOHO usage, i'v found 802.11a/b to be pretty useless. 72m/bit at best, i'v tryed top of the line name brand 802.11a / b access points, and had trouble finding one that would bounce a signal thru a sheet of paper without loss of speed / strength / reliability. I'm not a big wireless networking buff, but it seems to me that 802.16 could make wireless more practical for everyone.

Re:Potential?

[oscast]

I use wifi on a regular basis. It seems to me that you either bought a bad card, are trying to access a hot spot from too far away or are using the wrong technology. Rather than using "A" for your fast-wireless needs, you ought to use the "G" standard instead. not only is it known for being more reliable, but its also combatible with "B" access points... unlike the "A" standard.

What ...

[Anonymous Coward]

... are you talking about? 802.16 is a completely new standard designed specificly for last mile connectivity. 802.11 was never intended for many of the applications that it is now being aplied to. That is why there is the quick succession of bolt on standards.

None of the above

[SuperKendall]

It's 802.16a. All of the three technologies you listed are just for short range networks, not the kind of MAN network that they are addressing with 802.16a.

I think the way it would work is you'd get an 802.16a "modem", just like you get a cable or DSL box right now to connect your network to.

Personally, I find wireless access a choice of last resort - if I can get cable or DSL I'd take that every time over wireless.

Re:None of the above

[div_2n]

As opposed to amplified 802.11a/b/g capable of 12 miles range without breaking a sweat?

Re:None of the above

[Anonymous Coward]

ahem...

when I had a wireless link to my ISP (first and only after long negotiations) I had killer counterstrike ping times and the equivalent of around 6Mb/sec. They caught onto my bandwidth utilization when they got their upline bill. To bad they went outta business woulda been nice to keep them around.

Miracles from Where?

[LordMyren]

How does 802.16a hope to achieve these lofty goals?

What band does it use? Considering its long distance, the 802.11a 5Ghz range seems a bit out of the question, just too energy hungry. If its 2.4, i cant see how they expect it to compete with every other signal under the sun and still pull off such spectactamundo specs.

Typical transmition power?

Now wouldnt it be nice to have a frequency not in tune with water? So maybe vegitation isnt a big iron curtain between you and your data? Bring that critical LoS step another twenty feet down to earth? Course, thats probably not gonna happen.

Re:Miracles from Where?

[Wesley Felter]

802.16 can run in pretty much any band. Some ISPs will use licensed (e.g. 3GHz) spectrum; some will use the unlicensed 5GHz band.

I don't think range is as bad in the 5GHz unlicensed band as people say; there's already equipment out there that gets multi-mile range.

Get the Pringles and call the SETI people!

[worst_name_ever]

If a Pringles can is able to extend the range of 802.11 wireless LAN to several km, then a similar application of tubular snack food waveguide technology to this new standard ought to solve the question of "are we alone in the universe" once and for all!

By the time this arrives...

[path_man]

...the CDMA carriers (SprintPCS and Verizon) will have 2Mbps 1xEVDO (TRUE 3G networks) up and active. The biggest single limiting factor to creating a wireless infrastructure is that somewhere it has to tie into fibre optics. Wireless carriers, nacent though the technology is today, have this figured out. Some xx,000 wireless radio towers all terminate at a base station connected to real telco networks.

Creating new wireless networks for purposes of roaming inside a metropolitan area seems like a big waste of resources -- especially considering that wireless carriers have already figured this out.

Re:By the time this arrives...

[iCEBaLM]

And they will happily charge you $5 per minute for that access.

Something tells me 802.11b is a little cheaper.

-- iCEBaLM

Re:By the time this arrives...

[dieman]

And before then we will see EDGE from T-Mobile and AT&T. (I'm guessing AT&T will beat them all with 3g, they have cool features like buddy location allready, too.)

Re:By the time this arrives...

[vought]

...the CDMA carriers (SprintPCS and Verizon) will have 2Mbps 1xEVDO (TRUE 3G networks) up and active. The biggest single limiting factor to creating a wireless infrastructure is that somewhere it has to tie into fibre optics. Wireless carriers, nacent though the technology is today, have this figured out. Some xx,000 wireless radio towers all terminate at a base station connected to real telco networks.

You know, I've been hearing this exact verbiage for four years now, and I don't believe it any more. When I worked at Metricom, Ricochet was the product that was going to be 'killed' by 3G. Luckily for 3G, Metricom's brain-dead, overspendy management and ridiculous pricing model killed the company instead. Curiously, the arrangement Intel seems to be proposing here is strikingly similar to the dual-band microcellular architecture Ricochet used/uses. Microcellular architecture has some unique strengths, as evidenced by the fact that Ricochet was the ONLY way to get data to ground zero in the days immediately following the WTC attacks.

Now the previous poster is saying this uplink and backhaul arrangement will be obviated by 3G. You know what? Show me. Then I'll believe it. Until then, I don't think 3G will ever solve anything for anyone.

3G sounds like great technology. But it isn't shipping, and there are LOTS of caveats. have you ever seen a technology that worked out of the box? 3G is still "months" away, and it probably won't work as advertised when it does ship, if ever. Perhaps 3G should be renamed "Duke Nukem Forever Wireless".

I'm tired of hearing "wait until 3G". Hell, I'm tired of waiting.

Re:By the time this arrives...

[gad_zuki!]

>...the CDMA carriers

So? GPRS is here today and you pay through the nose and sometimes through other orifices I'm too polite to mention.

Cell carriers have a huge incentive to bill per megabyte (or kilobyte). It may be the only way they're going to turn a profit. Wireless designed for a MAN has a huge incentive not to pay per byte, but to give it away and bill as a "last mile" carrier a la DSL, Cable, etc.

Heh, I can't wait to see my 3G bill after getting those new Mandrake ISOs.

"Honey, did you download that new microsoft patch?"

"Oh yes, I always patch."

"That damn thing cost us $59.24 in transfer fees plus a $40.00 surchage for using that much bandwidth during peak business hours."

"You're the one that wanted the wireless not me!"

"So? I told you to only use it between 2 and 4:30 am CST!"

"That's it, I'm spending the night at my sister's."

Community Wireless

[composer777]

I wonder if this will be cheap enough to use with community wirless networks, or if we'll be stuck trying to extend the range of 802.11b? I would hate to see the wireless spectrum sold to the highest bidder the way domains were, but it seems that this may in fact happen unless laws are created to protect non-profit community access networks.

We have one such group here in Atlanta called atlantafreenet.org
The project looks fairly promising, and they already have a backbone up, but it requires a line of site. Does anyone have any prices on this equipment? I would hate to see the price of this technology made artificially high or have the bandwidth used up by the highest bidder. Hopefully we'll see communities creating their own free networks out of this.

Contradictions in Intel Strategy

[ldesegur]

On one side the marketing at Intel is pushing for 802.11a, and on the other side, the company offers technology with 802.11b only. What you can get from Intel as far as Wi-Fi in the new line of x86 laptops is an inferior 802.11b. Intel 802.11b chipset is significantly worse than other players like broadcom that even reviewers of ZDNet flag the chip has been mediocre.

And currently, if I want to get a laptop with 802.11a or both 802.11a/b (which makes more sense currently since a is not so popular), I can not buy anything with Intel Mini-PCI chipset in it (Centrino technology, not the banias but the wireless stuff, is 802.11b only).

The new laptops with 802.11a/b all come with the superior Broadcom chip that has been licensed to Philips, IBM, Dell, etc...

It appears that Intel marketing droids are at work on some cool idea, and the engineers are developing something else. Not too uncommon for a company of this size.

Re:Contradictions in Intel Strategy

[thesadjester]

This is 802.16a ...not 802.11b...totally different standards.

Don't worry. You'll probably never use 802.16a explicitly, but rather through an 802.11b system.

Re:Contradictions in Intel Strategy

[oscast]

"Don't worry. You'll probably never use 802.16a explicitly, but rather through an 802.11b system."

You got that wrong. To suggest that somebody will access an 802.11a network over an 802.11b connection is incorrect.

802.11a (high-speed) is incompatible with 802.11b (slow-er speed). Only 802.11g (high-speed) is compatible with 802.11b (slow-er speed).

Re:Contradictions in Intel Strategy

[thesadjester]

I was suggesting that they would access an 802.11b router, which would then itself be bridged somehow to the 802.16a interface..I'm not an idiot....the 802.16a would be the backend and wired somehow to the 802.11b.

Also, it's 802.16a, not 802.11a,.,.,I wish people would get it correct. Don't imply I'm wrong when you aren't even using the correct standards...802.11a is completely different from 802.16a (802.11a is yet another hotspot standard and 16a is the longrange one for non hotspot usage)

Re:Contradictions in Intel Strategy

[Anonymous Coward]

er slo down there kiddo.
The a and b in the decimel side of the figure don't translate from one standard to the other. .11 and .16 are two seperate standards with their own values for a,b,c ad infinalphabetium.

2Mbps SHARED, max few 100Kbps

[univgeek]

You wanna share 2Mbps,( and pay through your nose), or you wanna share 11Mbps (.11b), 54Mbps(.11a,g) ??

Especially if this is a fixed application, and doesn't need to be truly mobile?

802.16 is not wifi, not 802.11 at all

[Anonymous Coward]

Totally different standards. And for a typical long-haul connection both endpoints are staticly configured, so the security protocols like WEP and AES aren't needed at the layer2/1 level. Instead, each endpoint should just run a vpn. Still vulnerable to denial of service due to spoofing, but it's wireless - that's unavoidable. The key is to make it unlikely by limiting its usefulness, and with a vpn running, an attacker can only deny service, never gain free service or snoop the medium for anything useful.

Re:802.16 is not wifi, not 802.11 at all

[Anonymous Coward]

So you're telling me some closed source vpn solution is secure? Might as well use ppp over ssh.

Re:802.16 is not wifi, not 802.11 at all

[LordMyren]

VPN: absolutely fantastic idea. Honestly, i think all wireless communications should use VPN's. Pitty freeswan doesnt move, and I'm not ready to put production servers on 2.5, much less a componenet i'm fairly sure has a long long ways to go (what do you expect from a component as complex as the kernel itself).

I'd really like a DHCP server that points to a web proxy where you login to the VPN, and that helps to configure the client VPN. Unfortunately, VPN is such a nightmare right now in linux that I cannot provide the free secure hotspot I would like to.

Ultimately, some sort of dynamic mesh routing based off of VPN's is the ultimate goal. But its going to be a while.

Myren

wires

[stratjakt]

Plug em in

Wires are the future

When all you wireless guys cancer ridden corpses are long since buried, those of us with wires will be enjoying the fruits of the new millenium.

Ever try to assasinate someone with piano 'air'? No. You need wire.

Re:wires

[Anonymous Coward]

Ever try to assasinate someone with piano 'air'?


You have yet to hear me play...

Re:wires

[outsider007]

When all you wireless guys cancer ridden corpses are long since buried, those of us with wires will be enjoying the fruits of the new millenium.

in the days before the first robot wars, machines made MAN after their own image and saw that it was good...
- excerpt from the robot bible (circa 2600 AD)

Re:wires

[MidnightBrewer]

The average person is already exposed to more radiation than the average computer geek; it's called "sunlight." Big glowing ball of gas in the sky. Lights up everything. Amazing thing, really. ;) (You may know it as, "that damn thing that keeps causing glare on my monitor and interrupts my sleep at two in the afternoon...)

Seriously, I was wondering about the radiation. The obvious question is, is there more radiation now than before, or are we simply utilizing a frequency that already has naturally-occurring radiation, and therefore not a big deal?

It's also going to depend on strength-of-signal, obviously.

Re:wires

[Anonymous Coward]

Ever try to assasinate someone with piano 'air'? No. You need wire.

It's called chlorine ;-)

Seems a little fishy to me . . .

[div_2n]

Something doesn't add up to me. You can already go up to 72mi/115.8km with 802.11b in the 2.4 range. I know you don't get that high of data transfer that way but you can get really good transfers easily up to 24mi/38.6km and higher.

The only good I see coming from this will be more non-overlapping channels. But I noticed that some of the frequencies they are talking about are in the licensed bands. I really don't see how they are going to make that affordable unless the FCC opens up some frequencies.

It seems to me that cost effective deployment of such technology might be a good ways away unless I am missing something. If I am, please someone clear things up for me.

Re:Seems a little fishy to me . . .

[Anonymous Coward]

you need line of sight to do those kind of ranges with 802.11*

Re:Seems a little fishy to me . . .

[div_2n]

And that really doesn't change that much just because they throw another standard on top of the same signals UNLESS the FCC ups the amount of total EIRP allowed. But there is a reason that there is a 4watt max in the 2.4 range. If they let people have 1000watt 2.4 devices very interesting things would happen most notably cooked flesh.

Besides, what good would it do if they have 4-8 watt devices that plug into a laptop (which WOULD be necessary due to pure physics) how long would their laptop battery last?

Re:Seems a little fishy to me . . .

[Anonymous Coward]

> You can already go up to 72mi/115.8km with 802.11b in the 2.4 range.

Wrong! The guys that did that (and /. posted a link to the story) admitted they were breaking the law. The previous two long-range 802.11 stories admitted they were frauds, and the wireless links they claimed to have work, actually did not. Check the facts before posting wild claims like 72 miles. 802.11 doesn't work from one side of the average american house to the other. I've seen dozens of 802.11 networks with work, and every single one did not work well. Please check the facts before posting wild garbage like that. The longest range 802.11 network I've personally seen was about 200 yards. Even with very nice cables (LMR-400), a 250 mW amp, 19 dBi parabolic dish antennas, and a lot of tweaking on the antenna positions, it still had a huge amount of packet loss (nearly 1%). 802.11 is not something a business can depend on.

I think a lot are misunderstanding the uses...

[thesadjester]

When I thought about this for a second...

It seems to me that this is the backend for the 802.11b and/or 802.11g wireless hotspots. Try this scenario..

Say I own a lot of starbucks (if they allow franchising, but this is an example so it doesn't matter).

I have a big fat oc3 sitting in downtown LA ready to serve bandwith to a lot of starbucks in LA. It would be MUCH more economical to pay for just one oc3 rather then a bunch of t1's or even cable modems for EACH starbucks. Using this technology, you pay once for the hardware and can send the bandwith from your oc3 over to each and every starbucks you want to and siphon it through the 802.11x clients without having to have the cable/phone companies come in and install land lines in each location.

Also, say that it was decided that you needed more bandwith for your coffee shops. Welp, you just add more bandwith at the home site (assuming each starbucks doesn't NEED 70mbits or whatever the new standard allows), and you now have more bandwith for all your starbucks shops...i think it's a great idea and is extremely expandable.

Re:I think a lot are misunderstanding the uses...

[DrInequality]

Sounds ok, BUT suppose now McDs, KFC and every other fscking chain under then sun gets in on the act? How much bandwidth is being shared by how many people then?

This is going to be slow.

[Guspaz]

With a radius of 50KM, we're talking 24649 square kilometers. At 72mbit, that's a total of 3 kilobits per square kilometer. Such speed! And imagine if you had more than one customer per square kilometer, as one would expect in a metropolitan environment...

Re:This is going to be slow.

[Anonymous Coward]

you assume that you would be using an Omni antenna trying to cover a 360 degreee range. Bad assumption, since an omni would simply never give you that range both physically and theoretically. Any provider hooking this up for multipoint access will likely use an array of antennas with a limited degree of scope just like cell sites do today, (alpha, beta, gamma). So your bandwith/square KM ratio would be quite different IMHO.

Yes, but point to point.

[SuperKendall]

I don't think 802.11x is built to handle the number of connections that the larger scale 802.16a is built for.

I can make my car orbit the earth by taking it up in a shuttle and throwing it out the bay, but that doesn't make it a real spacecraft.

Re:Yes, but point to point.

[div_2n]

With 802.11b you can easily get 50 clients per channel. With 802.11g you SHOULD be able to get at least double if not quadruple that.

With three non-overlapping 802.11b/g and 8 non-overlapping 802.11a channels, you could easily get over 1000 clients on one tower albeit a little bit of money in it.

What are they offering with this new standard?

Re:Yes, but point to point.

[SuperKendall]

Read this. They know more than I do.

1000 clients sounds incredibly small to me when you're talking about feeding a large residential area networking, especially given the bandwidth of 802.11g spread out over all those users...

Here's a snippet from the article on what it offers:

In an ideal world, Marks said, 802.16a can serve as a backbone for 802.11 hot-spots. Still, some wireless LAN advocates promote 802.11's use as a MAN, even though its medium-access control protocol is fundamentally optimized for shorter-range topologies. At the same time, Marks said, others have talked of using 802.16a within the enterprise as an adjunct to 802.11a or 802.11g. If the 802.11e working group has trouble providing true quality-of-service prioritization for wireless LANs, then it might make sense to take 802.16a directly to an end user, Marks said. Otherwise, "it's more efficient and more cost-effective to look for the ways 802.11 and 802.16 complement each other."

Good thing it's not WiFi then

[bill_mcgonigle]

802.11[b,a,g] is WiFi, this is 802.16a. (Note 802.3 is Ethernet, but this isn't Ethernet either). Granted, there's a provision in the spec for linking 802.11 WAN's, but the much more interesting part of the spec is the MAN stuff, with 20km links. The IEEE usually gets these things right, so I wouldn't worry about Mr. Cooper's concerns.

I need this - the only low-latency broadband I can get at my house (in a lovely pastoral setting 7.2 miles from the CO my line is served from, but of course not the closest) runs $850 install plus $90/mo. SWMBO frowns upon such things, so let's hear it for standards!. Remember what Proxim radios cost before 802.11b?

I'm actually more concerned about press articles that were flying around today talking about how Intel was about to revolutionize wireless communications. Yay, they sponsored an industry group already promoting an IEEE spec, but it seems more of a case of "why actually do the work when you can just take credit for it?"

What people are saying about 802.16

[Dusty]

From grouper.ieee.org/groups/802/16/pub/buzz.html

What People Are Saying about 802.16 This dated list includes an incomplete but nonselective collection of external references. If you have items that you'd like added to the list, notify the Working Group Chair, who compiled it.

Re:This is going to be slow. -- silly proposition

[victim]

Now thats just silly. Correct arithmetic does not make correct conclusion. Oh wait, I just checked in preview, your arithmetic is wrong. PI*r*r... 3.141596*50*50 = 7853sqkm... ~9kbps/sqkm. Maybe you used PI*PI*r*r? Anyway, to continue...

Just like cellular phone cell size, you tailor the coverage area to match the number of subscribers. In an urban area you use small cells, as small as a block or 4, in rural areas crank it up and cover a whole county. (I'm from Missouri, ours fit. Nevadans and Austrailians not so.)

Of Course they are gonna Push it!

[mesach]

How else do you think I can get my Centrino(TM) Laptop to have a connection on:

the driving range
the 5 meter platform
the roof of my skyscraper

Re:Of Course they are gonna Push it!

[frieked]

Don't forget the edge of the diving board at the pool

Re:Of Course they are gonna Push it!

[mesach]

thats the 5 meter platform

War driving and gas prices.

[SourceHammer]

Just when gas prices had made it too expensive to go war driving. Whole cities can be owned by simply driving down the main roads.

802.12 power output, looking for technical info

[Anonymous Coward]

If anyone has found specific information about radio power output of this new 802.16 system, please share pointers to info. Thanks. (Power and range limitations really seem to limit 802.11, so I'm wondering if 802.16 is a big step forward.)

Link to REAL information

[Anonymous Coward]

It appears that almost the entire /. crowd today has not even a glimmer of a clue as to what 802.16 is all about sooooo .... 10101 ~ -< ))) "802.16" ((( >- ~ 10101.

TMDA, not peer-to-peer

[Anonymous Coward]

Oh, it looks like a lot of 802.16 work
is for point-to-multipoint systems.
I wonder if the intent is for subscription
fees and things like that instead of the
buy-it-and-set-it-up model of 802.11
equipment that has made 802.11 so popular.

Apple & 11g (Re:Contradictions in Intel Strate

[Anonymous Coward]

Or you can buy an Apple notebook and get 802.11g which is backwards compatible with 11b. :)

This is a protocol?

[anthony_dipierro]

Do we have any type of licensed or unlicenced spectrum to go along with that protocol? Otherwise it seems kind of useless.

I am wireless man!

[mao che minh]

"802.16a Wireless MAN", is that a new Marvel super hero?

wireless man

[b17bmbr]

uh, for us married folk, umm, ain't never gonna happen. they ain't never lettin' go of the riens.

Radio isn't in tune with water

[mmol_6453]

See the Wikipedia.

Anything is slow if you use it wrong

[Wesley Felter]

Of course, most deployments will use much smaller (1-5 mile radius) cells. Also keep in mind that the cells are sectored.

Killer App: Corporate Remote Access

[HaeMaker]

I bet you could hit most of your employees' homes with a 50km reach.

One tower. Bunch of 802.11/802.16 devices to send home, matched with 802.1x/p/q and IPphones...

Remote access in a box.

wireless MAN

[eupheric]

wireless MAN, wireless MAN does everything that a wireless can what's it like? that's not important wireless MAN (with apologies to They Might Be Giants)

Don't Forget the Maintenance

[Myriad]

Presumably these will be equipped with an 802.11g-spot?

Yes, except you won't know where to find it, and the equipment will always fake a link light, so even if you think you've found it, you can't be sure.

Don't forget that they also tend to be highly unstable, suffer from monthly outages, and require enough regular maintenance that you'll likely have less time to spend fragging with the guys.

Watch out for the frequently required diamond upgrade too!

Blockwars: a real-time multiplayer game similar to Tetris.

shared?

[Cynikal]

so a shared speed of 70mbits divided by a hypothetical 1 million users offers you a whopping 70bits a second..

WOW

Mesh UWB

[LordMyren]

Mesh ultra wide band is where its at. Small scale flexible networking. Path from point A to point B gets clogged? Just use the untouched path to point C, then back to B. Good ole almost sorta kinda grid networking.

Very high speed switching antenna arrays should really help, allowing for the dynamic directionalizing to focus beams on the fly, sans mechanical apparati. I dont really know much about UWB in relation to this, if its even possible with UWB, but if its possible, it'd be pretty crucial.

Myren

Long range is for licensed providers

[isdnip]

WiMAX is a technique, not bound to a specific frequency like 802.11b/g/a etc. It is mainly aimed at licensed providers. It could be used in the LMDS (29 GHz) range, where reliable coverage is 2-3 miles (because of rain fade), at 39 GHz (similar), or at other licensed frequencies. It could be used in the licensed 2.6 GHz MMDS band, whose rules are the topic of a currently-open FCC docket in the USA. (Among the options is opening up some more unlicensed spectrum there.) Or it could be used, with low power and thus lower range, in the 5.8 GHz unlicensed band.

The technology is better for a public MAN than any of the 802.11 family. But let's not expect miracles from shared, unlicensed frequencies.

Re:Long range is for licensed providers

[Netssansfrontieres]

Which opens up the wonderfully complex questions: what to expect from 802.16 and 802.11 ... it would be neat to think that .11 would be a homeLAN or short drop solution, but it's unlicensed, so there's no control. And 802.16 is far from being cost effective.

Is what you're saying that 802.16 is the future for broadband?

Kickass idea.

[Cinematique]

TV is moving off of the VHF band as the eventual HDTV revolution takes place. But what's stopping Acme Wireless from saying, "Hey, only channels 4,6, and 10 are being used in this area. Why don't we add some sort of auto-sensing feature to broadband wireless equiptment and start using parts of VHF today..." and then asking the FCC for help? Then, when the machinery starts hopping packets to other routers that are close to other television markets, it switches to another unused set of frequencies.

I'm just thinking of a solution like that little channel button on 2.4ghz wireless phones.

Technically infeasable?

If it worked, it'd be a hella way to jumpstart nationwide wireless Internet via the VHF band now and not a decade or two from now.

Thoughts?

Re:Kickass idea.

[Wesley Felter]

It's possible. People in the "open spectrum" movement have proposed similar ideas.

Re:Kickass idea.

[Cinematique]

Where are HDTV signals going to be? Just curious...

802.16 is not the same as 802.11

[JeffDwskn]

I attended a talk today by Roger B. Marks, a member of the IEEE 802.16 standards committee where he described the standard in detail. Many people say just add a pringles can to 802.11 to extend the range, but there are many other issues beyond range. 802.11 and 802.16 are designed for different purposes.

Among other things, Mr. Marks described that 802.11's MAC uses CSMA/CA (carrier sense multiple access with collision avoidance). The carrier sense means that it listens (or tries to listen) for other devices broadcasting and only sends when it detects silence since the receiver can only handle one transmission at a time. This is fine for wireless LAN's where for the most part, all of the devices can "hear" each others transmissions and figure out when its ok to send. In an 802.16 MAN (metropolitan area network), the users' devices can't receive each others transmissions so the base station assigns each device a time slot in which to send & receive its data. (For more information on IEEE802.16, see their website: http://WirelessMAN.org.)

So like, it's just 20year to wait now?

[pepper_pusher]

Yeah well, I guess now we're just 20 years away from actually using that technology, until it gets approved by the gov. 2003 and people are still offline on the street...

MAN already exists - in Alaska!

[mazor]

Metropolitan Area Networks are already in use in the field. WAY out in the field - like Nome, Alaska. Check it out: www.nook.net

-mazor

wireless

[signore pablo]

you know what would be very cool? just an idea, even if security (and a million other things) needs to be worked out... But make internet connection purely wireless for the end user in all parts of the world. Blanket the world in wireless access points, and make it accessible to anyone just by asking their gps coordinates or something like that. A vision for the future internet! would be very cool to have a fully connected world where wherever you were in the world you could just switch on and be able to communicate with anyone else. mm.. grandiose ideas that are impractical... but cool. But still, I consider humans communication technology to be in a rather primitive state... I mean what, we got our first real start in 1876? its only been 120 years, thats not much time to progress.. The scene is really exciting if you think about the prospects of what we could one day possess. Now for the specifics? :) dont know, but i can dream

802.11

[IAR80]

Even old 802.11 standard can reach 50km. There is no limitation of distance in the actual protocol. With high gain antenna, line of sight and enough power you can shoot much more. Actually there was a trial in Sweden using metorolagical baloons and they did more than 200km with 802.11b. The main concern is LOS blocking due to curvature of the Earth. If you want to shoot 60km and have one antenna at ground level, you need to have the other one at least 200m high. And this will aplly to 802.16 too.

Why not free?

[Anonymous Coward]

Currently cities provide free em in the form of street lights. If you only needed one AP per block your total cost could be much less than the cost of all those street lights. The main cost would be the network connections to the APs. Or they could use Mesh APs. The city could use the Wireless network for police, Fire, or city services. The wired network that they tie it all to could be used for other types of monorting systems like traffic cams.
The only cost would be the connection to the internet.

The way it is to play out.

[Zebra_X]

As I see it, this is the beginning of something truly great. Wi-Fi in the home, around the block, in the coffee shop has been a truly nice thing. A convience that we all find ourselves saying "I wish more places had this (and I didn't have to pay an arm and a leg)".
It won't be much longer before that is the case. All we need now in boston, is one or two major providers to come in and set up MAN AP's on the tallest buildings in town, alternatively several large towers within a few miles of the city. We'll find that any jo-shmo will be able to uplink to those tall buildings or towers and instantly be connected to an "Aether-Pipe". The provider of the MAN uplink would obviously charge his or her customers for access but that wouldn't be a problem because the hotspot operators would then charge the end user, much like the myriad of DSL providers out there now.
Much of the cost difficulties with setting up an AP will be solved by this "AEther-Pipe" a ubiquitous backhaul that, provided you've paid your MAN provider you'll have access to. I'm sure we'll see AP wars in the future. My guess is that we'll need a new incarnation of 802.11x that will be able to put you on the MAN network and hand off from one spot to another as you cruise around the city. Just think an 11Mb/sec enable wrist phone... I can watch DVD's now!

-Hal

Last Post!

[alpg]

I:
The best way to make a silk purse from a sow's ear is to begin
with a silk sow. The same is true of money.
II:
If today were half as good as tomorrow is supposed to be, it would
probably be twice as good as yesterday was.
III:
There are no lazy veteran lion hunters.
IV:
If you can afford to advertise, you don't need to.
V:
One-tenth of the participants produce over one-third of the output.
Increasing the number of participants merely reduces the average
output.
-- Norman Augustine

- this post brought to you by the Automated Last Post Generator...