Slashdot Mirror
155Mbs Over Copper Lines
FIGJAM writes: "Actelis, based in Fremont, California, have raised $26 million to enable speeds up to 155Mbs over copper lines by overcoming crosstalk interference with error-correction." The article is unfortunately short on technical details, long on current telecom market financial conditions. There's a bit more information on their website, but be warned, it's nearly as buzzword centric and glib. The speed of fiber over existing copper is still too tantalizing to ignore, though, even if it's not as sexy as actual fiber to the home.
That would be milli-bits per second.
1/1000(or is that 1024?)th of a bit per second.
155mbs over copper is a piece of cake! We had those speeds years ago.
Fiber optic cable is not just another type of wire. It's much more difficult to work with, and installation (on a large scale) requires more skill than your average phone/cable company truck monkey posesses.
Besides, what in the world would make communications companies WANT to do such a thing? If the current broadband market has shown us anything is that residental internet is a suckers game. There's no profit in it.
I have the first issue of Wired magazine, and there is a brilliant column (back page) by Nicholas Negroponte (of the MIT media lab) -- he makes a fairly profound and accurate statement: by in large, the massess are not willing to pay extra money for extra quality. The article was about HDTV an his assertion was that it would never catch on (by market forces alone), unless the government forced it on us via regulations.
The same can be said of broadband -- the masses are overjoyed with AOL and MSN...the rest that arent arent big enough of a market to bother with.
I've got wireless 11MBps Internet connectivity from my laptop. Unfortunately there's a bit of a bottleneck where it goes into my ISDN line upstream limiting me to 128KBps but hey though, I've still got 11Mbps wireless from my laptop!!! I bet the ISPs are lining up to turn up gigabit speeds from the ISP to the home for $40/month.. afterall, ISPs don't pay for their bandwidth right? :-)
http://www.qwest.com/about/qwest/QwestCyberCenters /NA_IP_Network_map_large.jpg
http://www.above.net/network/index.html
It's wonders what you can do when you also own the dark fiber/DWDM gear... Split a few lambdas off for yourself, sell the rest to cover your costs... Not too bad if you can afford it.
- ------------
-----------------------------------------------
---
Segmentation Fault ( core dumped )
More importantly, if everyone had fiber at their home, no one's connection would be any faster than when we were all using 14.4K modems.
All the widespread existence of ridiculously fast connections to the backbone will yield is a ridiculously clogged backbone. Running fiber to businesses and homes will just push the bandwidth block back a step. It won't fix jack.
--
Do you have a
That's what I get for doing math in my head at this early hour. 1000 mbps would be 1 bit per second. I was obviously thinking something about seconds per minute for some reason.
I see even classic Slashdot is now pretty much unusable on dial up anymore.
If you're so far away from the Central Office that voltage drop eats up that 48 volts, then your phone wouldn't work either. The problem with DSL and distance is that DSL runs at a higher frequency, i.e, needs more bandwidth, better high frequency response, than your phone. As wires get longer not only does resistance go up, but impedance does as well, and that tends to roll off high frequencies, so there's a point beyond which the length of the wire between you and the phone company is too long to deliver those higher frequencies.
I see even classic Slashdot is now pretty much unusable on dial up anymore.
mbps=millibits per second. 6000 mbps would be 1 bit per second, meaning about 10 seconds, including overhead, to transmit one byte.
I see even classic Slashdot is now pretty much unusable on dial up anymore.
(In the case of power lines I don't mind it so much. 60 Hz 120 volt AC is pretty much the same no matter from whom you buy it. The company from whom we buy electrical power is the same company from whom we buy delivery of that power. If we were buying power from some company other than the one that owns the wires that run to the house, the people with the package deal would probably get any of their problems dealt with first.)
I'm a lot less happy that our choice of cable TV providers is either Time-Warner or Time-Warner.
I'm beginning to wonder if the local telco (Carolina Tel&Tel as swallowed by Sprint--they've been saying "real soon now" on DSL for a couple of years) isn't waiting for Road Runner to put all the (few) local ISPs (which are all owned by regional companies now anyway, my account's been absorbed twice so far) out of business before they offer DSL so that they can be the only ISP available over DSL, and then they'll undercut cable by a few bucks to steal some Road Runner customers and grab all the new ones that come along.
I wonder if a lot of people will stay with cable anyway (I'm sure TW will offer some sort of TV/Internet package deal and make up the difference overcharging cable-only customers) just to avoid changing their email addresses.
I see even classic Slashdot is now pretty much unusable on dial up anymore.
I don't see telco's ever offering this service. Look how long they've drug their feet with DSL and the like. Hell ISDN has been out since the mid 80's, and telco's still can't seem to get it installed right. (At least Ameritech can't) DSL? Hell my ex back in 97 was beta testing ADSL from Ameritech in Ann Arbor, and yet years later, Ameritech has yet to fully roll it out. Honestly, how hard is it to support? Just add a few boards into the telco switch and volia, you have support.
If this was something that was available and at a reasonable price, it's something that I can see many many people picking up. I'd of course be in line for that much bandwidth. (Voice over IP comes to mind). But knowing how slow and lethargic telcos are, it will be at least 2010 before you start seeing anyone in the US with this service and much much longer before it's available nationwide as well as globally.
</rant>
--
"If you insist on using Windoze you're on your own."
"If you insist on using Windoze you're on your own."
In Canada, the telcos generally lay fiber right to the foundation wall of new homes.
:-)
They're laying copper to homes anyway, and the big expense isn't the fiber -- it's cheaper than copper -- but installation. Expensive to dig trenches, lay conduit, and all that shite. Costs damn near nothing to toss a bit of fiber in the pipe at the same time.
So there's a lot of dark fiber out there, just waiting for the profitability point to make it worth it for the telco to turn it on.
You should move to a modern country. We've got ADSL all over the damn place. Can't get away from it these days...
--
--
Don't like it? Respond with words, not karma.
Hmmm, I'd personally rather have fiber. Why would I rather see fiber widely distributed? Because it is multifunction. A single fiber line can carry hundreds of TV channels, tons of voice channels and even data channels for us internet users. Coax could do the same thing if cable companies and telephone companies were allowed to be a little more friendly with one another (ergo, getting higher throughput on copper means it could compete with fiber thus this article is is interasting however this is pushing the limits of copper wiring's data rate while optical fibers still have room left for bandwidth). Using fiber networks wouldn't necessarily destroy the country's backbones either, if anything it would spurn the usage of higher data rate routers and lines. IIRC the fastest backbone in the US is the OC-192 between Chicago and St.Louis owned and operated by MCI. RUnning fiber to homes and businesses is not going to automatically mean they've got an OC-3 in their garage. The owners of the fiber would just allocate data channels (just like on copper lines) and sell them to whoever wanted to use them. Video could be transmitted on channels a-b whilst data/voice is allocated to channels c-d with e-f (the letters obviously represent ranges of individual data channels) reserved for other usage. Wiring inside houses could easily be preserved by connecting it to a small autonomous bridge in a box on your garage. I could envision fiber to homes but this is marred by the reality that cable companies and telephone companies are both pretty stingy and the only real way to get fiber to the home is for the government to do it and control the lines themselves letting others fill them with content.
This brings up an interesting point, should municipalities get into the fiber game? Not for profit but merely to seed the technology's availability, especially since private organizations are not willing to do it. Routine maitenance of sewers, streets, and power grids could be used to lay dark fiber for later illuminating. If the process was tacked onto work already being done it could be made cost effective and it offers greater local competition for service. You'd get the fiber run to your house and opt to either keep your existing copper service to upgrade to fiber. With bridges linking the copper in your house and the fiber outside you wouldn't need to spend several thousand dollars on new equipment right off the bat. I guess I'm warm to the idea because my city has a municipally owned electricity company and is thus fairly exempt from the troubles of the rest of the state's privatized electricity generators.
I'm a loner Dottie, a Rebel.
On their website, they have a press release that calls it patent pending 'SPATIAL DIVISION MULTIPLEXING TECHNOLOGY' (SDM), and they only claim DS3 (45Mbit/s).
They seem to have already tested it in the field, which explains the funding.
On another page of their website they mention a bit-error rate (BER) of 10E-10, one bit error per 1 billion bits at 10-100Mbit/s.
All they're saying that they are 'controlling the crosstalk interference'. Maybe they're just the first to try that, or maybe they did some advanced crosstalk modeling. I wonder how well that works if the crosstalk is caused by another SDM line?
Will this be another bluetooth? Works well in the lab, but when you cram a couple of devices close together it doesn't work as well?
--- Hindsight is 20/20, but walking backwards is not the answer.
waittta
/.
shoulda
lotsa
Goodness. Jar-Jar is trolling
--
--
"Outlook not so good." That magic 8-ball knows everything! I'll ask about Exchange Server next.
Good point... even if these can get out to homes at reasonable prices, this is gonna be worse shared bandwidth than cable modem technology; cables are 4mbps average (around here anyway), sharing a couple T3's or maybe an OC3 for larger area isn't that bad - but one of these copper links is the speed of just one of those kinda-expensive OC3's, and i don't think the phone companies are going to be too eager to install huge pipes to fully take advantage of a 155mbps link to each home...
Try getting the price on 1000 km and compare *those* instead.
The high price of your tiny cable is not because of the fiber, but because of the connectors (5-10 bucks each, IIRC) and the labour, which was jacked up because fiber guys are very well paid.
Anybody can get a Cat5 crimper and a bag of cheap-o connectors and start cranking out cables. I've done fiber work and the equipment is *very* expensive, not to mention the attention to detail it requires to keep your signal losses minimised.
Go check out the proposed MPLampS, Electricity over IP draft.
;-)
There is hope for the people of Fremont.
the AC
Hemos is like...sci-fi fans;he thinks technology is cool, but he hasn't bothered to understand the science it's based on
That isn't funny; it's probably the trick, the key technology! Sometimes called an inverse mux, you take multiple lower-speed channels and combine them.
So they pump copper to its limit, do it over a bunch of pairs in parallel, and get a lot of bandwidth in the aggregate. Big whoop!
There is some effort needed in controlling interference, because most DSL technologies are mutually interfering, and degrade in speed when the subscription rate increases. But it's not going to give you 155 Mbps over a 5-mile loop from the CO to your house. Never will.
It's so nice to live in a small enough city that doing a complete rebuild of the cable system is actually feasible.
------
It's not theft; It's copyright infringement. Since that's been discusses 800 gazillion times already, I won't discuss it here.
------
Don't go putting Canada down. It's only 40% of our income...
------
...or actually something like 15,000 feet, which is the telco's DSL limit.
I'm no telco guy, but I know a couple and dimly recall conversions like '...some people are 40,000 ft or more from the c/o...' and '...the signal quality decreases rapidly when the DSL limit is reached'
1 Gbit @ 100 meters is here now. But if you recall from geometry, the area of a circle increases rapidly with the radius, so you get a lot of population outside the current DSL limit...and a horrendous amount outside the 100 meter limit.
This technology could change all that.
Treatment, not tyranny. End the drug war and free our American POWs.
Treatment, not tyranny. End the drug war and free our American POWs.
See my user info for links.
Because where I live, my connection is the equiv. to a post-nuclear wasteland... I don't know if that made sense..
anyway, 26k dialup is all I can get.
no 56k
no cable
no dsl
no 128k ISDN, but I can get 64k ISDN for $115 a month.. don't think so
wireless, I don't know yet.. probably not, considering I have trees around me
satellite, high latency sucks
tin cans and fishing line, I don't have enough to reach that far
I hate the internet,
Klowner
Not all telephone lines are CAT3... my home (built in 1994) has two-pair (4 conductors) and about 8 twists per foot (somewhere in the middle of cat3 and cat5). My older home had flat (no twists) two-pair running in the walls. I have no idea what sort of cable comes from the telco to the pedestal in my yard, but from the pedistal to the box on the side of my house is six-pair (12 conductors) with probably 8 twists per foot and it's thick (22 guage versus the 24 guage normally used for telephone and network wiring).
Good point... even if these can get out to homes at reasonable prices, this is gonna be worse shared bandwidth than cable modem technology; cables are 4mbps average (around here anyway), sharing a couple T3's or maybe an OC3 for larger area isn't that bad - but one of these copper links is the speed of just one of those kinda-expensive OC3's, and i don't think the phone companies are going to be too eager to install huge pipes to fully take advantage of a 155mbps link to each home...
VERY wel put! The nearby ISP here pays $13,000 per month for each of their upstream DS3/T3 connections (45mbps to UUNET/Worldcom). OC3 is 155mbps and is in the neighborhood of $35,000 per month.
True, the fiber backbone isn't the bottleneck -- but the routers and switches *are*. Points where you need to convert from optical to electrical and often back to optical again. That's why optical switching companies were the darling of the tech sector until the bust -- while there's more than enough fiber to last us for years to come, the switching technology is still woefully inadequate.
---
Sorry, but you can stop holding your breath. Fiber is an excellent backbone technology but it's too cumbersome, expensive, and difficult to deploy to ever become a "last mile" solution.
People clued in long ago that fiber to the workstation on corporate LANs was far more expensive than copper and offered no speed increases. Fiber to the home is the same -- it's cheaper to just install better quality phone cable or find better transmission methods for existing cabling. Why spend all that money on fiber?
---
Actually it is no where near that simple. There are a lot of problems with rolling DSL service out, and most of them are a function of lack of resources. Ameritech in particular lacks the manpower and cable availability to properly support a rollout of DSL.
Let me see if I can explain why a little bit. Back in the day (~15 years ago and earlier) the local telco would provision for basically 2 lines per household. At some point, shortly before the explosion in popularity of modems, cell phones and all the other toys we love to use, it was decided that this was wasteful and that fewer lines would be provisioned. This meant that that new subdevelopment would have only a minimal number of extra lines available to it. This also meant that once folks started using extra lines for modems, working at home, and the like, that upgrading them was no longer simply a matter of adding a few switches. Now they had to go and dig up that line and add more lines.
But it gets worse. DSL only works when the customer is close to the central office and when they have an uninterrupted connection to it. If they don't the line has to be "conditioned" to make it work. Have a bit of fiber cable between you and the office? Chances are you'll never get DSL any time soon. Conditioning the line is expensive and time consuming. Since the phone company isn't going to eat the cost of doing it so they can take a loss by billing you for $40/month which is much less than the several thousand dollar cost of doing the line.
Even if you are all set up from a technical point of view, there still are the issues with getting the equipment on the user's computer to work. The phone companies are simply not experienced in or set up for supporting end users with complicated copmuter equipment. Phones are easy, computers aren't. Remember, the local telco is still esentially a monopoly and has the same bureacracy and mentality of one even 20 years since AT&T split. I've seen it first hand. Anyone who thinks that these companies became nimble and customer responsive just because AT&T was broken up has zero knowledge of how they work on the inside. They are still not set up to hand customer service on the scale DSL requires.
There's a lot more to it than this but I'm tired of typing. Point is, there are a lot of issues we as customers don't generally see. General DSL rollout is going to take a LONG time. Learn to live with it.
My brother is the manager of the IT department for a local utilities concern. He got the job when the power co bought the telco he worked for. They've started rolling out DSL service in addition to voice and power. They have a few DSLAMs(a fairly small service area, one county in North Texas) and on the other end of the DSLAM is a tie into TWO DS3s. That's right, they have TWO DS3s as their pipe into their DSLAM, and a T1 for backup. Not only that, the DS3s are tied to two different networks so if one is having problems, they re-route. It's a great setup and makes me wish I lived in their service area.
Steven
-- I have marked myself unwilling to moderate-- I don't have other accounts to artificially inflate the karma of
You guys are killing me here. I work in the Telco industry, and I have a few points to address.
1) This technology already exists. It has for a while. It's called HDSL. When we need bandwidth at a remote site that is not fibre fed, we will install HDSL shelves to carry the payload. Install enough of them, and you will get your 155MB/s rate.
2) All the people that are whining about "I live in the city, less than 10,000 ft from a C.O., and I can't get DSL." . YOU may live 10,000 ft from the CO, but how far does that wire that feeds your house extend PAST your house. It may continue on for a mile or more past your house.
3) People who say "X" company has an OC-192 connection. No, actually they don't. Show me a company that has a OC-192 connection, and i'll show you a company that is blowing smoke out their ass. OC-192's (and recently OC-768) and the DWDM technology are used by Telco's as backbones. Do you really think a Telco is going to allow one of their backbones to be saturated by ONE customer?
4) Fiber to the house. Unless your house was built with it, don't expect to get it. The costs would be ASTRONOMICAL to lay a fiber to every house that currently has copper. Opening trenches is VERY labor intensive. (read:$$$)
5) To put DSL in a CO, its "just a couple of cards in the telco switch". WRONG. The whole point of DSL (other than speed increase), is to offload the processing/routing of the data away from the voice switch. To put DSL into a CO, they have to install an entirely different piece of equipment, called a DSLAM (Digital Subscriber Line Access Multiplexer). This box strips off the higher frequencies containing the data traffic before it hits the voice switch and routes it over the telco/ISP's ATM network, freeing up resources in the voice switch which would otherwise be dedicated to dial-up connections.
6) "It would be cheaper to use fiber than copper, because of the cost of a CSU/DSU and the setup." WRONG. Do people think that you can just plug this fiber into a fiber modem and BAM!, you have an OC3? THE most expensive part of a dedicated fiber connection is the ADM, which sits at the customer end and pares off/adds on the customers traffic from the other traffic on the ring or linear connection. (Add/Drop Multiplexer) The cost of a ADM is substantially more than that of a CSU/DSU.
7) And finally, to the person who stated "Forget copper, I want to get SONET". You can get SONET over an electrical connection (read: copper), its called an STS-x connection rather than an OC-x connection. I believe the STS-x connections now have a rate as high as STS-12.
/end rant.
Cool, people in Fremont will be able to get high speed Internet through their powerlines during blackouts.
"A door is what a dog is perpetually on the wrong side of" - Ogden Nash
Unfortunately, the link to the home is just one of the bottlenecks. I've had a cable modem for a couple of years now, and speeds still suck. It isn't because of the wire to my house, either; downloads from @Home's newsgroups, which are hosted on their own servers, hit 3-4Mbps. For most downloads from non-@Home servers, it's rare if the speed exceeds 300-400 kbps. The servers themselves and paths on the internet at large are going to have to be improved for high-speed access to really meet its full potential.
"If I have seen further than other men, it is by stepping on their glasses." - Michael Swaine
imagine a beowulf cluster of these.....
waittta minute....
sorry wrong obligatory....
shoulda been:
lotsa more porn...a video for each of my monitors....
...downloads from @Home's newsgroups, which are hosted on their own servers, hit 3-4Mbps.
I suppose the TCP window would only get big enough for you to clock a speed like that if you were downloading a...let's see...a large binary image? From a newsgroup?
And just where exactly have you been hanging out on this "Internet," young man?
I got my Linux laptop at System76.
Don't feel too sorry for yourself. Because of crappy wiring, I max at 31.2k, and I live in a VERY large college town, with a population of 200,000!
I wouldn't assume that the average phone/cable company employees are too dumb to lay fiber. I've seen wiring jobs by the telco, and they quickly approach a degree of complication that I don't want to mess with. I've also seen the tools used by telco employees, which seem to have a level of sophistication that monkeys would be unable to use.
Its like assuming that since I can use the new iMac at home, those network admins in the back room must not require that much training. Hell, its just a computer, you turn it on or off, what more do you need to know. :)
Spatial Division Multiplexing is just another way of assigning specific areas to transfer data in parallel, instead of using a serial transfer method. It's basically dividing a space up into several channels.
I know you can do this in optics: for example, use only one part of an optic for a specific channel, etc. but since copper is conductive you can't define boundries, which is where I assume the cross talk correction comes in.
However, SDM is very inefficient for copper networks. A cousin, frequency division multiplexing, which is used in optical networks, DSL, cable networks (frequencies divided up into channels) is much better suited. Fiber uses it as well.
Time division multiplexing is also used, and on high capacity optical pipes DWDM systems are used which multiplex channels over several wavelengths, which can also 'layer' FDM and other modulation and frequency style methods to get even more bandwidth. A single optical fiber has as much bandwidth as the diffraction and interference limited wavelengths seperation times the data frequency. You could very easily fit all the world's Internet traffic onto a single fiber using a very sophisticated DWDM-FDM (with wavelength spacings at the physical limits). But then you need to be able to seperate each very close wavelength out, and demultiplex the several GHz modulation of each thereof (picosecond response, anyone?). Not to mention the generation which involves inverse problems of the same nature.
But as far as SDM on a copper line, this is defeating its own purpose. IMHO, its better to have a high clock and go up in bandwidth on the line. Copper can handle 100's of channels of UNCOMPRESSED video, just look at your cable system. Your cable modem can barely handle 2 Mbps, which is a far cry from raw NTSC. We aren't even using a fraction of that for gigabit networking fiber-over-CAT5 as it is.
This may be an interesting method, but it's a step back. It might be good if it can do it over all the ancient POTs stuff and go to consumers, but then again it relies on self-correcting cross talk and interference models, which can only be taken so far and are only ideal in *ideal* conditions.
What we really need is a good CAT5 going to every house providing all data services, and DWDM hubs to transfer and enable dedication and not bandwidth sharing. But as someone else mentioned, the telco's like to keep us in the dark ages, so they can milk out as much as they can from their nth generation system. They'll take it for as long as possible, before they're forced to get up and realize those research dollars back in the 1960s have long since been payed for.
"I'll just chip in a bit for RedHat: I actually have that installed on my university machine." - Linus, '95
Your local loop provider does not want to deploy this technology....here's why: Local Verizon T1 (1.5 mbps) fiber charge: about $400.00/mo. Local Verizon DSL (1.5 mbps if you can get it): about $150/mo. Verixon is not really pushing business DSL at these speeds because that would canabalize their T1 business. Do you think for a minute that Verizon would allow cheap, 155 mbps over copper circuits to destroy their T3/DS3/OC192 business? No way! The local telcos like selling fiber because it's viewed as "slightly exotic" and worth the higher monthly fees.
For some reason, when I saw the article I thought of cat5.
The article had me going for a while as I tried to figure out what the big deal was, I mean, we've had 1000Mb for copper for years now : )
Based on upvotes, Ageism is the only "-ism" Slashdotters care about and think isn't SJW
that much more income for the cable companies who insist that they're pushing all they can out of the lines by providing me with an 80 KB downstream cap
Screw 3...