Copper Wire Could Push 4 Times the Bandwidth of Fiber Optics
(5 votes, average: 4.2 out of 5)
Loading ...
by Mike Zazaian October 10, 2006 - 7:37pm, 12 Comments
By using a technique called Dynamic Spectrum Management, internet providers would be able to achieve DSL bandwidth speeds as high as 10Gpbs through copper wires. At such immense speeds, the already-laid copper wires could achieve speeds up to four times faster than the 2.5Gbps throughput at which standard fiber-optic cables top out.
Current bottlenecks on DSL connections are caused by static spectrum management, a technology that prevents severe service degradation in wires that may be used by multiple services. It essentially ensures that nobody gets terrible service, also hinders service symmetry and greatly lowers the highest possible speeds in copper wire solutions. DSM improves upon the current SSM implementation, allowing each DSL connection to be regulated in real time based on the immediate data needs of each customer. By just changing the way data is transferred through copper wires, individual users could achieve download rates as fast as 100 megabytes per second, more than twice as fast as Verizon plans to offer with its Fiber-to-the-Home (FTTH) FiOS network.
The main obstacle for the advancement of DSL technology is the interference generated from different DSL lines that share the same telephone cable binder,
said Professor John Cioffi, Professor of Engineering at Stanford University. DSM is a promising technology for the future evolution of broadband access networks using existing copper infrastructure.
The news couldn’t come at a worse time for Verizon, who has already committed over $18 billion to the expansion of the aforementioned FiOS network. The primary costs of such FTTH networks lie not with the actual implementation, which at $933 per home costs only slightly more than an $800 per-home cable installation. Instead, Verizon’s core costs come from the need to pass fiber optic cables by every single house that it plans to eventually connect with, even if those houses choose to opt out of service in the interim. If only 10 percent of customers then choose to adopt the service, Verizon’s costs ring up at around $10,000 per home.
And while such FTTH networks are already prevalent in Japan and South Korea, the increased bandwidth of DSM solutions would eliminate the primary costs that Verizon faces as it plans to extend its FiOS network to 18 million homes by 2010. And while DSM is still in development, its implementation would allow for a less costly Fiber-to-the-Curb solution, in which service providers could avoid the enormous costs of passing fiber by every house in a given network.
Still, DSM is in its infancy, and doesn’t yet mean the downfall of Verizon’s FiOS or Fiber-to-the-Home. If DSM gets off the ground any time soon however, Verizon may have a whole lot of ’splainin’ to do to its investors about the $18 billion hole in its pocket
[via ars technica]
.
I just hope they companies won’t price gouge the new speeds.
Tommy Lee
First, I’ll believe it when I see it. Second, 10 GBps is nothing compared to what the fiber optic strand can handle, the only limitations so far are the lasers at either end. If this new copper technology comes out, they won’t be the fastest on the market for very long. Copper wires are destined to go the way of the dinosaur.
Jeff
Given the upcoming bandwidth requirements to the average home, this should be just what the doctor ordered. To compare this with the potential of fibre is a bit pointless, given that fibre bandwidths are far in excess of what an average consumer will need for the forseeable future.
CB
Found your DSM article very interesting. Any information on the distance limitation of this technology?
Shaun Hollingsworth
Looking back through the documentation for the technology I couldn’t find any specific distance — but it seems like this would apply to any DSL installation. I suppose that means that it shares the same limitations as DSL, which at present is a maximum of 18,000 feet. For more info on it check out the DSM homepage from Stanford.
Mike Zazaian
“And while DSM is still in development, its implementation would allow for a less costly Fiber-to-the-Curb solution, in which service providers could avoid the enormous costs of passing fiber by every house in a given network.” - the article
This suggests to me a much shorter loop requirement, in the less than 300-m ballpark. Otherwise, the best solution would be from the node, not one that involves placing fibre to the curb and installing electronics in the pedestals, no?
Blair Lemon
we are doing 10G optically today (have been for several years)…..
fiberguy
The author of this piece, Mike Zazaian, should go back to covering high school sports for the local paper instead of writing technical articles. I do not work for Verizon so I’m not biased that way. However, feeling the need for full disclosure, do work for an ILEC/ISP. I know in 1996 AT&T muxxed 100 - 10GbE streams siuiltaineously over DWDM equipment and that was 12 years ago, back when residential broadband was in its early infancy. Yes, ONE TIGABIT PER SECOND!)Who ever tells you that 24 gauge twisted pair copper over distance can carry even as much bandwidth as fiber is as big idiot as the author of this article
The only thing in this article that is relevant is Verizon’s 2.5GbE over a 1X32 PON splitter using , I believe GPON. There is already available CWDM and DWDM PON equipment for use in FTTP networks. Since the limitations of bandwidth is equipment based, rather than theoretical limits of the glass itself, fork-trucking the OLT & ONT is the only requirement for bandwidth upgrading.
George
Yes the limitations on fiber-optics are the lasers on the end of the runs. But to replace all the copper wire that has been laid for POTS (plain old telephone service) would be not cost effective for the phone companies.
Mike