RE: [EFM] EFM Requirements
I lot of comments have been made about the Shannon Capacity calculations
over the past week or so. The Shannon Capacity calculation is the foundation
of all rate v. reach calculations, but it is only the beginning. The rate
reach calculations require many variables to be entered into the capacity
equation in order to calculate the rate, and there is not a single answer.
One obvious set of variables are the background noise, which standards have
set as -140dBm/Hz, and the number and types of disturbers. Another important
variable in the equation is the directional duplexing method used. The 3
major duplexing methods in use today are full-duplex, single frequency band,
echo cancelled (EC) (such as ISDN, HDSL, HDSL2, g.shdsl), full-duplex
frequency division duplexing FDD (such as ADSL, and VDSL), and time division
duplexing (TDD)(some with centrally coordinated transmission times, such as
Japanese ISDN, or past TDD VDSL proposals, or others with independent,
dynamic transmission times and symmetry ratios, such as EtherLoop, which is
deployed today, and 100BASECu, which Elastic and others are proposing for
EFM copper PHY.)
Even after the duplexing method has been selected, other details need to be
factored in to calculate the rate, such as, what are the bandplans, PSD
levels, and what power back-off method is used, if any.
An independent SNR margin engine, based on T1.417, can be found at:
click on registered users
Login Name: spectral
for questions contact: Ken Kerpez, Telcordia, 973-829-4747
You can use this to calculate, for instance, that the ADSL capacity, with 49
Spectrum Management Class 5 (ADSL) disturbers on 12kft of 26 AWG wire is:
522.66kpbs upstream, and 2879.62kpbs downstream, which, of course, is
greater than 1Mbps.
You can also use it to calculate that the symmetric rate for G.shdsl, with
49 self-NEXT disturbers, on 12kft of 26AWG wire is 1088kbps, again greater
than 1Mbps, but you will also note is different than the rate for ADSL.
I give the examples above to show that there is not a single "capacity
calculation" for a given loop length.
From: Steven.Haas@xxxxxxxxxxxx [mailto:Steven.Haas@xxxxxxxxxxxx]
Sent: Monday, August 20, 2001 9:32 AM
Subject: RE: [EFM] EFM Requirements
The only way to calculate what can be carried over a copper pair is with
capacity calculations. These take in to account the mutual affects between
pairs and the limitation on the transmitted power density. A additional
method is to look at the rates provided by existing systems that are
standardized for the public network. These systems (SHDSL, ADSL, VDSL)
provide an aggregate rate (up and down) that is indicative of what can be
acheived at the different ranges.
The public network standard bodies have performed calculations for 1/2 a
mile during the VDSL standardization. These calculations assumed 20
additional systems in the binder, a limit of -60 dBm/Hz on the PSD and of
course, legacy systems in the cable plant. At 1/2 a mile, with AWG26 cables
and all 20 systems providing service, a system can provide a symmetric rate
of about 13-15 Mbps.
For 12 Kfeet, I suggest you look at ISDN, ADSL and SHDSL for a feeling of
what you will get. From a capacity calculation point of view, with a PSD
level of -40 dBm/Hz, you can provide less than 1Mbps in either direction.
The FSAN, a group of operators, has a capacity calculation tool on their
website. This tool is open for anyone to calculate what can be provided over
the public copper plant. It provides realistic values and is an independant
tool. It is at: http://www.fsanet.net/index.html
Infineon Technologies Savan
6 Hagavish St.
Poleg Industrial Park
Email: steven.haas@xxxxxxxxxxxx <mailto:steven.haas@xxxxxxxxxxxx>
Tel: +972-9-8924186 (direct)
Tel: +972-9-8924100 (switchboard)
From: ramu [mailto:ramu_raskan@xxxxxxxxxxxxx]
Sent: Monday, August 20, 2001 12:21 AM
Subject: Re: [EFM] EFM Requirements
Rich, you may not be _advocating_ new copper drops (hopefully no one is!),
but the scenario you describe would require it. The vast majority of
in-place copper simply does not drop into the scenario you describe.
But, for the sake of argument, let's say new copper was laid so that
everyone was within a half mile. What kind of sustained, symmetric bandwidth
are you suggesting you can provide at 1/2 mile distances over 26AWG pairs,
and what happens if this ever needs to be increased? Anyone else who has a
copper solution that would like to answer is welcome to also. For comparison
purposes, what sustained, symmetric bandwidth you could provide at 12,000
feet would be useful to hear also, since 12000 feet seems to be a number
that is of interest to many here. Please provide any other assumptions made
to arrive at the bandwidth numbers. Patrick Stanley from Elastic, can you
provide you numbers for this scenario please?
On Sun, 19 Aug 2001 12:59:38
Rich Taborek wrote:
>It seems we have a major misunderstanding. I'm not advocating any new
>copper drops. I believe that Vladimir's note, which I responded to, was
>not advocating new copper drops for the first/last 0.5 mile as well.
>This is what Vladimir referred to as the "copper tail". The copper tail
>connects to the residence/business on one end and fiber on the other.
>The ONLY issue my response addressed was whether the fiber architecture
>0.5 miles away from the residence/business was PtP or PON. It is at the
>copper/fiber juncture that E/O conversion is obviously required. I
>further assume that the bandwidth requirements at this 0.5 mile central
>point are such that high bandwidth technologies such as 1000BASE-X or
>even 10GBASE-X make more sense than PON.
>I do like the idea of FTTH, but I'm being practical and noting that if
>the fiber gets to 0.5 miles of the home that high enough bandwidths can
>By "more scalable" I mean that a PtP fiber architecture is more scalable
>than PON. I note that one can deploy existing cost effective technology
>such as PtP 1000BASE-X and achieve 1 Gbps throughput. I expect that the
>relative cost of achieving the same throughput with PON would run two
>orders or magnitude or so more.
>I hope this helps.
>> Rich, not sure I understand the logic behind some of your points, but the
simple answer appears to be that it requires new trenching for the new
copper drops, which will never fly. If I misunderstand, please elaborate.
>> I don't quite get you conclusion: 'since E/O is required, 1KBaseX is
significantly more cost effective, scalable, and simpler than PON.'
>> If E/O is required in the field it can't be PON. Whether it is
significantly more cost effective cannot be judged without detailed designs
of each. Whether it is more scalable I guess depends on your definition.
Simpler is hard to imagine, but again may depend on definition. If you could
elaborate your view I would be interested.
>> Since you are suggesting an active architecture with new electronics in
the field, I for one would welcome a discussion of such a network
architecture that had fiber for the last thousand feet. No one has addressed
that at all to my recollection. None of the architectues is perfect in all
respects so an all-fiber active architecture undoubtedly has some
>> It would seem that the most cost effective approach for a 10 mile EFM
>> solution would be to use standard point-to-point 1000BASE-X or 10GBASE-X
>> for the first 9.5 miles and then a 0.5 mile copper tail for the
>> first/last half mile. Since E/O conversion is required at the 9.5 mile
>> mark, standard 1000BASE-X or 10GBASE-X technology would be significantly
>> more cost effective, scalable and simpler that PON at that point. I
>> expect that there will be switching equipment located at the upstream
>> (10 miles away) side, negating any benefit of a PON split at that point.
>> Please tell me what's wrong with this picture?
>> P.S. I understand that this does not address the rural market portrayed
>> by Frank Miller in this thread, but neither does PON.
>> Best Regards,
>Richard Taborek Sr. Intel Corporation
>XAUI Sherpa Intel Communications Group
>3101 Jay Street, Suite 110 Optical Products Group
>Santa Clara, CA 95054 Santa Clara Design Center
>Cell: 408-832-3957 mailto:rich.taborek@xxxxxxxxx
>Fax: 408-486-9783 http://www.intel.com
Get 250 color business cards for FREE!