No Subject
From owner-stds-802-3-hssg@xxxxxxxx Tue Feb 29 19:54 GMT 2000
Received: from gatekeeper.pdd.3com.com (gatekeeper [161.71.169.3])
by isolan.pdd.3com.com (8.9.1b+Sun/8.9.3) with ESMTP id TAA21714;
Tue, 29 Feb 2000 19:54:25 GMT
Received: from ruebert.ieee.org ([199.172.136.3])
by gatekeeper.pdd.3com.com (Netscape Messaging Server 3.6)
with ESMTP id AAA156F; Tue, 29 Feb 2000 19:52:44 +0000
Received: by ruebert.ieee.org (8.9.3/8.9.3) id OAA23964; Tue, 29 Feb 2000 14:11:57 -0500 (EST)
From: "Edward Chang" <edward.chang@xxxxxxxxxxxxxxxx>
To: "jmw" <mwincn@xxxxxxxxxxxx>, <NetWorthTK@xxxxxxx>, <vivek@xxxxxxxxxxxxxxx>,
<stds-802-3-hssg@xxxxxxxx>
Subject: RE: PAM-5, what are your BERs ?
Date: Tue, 29 Feb 2000 14:17:06 -0500
Message-ID: <NDBBLFICGKDCKJBHKIAEMEDMCBAA.edward.chang@xxxxxxxxxxxxxxxx>
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2910.0)
In-Reply-To: <4.2.0.58.20000229064314.00a867e0@xxxxxxxxxxxxxxxx>
Importance: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300
Sender: owner-stds-802-3-hssg@xxxxxxxx
Precedence: bulk
X-Resent-To: Multiple Recipients <stds-802-3-hssg@xxxxxxxxxxxxxxxxxx>
X-Listname: stds-802-3-hssg
X-Info: [Un]Subscribe requests to majordomo@xxxxxxxxxxxxxxxxxx
X-Moderator-Address: stds-802-3-hssg-approval@xxxxxxxxxxxxxxxxxx
X-Lines: 110
Status: RO
Content-Type: text/plain; charset="us-ascii"
Content-Length: 4088
Mike:
We are discussing the product developments, which are based on proved
theorems and equations to mathematically calculate all designs. After that,
setting up tests to confirm that the test data, and the analyses are
correlated. Otherwise, repeat the procedures to achieve the goals.
Finally, it will go through extensive system tests and BER tests to certify
a product.
Engineers do not just "looking at a scope", then make the comment. They
theorize it, quantify it, then make comments.
Unless, some one has extensive communication product development experience
in the past, it is hard for a one to grasp, and appreciate the contents of
the comments put forward on reflectors by those serious contributors.
The term, S/N ratio, is a very abstract term to those who are not circuit
designers. However, to the circuit designer, S/N ratio can hurt you as real
as putting your finger on top of a fire.
The cascading of the frequency response of each component to predict the
over-all system frequency response is a well known basic tool implemented by
all circuit designers. It is as predictable as, 1+1 = 2.
Regards,
Edward S. Chang
NetWorth Technologies, Inc.
EChang@xxxxxxxxxxxxxxxx
Tel: (610)292-2870
Fax: (610)292-2872
-----Original Message-----
From: owner-stds-802-3-hssg@xxxxxxxx
[mailto:owner-stds-802-3-hssg@xxxxxxxx]On Behalf Of jmw
Sent: Tuesday, February 29, 2000 10:00 AM
To: NetWorthTK@xxxxxxx; vivek@xxxxxxxxxxxxxxx; stds-802-3-hssg@xxxxxxxx
Subject: Re: PAM-5, what are your BERs ?
Edward, my $0.02:
it is not always the case that because a thing seems to be true on first
examination, it must therefore always be true.
SNR is certainly a critical issue for any receiver in any communication
system. Claude Shannon proved that a long time ago. but aside from
bandwidth and data rate issues (for the moment, of course), if "signal"
and "noise" are sufficiently statistically independent, then the two can
be separated and the signal recovered.
for this case it is not enough to look at a 'scope display, say to yourself
"gee, that looks like nothing but noise to me", and decide you're done.
--
J M Wincn
Cielo Communications, Inc.
325 Interlocken Pkwy, Bldg A
Broomfield, CO 80021-3497
mwincn@xxxxxxxxxxxx
Voice: 303-464-2264
Cell: 408-394-5283
At 01:15 AM 29-02-2000 -0500, NetWorthTK@xxxxxxx wrote:
>Hi Vivel:
>
> >From theoretical point of view, you reasoning makes some points.
However,
>from the real implementation point of view, it is not quite true. Before
>starting analyzing the frequency response, just ask a question: "If we can
>simply keep equalizing the receiving signals to bring them back to the
>looking-alike to the original, transmitting signal, why we bother all those
>bandwidth issues? There must be some limitations to the equalization
>technique.
>
>The eye closure is caused by the insufficient bandwidth of a receiving
path;
>as a result, the narrow pulse (higher frequency pulse) is much more
>attenuated than the wider pulse (lower frequency pulse). We can cascade a
>high-pass frequency response equalizer to suppers the amplitude of a wide
>pulse, and keep the amplitude of the narrow pulse remain unchanged (but not
>amplified) to open the eye. However, if the amplitude of a narrow pulse is
>already too small to meet the minimum S/N requirement, the equalizer is
>useless. Theoretically, a linear amplifier can be added to bring the
signal
>amplitude up to meet the minimum S/N requirement. The linear amplifier
will
>need a BW larger than the transmitting signal rise time. Furthermore, any
>deficiencies in the linearity, will add both timing and amplitude
distortion
>to the received data. The additional distortion is not included in the
>jitter specification; as a result, the link will cause higher BER.
>Especially in a high data rate link, a linear amplifier may cause more
errors
>than the expected benefit. In practice, it is impractical to add a linear
>amplifier.
>
>The right way is to keep eye open at the receiver input.
>
>Regards,
>
>Ed Chang
>
>NetWorth Technologies, Inc.
From owner-stds-802-3-hssg@xxxxxxxx Tue Feb 29 20:00 GMT 2000
Received: from gatekeeper.pdd.3com.com (gatekeeper [161.71.169.3])
by isolan.pdd.3com.com (8.9.1b+Sun/8.9.3) with ESMTP id UAA21818;
Tue, 29 Feb 2000 20:00:14 GMT
Received: from ruebert.ieee.org ([199.172.136.3])
by gatekeeper.pdd.3com.com (Netscape Messaging Server 3.6)
with ESMTP id AAA1581; Tue, 29 Feb 2000 19:58:32 +0000
Received: by ruebert.ieee.org (8.9.3/8.9.3) id OAA25120; Tue, 29 Feb 2000 14:16:52 -0500 (EST)
From: "Edward Chang" <edward.chang@xxxxxxxxxxxxxxxx>
To: "David Martin" <dwmartin@xxxxxxxxxxxxxxxxxx>, <stds-802-3-hssg@xxxxxxxx>
Subject: RE: PAM-5, what are your BERs ?
Date: Tue, 29 Feb 2000 14:22:58 -0500
Message-ID: <NDBBLFICGKDCKJBHKIAEAEDNCBAA.edward.chang@xxxxxxxxxxxxxxxx>
MIME-Version: 1.0
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2910.0)
In-Reply-To: <C7919D8389D9D111A3930000F80824AE043F19FB@xxxxxxxxxxxxxxxxxxxxxx>
Importance: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300
Sender: owner-stds-802-3-hssg@xxxxxxxx
Precedence: bulk
X-Resent-To: Multiple Recipients <stds-802-3-hssg@xxxxxxxxxxxxxxxxxx>
X-Listname: stds-802-3-hssg
X-Info: [Un]Subscribe requests to majordomo@xxxxxxxxxxxxxxxxxx
X-Moderator-Address: stds-802-3-hssg-approval@xxxxxxxxxxxxxxxxxx
Content-Type: multipart/alternative; boundary="----=_NextPart_000_003D_01BF82C0.79F36830"
X-Lines: 1095
Status: RO
Content-Length: 44411
------=_NextPart_000_003D_01BF82C0.79F36830
Content-Type: text/plain; charset="iso-8859-1"
X-Sun-Content-Length: 12910
RE: PAM-5, what are your BERs ?Dave:
Are you suggesting, that if the BER of GbE is 10^-12, then the BER of the
10xGbE should be less? For example, 10^-13 to maintain the expected
throughput gain from 10xGbE?
Regards,
Edward S. Chang
NetWorth Technologies, Inc.
EChang@xxxxxxxxxxxxxxxx
Tel: (610)292-2870
Fax: (610)292-2872
-----Original Message-----
From: owner-stds-802-3-hssg@xxxxxxxx
[mailto:owner-stds-802-3-hssg@xxxxxxxx]On Behalf Of David Martin
Sent: Tuesday, February 29, 2000 10:12 AM
To: stds-802-3-hssg@xxxxxxxx
Subject: RE: PAM-5, what are your BERs ?
Jaime,
A clarification on backbone transport BER spec's. For OC-48 systems,
although BCR GR-253 spec'd
1E-10 for worst case end-of-life, the defacto customer requirement was
1E-12. Similarly, for OC-192
systems, although BCR GR-1377 spec'd 1E-12 for worst case end-of-life, the
defacto customer
requirement has been 1E-15. Note that the current target proposed for the
ITU defined Optical
Transport Network is also 1E-15.
...Dave
David W. Martin
Nortel Networks
+1 613 765-2901
+1 613 763-2388 (fax)
dwmartin@xxxxxxxxxxxxxxxxxx
========================
-----Original Message-----
From: Jaime Kardontchik [SMTP:kardontchik.jaime@xxxxxxxxxxx]
Sent: Monday, February 28, 2000 12:05 PM
To: stds-802-3-hssg@xxxxxxxx
Subject: Re: PAM-5, what are your BERs ?
Hello Vivek,
A few months ago there was a discussion on the Reflector where
some members claimed that the BERs in some optical backbones
reached easily the 10^(-15). The repply to these claims was that
although this might be true, the philosophy of the 802.3, that made
its standards so successful, was to design based on worst case
conditions. Therefore, the Ethernet optical networks were specficied
to a BER of only 10^(-12).
The same holds for the Ethernet Copper links.
Although it might be true that some links of 100BASE-TX
reach a BER of 10^(-12), the official BER adopted in the
Standard was only 2.5*10^(-10). And the same for the
1000BASE-T, where the Standard adopted to support a
BER of 10^(-10).
The reason for this higher BERs in the Copper networks is
ISI (Inter Symbol Interference). Or, in layman's language:
is the eye open or not at the input of the receiver ?
Jaime
Jaime E> Kardontchik
Micro Linear
San Jose, CA 95131
Vivek Telang wrote:
> Hello Jaime and 10G'ers,
>
> I had to jump in to try and correct this misconception of the "eye
opening requirement". While it is true that the "open eye" is always a good
thing, it is definitely not a requirement for low BER reception. If it were
a requirement, the tried and true 100BaseTX Ethernet would not be
functioning today. Many existing implementations of 100BaseTX silicon easily
achieve BERs of 1e-12 over 170+ meters of twisted pair cable, and the eye is
very much closed in this case. Not to mention that near-end crosstalk (NEXT)
makes matters significantly worse. Also, in 1000BaseT, if the only
impairment was ISI, and Echo, NEXT and FEXT were absent, BERs of 1e-12 could
be easily achieved, even with a 5-level symbol constellation.
>
> An open eye is a requirement for systems that use simple
lock-and-slice techniques, where little or no processing of the signal is
done, other than maybe amplification, and a PLL locks to the data clock and
slices the data. In contrast, in a DSP-based system, the combination of
linear feedforward equalizers and DFEs using simple LMS adaptation can
easily deal with "closed eyes". Even with a blind start-up.
>
> The implementation of a DSP-based system (including a 6-bit ADC) at
5GHz in CMOS is a different matter altogether, and I am eagerly awaiting
Oscar's presentation at the March meeting.
>
> Regards,
>
> Vivek
>
> *---------------------------------------
> *
> * Vivek Telang
> * Cicada Semiconductor Inc.
> * 901 MoPac Expressway South
> * Building One, Suite 540
> * Austin, Texas 78746
> *
> * 512-327-3500 x114 voice
> * 512-327-3550 fax
> * vivek@xxxxxxxxxxxxxxx
> * http://www.cicada-semi.com
> *
> *---------------------------------------
>
> -----Original Message-----
> From: Jaime Kardontchik [SMTP:kardontchik.jaime@xxxxxxxxxxx]
> Sent: Sunday, February 27, 2000 6:37 PM
> To: stds-802-3-hssg@xxxxxxxx
> Subject: PAM-5, what are your BERs ?
>
> Hello 10G'ers,
>
> Edward Chang was right on target.
> He wrote on Feb 23:
>
> > In the past, the multiple voltage-level coding
> > was adopted by two LAN standards, ATM and
> > Ethernet. Both of them were twisted-pair
> > applications, and the BER were 10^(-10).
> > I proposed BER of 10^(-12) in both working
> > groups to be consistent with the LAN optical
> > links' BER; however, for some reason, they
> > remain 10^(-10), officially in both standards.
>
> What was the reason ?
>
> Multilevel voltage coding was not the reason.
> On the contrary, it was a remedy. The real reason
> was the strong ISI at the maximum link lengths
> that these Standards wanted to support.
> Multi-level voltage coding was adopted to
> lower the baud rate or the frequency content
> of the signal in order to make the ISI smaller
> and get a better BER.
>
> However, the remaining ISI in these Copper links
> remained high enough that the eye at the
> input of the receiver remained still
> completely closed. It is this closed eye that
> limited the achievable BER in the Copper media
> to 10^(-10). Post-equalizers, no matter how
> sophisticated they were, are not able to
> completely reverse the effects of eye closening
> on the achievable BER.
>
> ---> Conclusion ? In order to get a BER of
> 10^(-12) you have to have a clear
> open eye at the input of the receiver.
> The maximum supported link length
> is set accordingly to meet this basic
> condition: open eye.
>
> Now we can go to the basic two PAM-5 proposals:
>
> 1) PAM-5 serial at 5 Gbaud
>
> Using 1300 nm lasers lasers at 5 Gbaud the
> optical eye is completely closed already at
> ~170 meters due to heavy ISI (installed MMF,
> 500 MHz*km bandwidth).
>
> Oscar claimed a support of 500 meters of
> installed MMF and pointed out that
> DFEs (equalizers) have been successfuly used
> in 100 and 1000 Mbps Copper networks.
> However, the specs and experience of the 100 Mbps
> and 1000 Mbps links over Copper - where the eye
> at the input of the receiver is completely
> closed at the target link length due to ISI -
> put the achievable BERs around 10^(-10) only.
> Even using sophisticated equalizers.
>
> What is the experimental support that a BER
> of 10^(-12) could be achieved when a strong ISI
> closes completely the eye at the input of the
> receiver ? I think that none and that the
> experience points to the contrary. Why then
> could an equalizer running at 5 GHz achieve
> here what sophisticated equalizers running
> 40 times slower (125 MHz) were not able to achieve
> in the Copper Ethernets ? Parallel processing
> could enable perhaps to meet the timing
> constraints of the design by running multiple
> equalizers at a lower clock, but will not
> eliminate the basic limitation on achievable
> BERs once the eye at the input of the receiver
> is already completely closed.
>
> Let us see now the case PAM-5 at 1.25 Gbaud:
>
> 2) PAM-5 4-WDM at 1.25 Gbaud
>
> Let us compare this approach to two other well
> known on-off approaches: 1 GbE and 4-WDM at
> 3.125 Gbaud using the 8b/10b coding. Let us
> assume again 1300 nm lasers and installed
> 500 MHz*km MMF.
>
> The figures below (I hope will not get distorted
> during transmission) show the power levels of
> the three systems. For 1 GbE I assumed no ISI.
> For PAM-5 I assumed 400 meters link length
> (there is no ISI up to this distance, ISI= 0 dB).
> For 4-WDM at 3.125 Gbaud I assumed 300 meters link
> length and about 3 dB optical loss due to ISI.
>
> | |
> | |
> | | 1 GbE, 1.25 Gbaud, no ISI
> | |
> | |
>
> | | | | |
> | | | | |
> | | | | | PAM-5 4-WDM at 1.25 Gbaud
> | | | | | no ISI.
> | | | | |
>
> |**** ****|
> |**** ****|
> |**** ****| 8b/10b 4-WDM at 3.125 Gbaud
> |**** ****| 3 dB ISI loss
> |**** ****|
>
> ( the asterisks denote closening of the eye due
> to ISI)
>
> 2a) 1 GbE
>
> By definition, the optical power difference
> between the '0' and '1' levels in the 1 GbE
> case is 1:
>
> optical signal power = 10*log(1) = 0 dB
>
> optical SNR = 0 dB (reference)
>
> And we get a BER of 10^(-12).
>
> 2b) PAM-5 4-WDM at 1.25 Gbaud
>
> In the PAM-5 case, notice that using the
> "open fiber control" method that I described
> in a previous email, we get the same launched
> power per channel as in the 1 GbE Ethernet.
> However, in PAM-5 the power difference
> between levels is 0.25 (this is the well
> known 6 dB optical power penalty loss of
> the PAM-5 modulation). Let us now add
> the 6 dB electrical coding gain provided
> by the Viterbi decoding: the effective distance
> between levels doubles. Summarizing, in PAM-5
> 4-WDM at 1.25 Gbaud, the effective optical
> signal power difference between levels is:
>
> effec optical signal power diff =
> 10*log(2*0.25) = - 3 dB
>
> where the factor 2 inside the log comes
> from the coding gain.
>
> The noise power at the input of the
> receiver is the same as in 1 GbE because
> we use the same baud rate. Hence,
>
> effec optical SNR = - 3 dB
>
> This is not bad compared to the 0 dB of 1 GbE.
> Furthermore, notice that we could even bring
> back the SNR for PAM5 to 0 dB (the 1 GbE
> reference) if a new laser safety proposal
> to move the maximum safe power from -4 dBm
> to -1 dBm is accepted. See P. Kolesar et al
> presentation in the next March meeting.
>
> 2c) 8b/10b 4-WDM at 3.125 Gbaud
>
> In this case the eye is half closed due
> to ISI. Hence the optical power difference
> is:
>
> optical signal power = 10*log(0.5)
> = - 3 dB
>
> and the electrical noise power at the input
> of the receiver is larger, since the receiver
> needs more bandwidth:
>
> elec noise power = 10*log[(3.125/1.25)^2]
> = 8 dB
>
> Hence, the optical SNR is
>
> optical SNR = - 3 - 8/2 = -7 dB
>
> This case is clearly worse than the 1 GbE
> case in terms of optical SNR. And the
> Task Force considers that this system
> can achieve the needed 10^(-12) BER.
>
> Summarizing:
>
> PAM-5 4-WDM at 1.25 Gbaud has no
> ISI, a clear and wide open eye at
> the input of the receiver, an optical SNR
> only slightly less than the SNR of present
> 1 GbE transceivers and better than
> the proposed 8b/10b 4-WDM at 3.125G
> transceivers.
>
> Hence, it is a good choice if one
> wants to reach the required BER of
> 10^(-12).
>
> Note: the above analysis, based on simple
> static SNRs, is not a substitute for a
> much more careful analysis regarding
> the viabibility of any PAM-5 approach.
> However, a simple back-of-the-envelope
> analysis is very useful to make a quick
> comparison between different PAM-5 proposals
> in order to find out which PAM-5 architecture,
>
> a) one working under strong ISI conditions
> that close the optical eye completely; or
>
> b) another using a lower baud rate
> and working with minimal ISI so that
> the optical eye at the input of the
> receiver is widely open,
>
> which one has more chance to meet the required
> BER of 10^(-12), and discard accordingly the
> one that does not have a chance.
>
> Jaime
>
> Jaime E. Kardontchik
> Micro Linear
> San Jose, CA 95131
> email: kardontchik.jaime@xxxxxxxxxxx
------=_NextPart_000_003D_01BF82C0.79F36830
Content-Type: text/html; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
X-Sun-Content-Length: 31170
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD><TITLE>RE: PAM-5, what are your BERs ?</TITLE>
<META content=3D"text/html; charset=3Diso-8859-1" =
http-equiv=3DContent-Type>
<META content=3D"MSHTML 5.00.2314.1000" name=3DGENERATOR></HEAD>
<BODY>
<DIV><FONT color=3D#0000ff face=3DArial size=3D2><SPAN=20
class=3D468471819-29022000>Dave:</SPAN></FONT></DIV>
<DIV><FONT color=3D#0000ff face=3DArial size=3D2><SPAN=20
class=3D468471819-29022000></SPAN></FONT> </DIV>
<DIV><FONT color=3D#0000ff face=3DArial size=3D2><SPAN =
class=3D468471819-29022000>Are=20
you suggesting, that if the BER of GbE is 10^-12, then the BER of the =
10xGbE=20
should be less? For example, 10^-13 to maintain the expected =
throughput=20
gain from 10xGbE?</SPAN></FONT></DIV>
<DIV><FONT color=3D#0000ff face=3DArial size=3D2><SPAN=20
class=3D468471819-29022000></SPAN></FONT> </DIV>
<DIV><FONT color=3D#0000ff face=3DArial size=3D2><SPAN =
class=3D468471819-29022000><BR>
<P><FONT size=3D2>Regards,<BR><BR>Edward S. Chang<BR>NetWorth =
Technologies,=20
Inc.<BR>EChang@xxxxxxxxxxxxxxxx<BR>Tel: (610)292-2870<BR>Fax: =
(610)292-2872=20
</FONT></P></SPAN></FONT></DIV>
<BLOCKQUOTE>
<DIV align=3Dleft class=3DOutlookMessageHeader dir=3Dltr><FONT =
face=3DTahoma=20
size=3D2>-----Original Message-----<BR><B>From:</B>=20
owner-stds-802-3-hssg@xxxxxxxx =
[mailto:owner-stds-802-3-hssg@xxxxxxxx]<B>On=20
Behalf Of </B>David Martin<BR><B>Sent:</B> Tuesday, February 29, 2000 =
10:12=20
AM<BR><B>To:</B> stds-802-3-hssg@xxxxxxxx<BR><B>Subject:</B> RE: =
PAM-5, what=20
are your BERs ?<BR><BR></DIV></FONT><BR>
<P><FONT face=3DArial size=3D2>Jaime,</FONT> </P>
<P><FONT face=3DArial size=3D2>A clarification on backbone transport =
BER spec's.=20
For OC-48 systems, although BCR GR-253 spec'd</FONT> <BR><FONT =
face=3DArial=20
size=3D2>1E-10 for worst case end-of-life, the defacto customer =
requirement was=20
1E-12. Similarly, for OC-192</FONT> <BR><FONT face=3DArial =
size=3D2>systems,=20
although BCR GR-1377 spec'd 1E-12 for worst case end-of-life, the =
defacto=20
customer</FONT> <BR><FONT face=3DArial size=3D2>requirement has been =
1E-15. Note=20
that the current target proposed for the ITU defined Optical</FONT> =
<BR><FONT=20
face=3DArial size=3D2>Transport Network is also 1E-15.</FONT> </P>
<P><FONT face=3DArial size=3D2>...Dave</FONT> <BR><BR><FONT =
face=3DArial=20
size=3D2>David W. Martin</FONT> <BR><FONT face=3DArial size=3D2>Nortel =
Networks</FONT> <BR><FONT face=3DArial size=3D2>+1 613 =
765-2901 =20
</FONT><BR><FONT face=3DArial size=3D2>+1 613 763-2388 (fax)</FONT> =
<BR><A=20
href=3D"mailto:dwmartin@xxxxxxxxxxxxxxxxxx"><U><FONT color=3D#0000ff =
face=3DArial=20
size=3D2>dwmartin@xxxxxxxxxxxxxxxxxx</FONT></U></A> </P>
<P><FONT face=3DArial =
size=3D2>=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D</FONT> </P>
<UL>
<P><A name=3D_MailData><FONT face=3DArial size=3D2>-----Original=20
Message-----</FONT></A> <BR><B><FONT face=3DArial =
size=3D2>From: =20
Jaime Kardontchik [SMTP:kardontchik.jaime@xxxxxxxxxxx]</FONT></B>=20
<BR><B><FONT face=3DArial size=3D2>Sent: </FONT></B> =
<FONT face=3DArial=20
size=3D2>Monday, February 28, 2000 12:05 PM</FONT> <BR><B><FONT =
face=3DArial=20
size=3D2>To: </FONT></B> <FONT face=3DArial=20
size=3D2>stds-802-3-hssg@xxxxxxxx</FONT> <BR><B><FONT face=3DArial=20
=
size=3D2>Subject: </FONT></B> =
<FONT=20
face=3DArial size=3D2>Re: PAM-5, what are your BERs ?</FONT> =
</P><BR>
<P><FONT face=3DArial size=3D2>Hello Vivek,</FONT> </P>
<P><FONT face=3DArial size=3D2>A few months ago there was a =
discussion on the=20
Reflector where</FONT> <BR><FONT face=3DArial size=3D2>some members =
claimed that=20
the BERs in some optical backbones</FONT> <BR><FONT face=3DArial=20
size=3D2>reached easily the 10^(-15). The repply to these claims was =
that</FONT> <BR><FONT face=3DArial size=3D2>although this might be =
true, the=20
philosophy of the 802.3, that made</FONT> <BR><FONT face=3DArial =
size=3D2>its=20
standards so successful, was to design based on worst case</FONT> =
<BR><FONT=20
face=3DArial size=3D2>conditions. Therefore, the Ethernet optical =
networks were=20
specficied</FONT> <BR><FONT face=3DArial size=3D2>to a BER of only=20
10^(-12).</FONT> </P>
<P><FONT face=3DArial size=3D2>The same holds for the Ethernet =
Copper=20
links.</FONT> </P>
<P><FONT face=3DArial size=3D2>Although it might be true that some =
links of=20
100BASE-TX</FONT> <BR><FONT face=3DArial size=3D2>reach a BER of =
10^(-12), the=20
official BER adopted in the</FONT> <BR><FONT face=3DArial =
size=3D2>Standard was=20
only 2.5*10^(-10). And the same for the</FONT> <BR><FONT =
face=3DArial=20
size=3D2>1000BASE-T, where the Standard adopted to support a</FONT> =
<BR><FONT=20
face=3DArial size=3D2>BER of 10^(-10).</FONT> </P>
<P><FONT face=3DArial size=3D2>The reason for this higher BERs in =
the Copper=20
networks is</FONT> <BR><FONT face=3DArial size=3D2>ISI (Inter Symbol =
Interference). Or, in layman's language:</FONT> <BR><FONT =
face=3DArial=20
size=3D2>is the eye open or not at the input of the receiver =
?</FONT> </P>
<P><FONT face=3DArial size=3D2>Jaime</FONT> </P>
<P><FONT face=3DArial size=3D2>Jaime E> Kardontchik</FONT> =
<BR><FONT=20
face=3DArial size=3D2>Micro Linear</FONT> <BR><FONT face=3DArial =
size=3D2>San Jose,=20
CA 95131</FONT> </P><BR>
<P><FONT face=3DArial size=3D2>Vivek Telang wrote:</FONT> </P>
<P><FONT face=3DArial size=3D2>> Hello Jaime and 10G'ers,</FONT> =
<BR><FONT=20
face=3DArial size=3D2>></FONT> <BR><FONT face=3DArial =
size=3D2>> I had to jump=20
in to try and correct this misconception of the "eye opening=20
requirement". While it is true that the "open eye" is always a good =
thing,=20
it is definitely not a requirement for low BER reception. If it were =
a=20
requirement, the tried and true 100BaseTX Ethernet would not be =
functioning=20
today. Many existing implementations of 100BaseTX silicon easily =
achieve=20
BERs of 1e-12 over 170+ meters of twisted pair cable, and the eye is =
very=20
much closed in this case. Not to mention that near-end crosstalk =
(NEXT)=20
makes matters significantly worse. Also, in 1000BaseT, if the =
only=20
impairment was ISI, and Echo, NEXT and FEXT were absent, BERs of =
1e-12 could=20
be easily achieved, even with a 5-level symbol =
constellation.</FONT></P>
<P><FONT face=3DArial size=3D2>></FONT> <BR><FONT face=3DArial =
size=3D2>> An=20
open eye is a requirement for systems that use simple lock-and-slice =
techniques, where little or no processing of the signal is done, =
other than=20
maybe amplification, and a PLL locks to the data clock and slices =
the data.=20
In contrast, in a DSP-based system, the combination of linear =
feedforward=20
equalizers and DFEs using simple LMS adaptation can easily deal with =
"closed=20
eyes". Even with a blind start-up.</FONT></P>
<P><FONT face=3DArial size=3D2>></FONT> <BR><FONT face=3DArial =
size=3D2>> The=20
implementation of a DSP-based system (including a 6-bit ADC) at 5GHz =
in CMOS=20
is a different matter altogether, and I am eagerly awaiting Oscar's=20
presentation at the March meeting.</FONT></P>
<P><FONT face=3DArial size=3D2>></FONT> <BR><FONT face=3DArial =
size=3D2>>=20
Regards,</FONT> <BR><FONT face=3DArial size=3D2>></FONT> =
<BR><FONT face=3DArial=20
size=3D2>> Vivek</FONT> <BR><FONT face=3DArial =
size=3D2>></FONT> <BR><FONT=20
face=3DArial size=3D2>> =
*---------------------------------------</FONT>=20
<BR><FONT face=3DArial size=3D2>> *</FONT> <BR><FONT face=3DArial =
size=3D2>> *=20
Vivek Telang</FONT> <BR><FONT face=3DArial size=3D2>> * Cicada =
Semiconductor=20
Inc.</FONT> <BR><FONT face=3DArial size=3D2>> * 901 MoPac =
Expressway=20
South</FONT> <BR><FONT face=3DArial size=3D2>> * Building One, =
Suite=20
540</FONT> <BR><FONT face=3DArial size=3D2>> * Austin, =
Texas =20
78746</FONT> <BR><FONT face=3DArial size=3D2>> *</FONT> <BR><FONT =
face=3DArial=20
size=3D2>> * 512-327-3500 x114 voice</FONT> <BR><FONT =
face=3DArial=20
size=3D2>> * 512-327-3550 =
fax</FONT>=20
<BR><FONT face=3DArial size=3D2>> * vivek@xxxxxxxxxxxxxxx</FONT> =
<BR><FONT=20
face=3DArial size=3D2>> * <A href=3D"http://www.cicada-semi.com"=20
target=3D_blank>http://www.cicada-semi.com</A></FONT> <BR><FONT =
face=3DArial=20
size=3D2>> *</FONT> <BR><FONT face=3DArial size=3D2>>=20
*---------------------------------------</FONT> <BR><FONT =
face=3DArial=20
size=3D2>></FONT> <BR><FONT face=3DArial size=3D2>> =
-----Original=20
Message-----</FONT> <BR><FONT face=3DArial size=3D2>> =
From: Jaime=20
Kardontchik [SMTP:kardontchik.jaime@xxxxxxxxxxx]</FONT> <BR><FONT =
face=3DArial=20
size=3D2>> Sent: Sunday, February 27, 2000 6:37 =
PM</FONT>=20
<BR><FONT face=3DArial size=3D2>> To: =20
stds-802-3-hssg@xxxxxxxx</FONT> <BR><FONT face=3DArial size=3D2>> =
Subject: PAM-5, what are =
your BERs=20
?</FONT> <BR><FONT face=3DArial size=3D2>></FONT> <BR><FONT =
face=3DArial=20
size=3D2>> Hello 10G'ers,</FONT> <BR><FONT face=3DArial =
size=3D2>></FONT>=20
<BR><FONT face=3DArial size=3D2>> Edward Chang was right on =
target.</FONT>=20
<BR><FONT face=3DArial size=3D2>> He wrote on Feb 23:</FONT> =
<BR><FONT=20
face=3DArial size=3D2>></FONT> <BR><FONT face=3DArial =
size=3D2>> > In the=20
past, the multiple voltage-level coding</FONT> <BR><FONT =
face=3DArial=20
size=3D2>> > was adopted by two LAN standards, ATM and</FONT> =
<BR><FONT=20
face=3DArial size=3D2>> > Ethernet. Both of them were =
twisted-pair</FONT>=20
<BR><FONT face=3DArial size=3D2>> > applications, and the BER =
were=20
10^(-10).</FONT> <BR><FONT face=3DArial size=3D2>> > I =
proposed BER of=20
10^(-12) in both working</FONT> <BR><FONT face=3DArial size=3D2>> =
> groups=20
to be consistent with the LAN optical</FONT> <BR><FONT face=3DArial=20
size=3D2>> > links' BER; however, for some reason, they</FONT> =
<BR><FONT=20
face=3DArial size=3D2>> > remain 10^(-10), officially in both=20
standards.</FONT> <BR><FONT face=3DArial size=3D2>></FONT> =
<BR><FONT=20
face=3DArial size=3D2>> What was the reason ?</FONT> <BR><FONT =
face=3DArial=20
size=3D2>></FONT> <BR><FONT face=3DArial size=3D2>> Multilevel =
voltage=20
coding was not the reason.</FONT> <BR><FONT face=3DArial =
size=3D2>> On the=20
contrary, it was a remedy. The real reason</FONT> <BR><FONT =
face=3DArial=20
size=3D2>> was the strong ISI at the maximum link lengths</FONT> =
<BR><FONT=20
face=3DArial size=3D2>> that these Standards wanted to =
support.</FONT>=20
<BR><FONT face=3DArial size=3D2>> Multi-level voltage coding was =
adopted=20
to</FONT> <BR><FONT face=3DArial size=3D2>> lower the baud rate =
or the=20
frequency content</FONT> <BR><FONT face=3DArial size=3D2>> of the =
signal in=20
order to make the ISI smaller</FONT> <BR><FONT face=3DArial =
size=3D2>> and=20
get a better BER.</FONT> <BR><FONT face=3DArial size=3D2>></FONT> =
<BR><FONT=20
face=3DArial size=3D2>> However, the remaining ISI in these =
Copper=20
links</FONT> <BR><FONT face=3DArial size=3D2>> remained high =
enough that the=20
eye at the</FONT> <BR><FONT face=3DArial size=3D2>> input of the =
receiver=20
remained still</FONT> <BR><FONT face=3DArial size=3D2>> =
completely closed. It=20
is this closed eye that</FONT> <BR><FONT face=3DArial size=3D2>> =
limited the=20
achievable BER in the Copper media</FONT> <BR><FONT face=3DArial =
size=3D2>>=20
to 10^(-10). Post-equalizers, no matter how</FONT> <BR><FONT =
face=3DArial=20
size=3D2>> sophisticated they were, are not able to</FONT> =
<BR><FONT=20
face=3DArial size=3D2>> completely reverse the effects of eye=20
closening</FONT> <BR><FONT face=3DArial size=3D2>> on the =
achievable=20
BER.</FONT> <BR><FONT face=3DArial size=3D2>></FONT> <BR><FONT =
face=3DArial=20
size=3D2>> ---> Conclusion ? In order to get =
a BER=20
of</FONT> <BR><FONT face=3DArial=20
size=3D2>> =
10^(-12) you=20
have to have a clear</FONT> <BR><FONT face=3DArial=20
size=3D2>> open =
eye at the=20
input of the receiver.</FONT> <BR><FONT face=3DArial=20
size=3D2>> The =
maximum=20
supported link length</FONT> <BR><FONT face=3DArial=20
size=3D2>> is set =
accordingly to meet this basic</FONT> <BR><FONT face=3DArial=20
size=3D2>> =
condition: open=20
eye.</FONT> <BR><FONT face=3DArial size=3D2>></FONT> <BR><FONT =
face=3DArial=20
size=3D2>> Now we can go to the basic two PAM-5 proposals:</FONT> =
<BR><FONT=20
face=3DArial size=3D2>></FONT> <BR><FONT face=3DArial =
size=3D2>> 1) PAM-5=20
serial at 5 Gbaud</FONT> <BR><FONT face=3DArial size=3D2>></FONT> =
<BR><FONT=20
face=3DArial size=3D2>> Using 1300 nm lasers lasers at 5 Gbaud =
the</FONT>=20
<BR><FONT face=3DArial size=3D2>> optical eye is completely =
closed already=20
at</FONT> <BR><FONT face=3DArial size=3D2>> ~170 meters due to =
heavy=20
ISI (installed MMF,</FONT> <BR><FONT face=3DArial =
size=3D2>> 500 MHz*km=20
bandwidth).</FONT> <BR><FONT face=3DArial size=3D2>></FONT> =
<BR><FONT=20
face=3DArial size=3D2>> Oscar claimed a support of 500 meters =
of</FONT>=20
<BR><FONT face=3DArial size=3D2>> installed MMF and pointed out =
that</FONT>=20
<BR><FONT face=3DArial size=3D2>> DFEs (equalizers) have been =
successfuly=20
used</FONT> <BR><FONT face=3DArial size=3D2>> in 100 and 1000 =
Mbps Copper=20
networks.</FONT> <BR><FONT face=3DArial size=3D2>> However, the =
specs and=20
experience of the 100 Mbps</FONT> <BR><FONT face=3DArial =
size=3D2>> and 1000=20
Mbps links over Copper - where the eye</FONT> <BR><FONT face=3DArial =
size=3D2>> at the input of the receiver is completely</FONT> =
<BR><FONT=20
face=3DArial size=3D2>> closed at the target link length due to =
ISI -</FONT>=20
<BR><FONT face=3DArial size=3D2>> put the achievable BERs around =
10^(-10)=20
only.</FONT> <BR><FONT face=3DArial size=3D2>> Even using =
sophisticated=20
equalizers.</FONT> <BR><FONT face=3DArial size=3D2>></FONT> =
<BR><FONT=20
face=3DArial size=3D2>> What is the experimental support that a =
BER</FONT>=20
<BR><FONT face=3DArial size=3D2>> of 10^(-12) could be achieved =
when a strong=20
ISI</FONT> <BR><FONT face=3DArial size=3D2>> closes completely =
the eye at the=20
input of the</FONT> <BR><FONT face=3DArial size=3D2>> receiver ? =
I think that=20
none and that the</FONT> <BR><FONT face=3DArial size=3D2>> =
experience points=20
to the contrary. Why then</FONT> <BR><FONT face=3DArial =
size=3D2>> could an=20
equalizer running at 5 GHz achieve</FONT> <BR><FONT face=3DArial =
size=3D2>>=20
here what sophisticated equalizers running</FONT> <BR><FONT =
face=3DArial=20
size=3D2>> 40 times slower (125 MHz) were not able to =
achieve</FONT>=20
<BR><FONT face=3DArial size=3D2>> in the Copper Ethernets ? =
Parallel=20
processing</FONT> <BR><FONT face=3DArial size=3D2>> could enable =
perhaps to=20
meet the timing</FONT> <BR><FONT face=3DArial size=3D2>> =
constraints of the=20
design by running multiple</FONT> <BR><FONT face=3DArial =
size=3D2>>=20
equalizers at a lower clock, but will not</FONT> <BR><FONT =
face=3DArial=20
size=3D2>> eliminate the basic limitation on achievable</FONT> =
<BR><FONT=20
face=3DArial size=3D2>> BERs once the eye at the input of the =
receiver</FONT>=20
<BR><FONT face=3DArial size=3D2>> is already completely =
closed.</FONT>=20
<BR><FONT face=3DArial size=3D2>></FONT> <BR><FONT face=3DArial =
size=3D2>> Let=20
us see now the case PAM-5 at 1.25 Gbaud:</FONT> <BR><FONT =
face=3DArial=20
size=3D2>></FONT> <BR><FONT face=3DArial size=3D2>> 2) PAM-5 =
4-WDM at 1.25=20
Gbaud</FONT> <BR><FONT face=3DArial size=3D2>></FONT> <BR><FONT =
face=3DArial=20
size=3D2>> Let us compare this approach to two other well</FONT> =
<BR><FONT=20
face=3DArial size=3D2>> known on-off approaches: 1 GbE and 4-WDM =
at</FONT>=20
<BR><FONT face=3DArial size=3D2>> 3.125 Gbaud using the 8b/10b =
coding. Let=20
us</FONT> <BR><FONT face=3DArial size=3D2>> assume again 1300 nm =
lasers and=20
installed</FONT> <BR><FONT face=3DArial size=3D2>> 500 MHz*km =
MMF.</FONT>=20
<BR><FONT face=3DArial size=3D2>></FONT> <BR><FONT face=3DArial =
size=3D2>> The=20
figures below (I hope will not get distorted</FONT> <BR><FONT =
face=3DArial=20
size=3D2>> during transmission) show the power levels of</FONT> =
<BR><FONT=20
face=3DArial size=3D2>> the three systems. For 1 GbE I assumed no =
ISI.</FONT>=20
<BR><FONT face=3DArial size=3D2>> For PAM-5 I assumed 400 meters =
link=20
length</FONT> <BR><FONT face=3DArial size=3D2>> (there is no ISI =
up to this=20
distance, ISI=3D 0 dB).</FONT> <BR><FONT face=3DArial size=3D2>> =
For 4-WDM at=20
3.125 Gbaud I assumed 300 meters link</FONT> <BR><FONT face=3DArial=20
size=3D2>> length and about 3 dB optical loss due to ISI.</FONT> =
<BR><FONT=20
face=3DArial size=3D2>></FONT> <BR><FONT face=3DArial =
size=3D2>>=20
=
| =
=20
|</FONT> <BR><FONT face=3DArial size=3D2>>=20
=
| =
=20
|</FONT> <BR><FONT face=3DArial size=3D2>>=20
=
| =
=20
| 1 GbE, 1.25 Gbaud, no =
ISI</FONT>=20
<BR><FONT face=3DArial size=3D2>>=20
=
| =
=20
|</FONT> <BR><FONT face=3DArial size=3D2>>=20
=
| =
=20
|</FONT> <BR><FONT face=3DArial size=3D2>></FONT> <BR><FONT =
face=3DArial=20
size=3D2>> | | | =
| |</FONT>=20
<BR><FONT face=3DArial size=3D2>> | | =
| =20
| |</FONT> <BR><FONT face=3DArial size=3D2>> =
| =20
| | | =20
| PAM-5 4-WDM at 1.25 =
Gbaud</FONT>=20
<BR><FONT face=3DArial size=3D2>> | | =
| =20
| | no ISI.</FONT> =
<BR><FONT=20
face=3DArial size=3D2>> | | | =
| |</FONT> <BR><FONT face=3DArial size=3D2>></FONT> =
<BR><FONT=20
face=3DArial size=3D2>> |**** =
****|</FONT> <BR><FONT face=3DArial size=3D2>>=20
|**** ****|</FONT> <BR><FONT =
face=3DArial=20
size=3D2>> |**** =20
****| 8b/10b 4-WDM at 3.125 =
Gbaud</FONT>=20
<BR><FONT face=3DArial size=3D2>> =
|**** =20
****| 3 dB ISI loss</FONT> =
<BR><FONT=20
face=3DArial size=3D2>> |**** =
****|</FONT> <BR><FONT face=3DArial size=3D2>></FONT> <BR><FONT =
face=3DArial=20
size=3D2>> ( the asterisks denote closening of the eye due</FONT> =
<BR><FONT=20
face=3DArial size=3D2>> to ISI)</FONT> <BR><FONT face=3DArial=20
size=3D2>></FONT> <BR><FONT face=3DArial size=3D2>> 2a) 1 =
GbE</FONT>=20
<BR><FONT face=3DArial size=3D2>></FONT> <BR><FONT face=3DArial =
size=3D2>> By=20
definition, the optical power difference</FONT> <BR><FONT =
face=3DArial=20
size=3D2>> between the '0' and '1' levels in the 1 GbE</FONT> =
<BR><FONT=20
face=3DArial size=3D2>> case is 1:</FONT> <BR><FONT face=3DArial=20
size=3D2>></FONT> <BR><FONT face=3DArial =
size=3D2>> =20
optical signal power =3D 10*log(1) =3D 0 dB</FONT> <BR><FONT =
face=3DArial=20
size=3D2>></FONT> <BR><FONT face=3DArial =
size=3D2>> =20
optical SNR =3D 0 dB (reference)</FONT> <BR><FONT face=3DArial=20
size=3D2>></FONT> <BR><FONT face=3DArial size=3D2>> And we get =
a BER of=20
10^(-12).</FONT> <BR><FONT face=3DArial size=3D2>></FONT> =
<BR><FONT=20
face=3DArial size=3D2>> 2b) PAM-5 4-WDM at 1.25 Gbaud</FONT> =
<BR><FONT=20
face=3DArial size=3D2>></FONT> <BR><FONT face=3DArial =
size=3D2>> In the PAM-5=20
case, notice that using the</FONT> <BR><FONT face=3DArial =
size=3D2>> "open=20
fiber control" method that I described</FONT> <BR><FONT face=3DArial =
size=3D2>> in a previous email, we get the same launched</FONT> =
<BR><FONT=20
face=3DArial size=3D2>> power per channel as in the 1 GbE =
Ethernet.</FONT>=20
<BR><FONT face=3DArial size=3D2>> However, in PAM-5 the power=20
difference</FONT> <BR><FONT face=3DArial size=3D2>> between =
levels is 0.25=20
(this is the well</FONT> <BR><FONT face=3DArial size=3D2>> known =
6 dB optical=20
power penalty loss of</FONT> <BR><FONT face=3DArial size=3D2>> =
the PAM-5=20
modulation). Let us now add</FONT> <BR><FONT face=3DArial =
size=3D2>> the 6 dB=20
electrical coding gain provided</FONT> <BR><FONT face=3DArial =
size=3D2>> by=20
the Viterbi decoding: the effective distance</FONT> <BR><FONT =
face=3DArial=20
size=3D2>> between levels doubles. Summarizing, in PAM-5</FONT> =
<BR><FONT=20
face=3DArial size=3D2>> 4-WDM at 1.25 Gbaud, the effective =
optical</FONT>=20
<BR><FONT face=3DArial size=3D2>> signal power difference between =
levels=20
is:</FONT> <BR><FONT face=3DArial size=3D2>></FONT> <BR><FONT =
face=3DArial=20
size=3D2>> effec optical signal power diff =
=3D</FONT>=20
<BR><FONT face=3DArial=20
=
size=3D2>> =
=20
10*log(2*0.25) =3D - 3 dB</FONT> <BR><FONT face=3DArial =
size=3D2>></FONT>=20
<BR><FONT face=3DArial size=3D2>> where the factor 2 inside the =
log=20
comes</FONT> <BR><FONT face=3DArial size=3D2>> from the coding =
gain.</FONT>=20
<BR><FONT face=3DArial size=3D2>></FONT> <BR><FONT face=3DArial =
size=3D2>> The=20
noise power at the input of the</FONT> <BR><FONT face=3DArial =
size=3D2>>=20
receiver is the same as in 1 GbE because</FONT> <BR><FONT =
face=3DArial=20
size=3D2>> we use the same baud rate. Hence,</FONT> <BR><FONT =
face=3DArial=20
size=3D2>></FONT> <BR><FONT face=3DArial =
size=3D2>> effec=20
optical SNR =3D - 3 dB</FONT> <BR><FONT face=3DArial =
size=3D2>></FONT>=20
<BR><FONT face=3DArial size=3D2>> This is not bad compared to the =
0 dB of 1=20
GbE.</FONT> <BR><FONT face=3DArial size=3D2>> Furthermore, notice =
that we=20
could even bring</FONT> <BR><FONT face=3DArial size=3D2>> back =
the SNR for=20
PAM5 to 0 dB (the 1 GbE</FONT> <BR><FONT face=3DArial size=3D2>> =
reference)=20
if a new laser safety proposal</FONT> <BR><FONT face=3DArial =
size=3D2>> to=20
move the maximum safe power from -4 dBm</FONT> <BR><FONT =
face=3DArial=20
size=3D2>> to -1 dBm is accepted. See P. Kolesar et al</FONT> =
<BR><FONT=20
face=3DArial size=3D2>> presentation in the next March =
meeting.</FONT>=20
<BR><FONT face=3DArial size=3D2>></FONT> <BR><FONT face=3DArial =
size=3D2>> 2c)=20
8b/10b 4-WDM at 3.125 Gbaud</FONT> <BR><FONT face=3DArial =
size=3D2>></FONT>=20
<BR><FONT face=3DArial size=3D2>> In this case the eye is half =
closed=20
due</FONT> <BR><FONT face=3DArial size=3D2>> to ISI. Hence the =
optical power=20
difference</FONT> <BR><FONT face=3DArial size=3D2>> is:</FONT> =
<BR><FONT=20
face=3DArial size=3D2>></FONT> <BR><FONT face=3DArial=20
size=3D2>> optical signal power =3D =
10*log(0.5)</FONT>=20
<BR><FONT face=3DArial=20
=
size=3D2>> =
=20
=3D - 3 dB</FONT> <BR><FONT face=3DArial size=3D2>></FONT> =
<BR><FONT face=3DArial=20
size=3D2>> and the electrical noise power at the input</FONT> =
<BR><FONT=20
face=3DArial size=3D2>> of the receiver is larger, since the =
receiver</FONT>=20
<BR><FONT face=3DArial size=3D2>> needs more bandwidth:</FONT> =
<BR><FONT=20
face=3DArial size=3D2>></FONT> <BR><FONT face=3DArial=20
size=3D2>> elec noise power =3D=20
10*log[(3.125/1.25)^2]</FONT> <BR><FONT face=3DArial=20
=
size=3D2>> =
=20
=3D 8 dB</FONT> <BR><FONT face=3DArial size=3D2>></FONT> =
<BR><FONT face=3DArial=20
size=3D2>> Hence, the optical SNR is</FONT> <BR><FONT =
face=3DArial=20
size=3D2>></FONT> <BR><FONT face=3DArial =
size=3D2>> =20
optical SNR =3D - 3 - 8/2 =3D -7 dB</FONT> <BR><FONT face=3DArial=20
size=3D2>></FONT> <BR><FONT face=3DArial size=3D2>> This case =
is clearly=20
worse than the 1 GbE</FONT> <BR><FONT face=3DArial size=3D2>> =
case in terms=20
of optical SNR. And the</FONT> <BR><FONT face=3DArial size=3D2>> =
Task Force=20
considers that this system</FONT> <BR><FONT face=3DArial =
size=3D2>> can=20
achieve the needed 10^(-12) BER.</FONT> <BR><FONT face=3DArial=20
size=3D2>></FONT> <BR><FONT face=3DArial size=3D2>> =
Summarizing:</FONT>=20
<BR><FONT face=3DArial size=3D2>></FONT> <BR><FONT face=3DArial=20
size=3D2>> PAM-5 =
4-WDM at=20
1.25 Gbaud has no</FONT> <BR><FONT face=3DArial=20
size=3D2>> ISI, a =
clear and=20
wide open eye at</FONT> <BR><FONT face=3DArial=20
size=3D2>> the =
input of the=20
receiver, an optical SNR</FONT> <BR><FONT face=3DArial=20
size=3D2>> only =
slightly=20
less than the SNR of present</FONT> <BR><FONT face=3DArial=20
size=3D2>> 1 GbE=20
transceivers and better than</FONT> <BR><FONT face=3DArial=20
size=3D2>> the =
proposed=20
8b/10b 4-WDM at 3.125G</FONT> <BR><FONT face=3DArial=20
size=3D2>> =20
transceivers.</FONT> <BR><FONT face=3DArial size=3D2>></FONT> =
<BR><FONT=20
face=3DArial =
size=3D2>> =20
Hence, it is a good choice if one</FONT> <BR><FONT face=3DArial=20
size=3D2>> wants =
to reach=20
the required BER of</FONT> <BR><FONT face=3DArial=20
size=3D2>> =
10^(-12).</FONT>=20
<BR><FONT face=3DArial size=3D2>></FONT> <BR><FONT face=3DArial =
size=3D2>>=20
Note: the above analysis, based on simple</FONT> <BR><FONT =
face=3DArial=20
size=3D2>> static SNRs, is not a =
substitute=20
for a</FONT> <BR><FONT face=3DArial size=3D2>> much more careful =
analysis=20
regarding</FONT> <BR><FONT face=3DArial size=3D2>> the =
viabibility of any=20
PAM-5 approach.</FONT> <BR><FONT face=3DArial size=3D2>> However, =
a simple=20
back-of-the-envelope</FONT> <BR><FONT face=3DArial size=3D2>> =
analysis is=20
very useful to make a quick</FONT> <BR><FONT face=3DArial =
size=3D2>>=20
comparison between different PAM-5 proposals</FONT> <BR><FONT =
face=3DArial=20
size=3D2>> in order to find out which PAM-5 architecture,</FONT> =
<BR><FONT=20
face=3DArial size=3D2>></FONT> <BR><FONT face=3DArial=20
size=3D2>> a) one working under strong ISI=20
conditions</FONT> <BR><FONT face=3DArial=20
size=3D2>> that close the =
optical eye=20
completely; or</FONT> <BR><FONT face=3DArial size=3D2>></FONT> =
<BR><FONT=20
face=3DArial size=3D2>> b) another using a =
lower baud=20
rate</FONT> <BR><FONT face=3DArial=20
size=3D2>> and working with =
minimal ISI=20
so that</FONT> <BR><FONT face=3DArial=20
size=3D2>> the optical eye at =
the input=20
of the</FONT> <BR><FONT face=3DArial=20
size=3D2>> receiver is widely =
open,</FONT> <BR><FONT face=3DArial size=3D2>></FONT> <BR><FONT =
face=3DArial=20
size=3D2>> which one has more chance to meet the required</FONT> =
<BR><FONT=20
face=3DArial size=3D2>> BER of 10^(-12), and discard accordingly =
the</FONT>=20
<BR><FONT face=3DArial size=3D2>> one that does not have a =
chance.</FONT>=20
<BR><FONT face=3DArial size=3D2>></FONT> <BR><FONT face=3DArial =
size=3D2>>=20
Jaime</FONT> <BR><FONT face=3DArial size=3D2>></FONT> <BR><FONT =
face=3DArial=20
size=3D2>> Jaime E. Kardontchik</FONT> <BR><FONT face=3DArial =
size=3D2>>=20
Micro Linear</FONT> <BR><FONT face=3DArial size=3D2>> San Jose, =
CA=20
95131</FONT> <BR><FONT face=3DArial size=3D2>> email:=20
kardontchik.jaime@xxxxxxxxxxx</FONT> =
<BR></P></UL></BLOCKQUOTE></BODY></HTML>
------=_NextPart_000_003D_01BF82C0.79F36830--