Re: [802.3_EPOC] EPoC Bandwidth Discussion
I agree that spectrum is not a substitute for RF bandwidth. There is a wide spectrum of definitions to the word spectrum, none relating to a particular bandwidth. Why do we not just use RF bandwidth and data bandwidth.
Sent from my iPad
On Aug 21, 2012, at 1:05 PM, "Volker Leisse" <volker.leisse@xxxxxxxxxxxxxxx<mailto:volker.leisse@xxxxxxxxxxxxxxx>> wrote:
Duane,
although I followed most of the discussions on this list very closely, I have not contributed to it, yet. However, now I would like to add my 2 cents to this discussion. Thanks for your very concise summary which I fully agree with. I am afraid, though, that I have to make an additional comment on the conclusion by adding another aspect.
I have lived with the mismatch of scientific correctness and market reality for quite some time. When I worked in the university as a researcher, my professor always required me to refer to bandwidth only in terms of a portion of the frequency spectrum despite the fact that this was in contradiction to the understanding most common within the audience.
On the physical/analog side, I want to refer to the difference between 'spectrum' and '(RF) bandwidth'. Particularly in Europe, we had to learn the hard way in recent months that 'spectrum' is a physical resource originally not related to any transmission technology. It is in many contexts a politically loaded term with many (also regulatory) implications. Therefore, I would strongly suggest that we do not refer to spectrum when we mean a portion of the electromagnetic resources that EPoC as a technology is using to perform its task. You certainly figured in the meantime, that I prefer a term like 'RF bandwidth' as has been suggested previously. This is correctly referring to a portion of the frequency spectrum between frequency x and y while at the same time removing any ambiguity by adding the prefix 'RF'. (Sorry, Jorge, for not being able to agree with you completely!)
On the data/digital side, I'm afraid we will not achieve to change the world which is using bandwidth synonymously for data rate (and I, definitely, don't think that it is worth the effort to change existing specs such as EPON for that). However, I don't have a good prefix for 'bandwidth' in this context. So, if you feel that the distinction between 'bandwidth' (aka as data rate) and 'RF bandwidth' is too small we should endeavor to go for 'data rate' at least in EPoC. And, by the way, I don't think that we automatically talk about 'channel capacity' if we refer to 'data rate'. Channel capacity is a characteristic of the transmission channel in terms of the amount of data it can maximally transport in a given time interval. 'Data rate' is more general as it is used in various protocol related meanings.
Consequently, I would propose to remove the definition of 'RF spectrum' from your list and use the same text for 'RF bandwidth'.
I hope that I didn't add to the confusion but rather to clarity.
Best regards,
Volker
--
-------------------------------------
Volker Leisse
Cable Europe Labs
Phone: +49 531 590 16 70
Mobile: +49 179 52 32 941
E-mail: volker.leisse@xxxxxxxxxxxxxxx<mailto:volker.leisse@xxxxxxxxxxxxxxx>
Am 21.08.2012 16:46, schrieb Duane Remein:
Given that I started this thread I guess I should add my two bits (pun intended :) ). Here’s what I gather from the discussion
1) We are writing an amendment and we shouldn’t try to change the entire history of Ethernet. So in existing spec. is what it is and we should strive to be consistent with that as much as possible without adding to the confusion.
2) We seem to agree that “bandwidth” by itself is confusing, due to the rightful or wrongful misappropriation by “the digital guys” (here I’ll have to plead guilty).
3) We seem to agree that “spectrum” is not confusing and is used to refer to a band of frequencies (either RF or light btw).
4) It seems that the term “data rate” is also not confusing and is well defined, or at least well understood, in the Ethernet Spec.
I would propose that for our clauses:
we restrict the use of “bandwidth” and only use it when prefaced by “RF” or “spectrum”,
we use the term “data rate” when talking about the number of information bits per unit time.
I suggest that Marek add the following terms to the definition file:
· RF Bandwidth – see RF Spectrum
· RF Spectrum – the radio frequency band, measured in Hertz (Hz) used to convey information via various modulation techniques.
· Channel width - a specific segment of contiguous RF Spectrum.
· Data Rate – the transfer rate of information bits, measured in bits per second (bps).
Best Regards,
Duane
FutureWei Technologies Inc.
duane.remein@xxxxxxxxxx<mailto:duane.remein@xxxxxxxxxx>
Director, Access R&D
919 418 4741
Raleigh, NC
From: Marek Hajduczenia [mailto:marek.hajduczenia@xxxxxxxxx]
Sent: Tuesday, August 21, 2012 2:16 AM
To: STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx<mailto:STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx>
Subject: Re: [802.3_EPOC] EPoC Bandwidth Discussion
Geoff,
I do not disagree with you, but the scope of this project is not to fix anything in 802.3, or change EPON for that matter. That would be a task for maintenance or revision projects. It so happens that revision is done. So while you’re right in that there is inconsistency in terminology, it is not EPoC task to fix it in EPON, thus we just have to live it with for the time being, making sure that what we add is free of such artifacts.
Marek
From: Geoff Thompson [mailto:thompson@xxxxxxxx]
Sent: Tuesday, August 21, 2012 06:06
To: Marek Hajduczenia
Cc: STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx<mailto:STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx>
Subject: Re: [802.3_EPOC] EPoC Bandwidth Discussion
Marek-
Matt is correct and you are incorrect.
We are writing an amendment to the 802.3 Ethernet standard not an amendment to EPON.
If EPON terminology has flaws with respect to the rest of the Ethernet standard then they should get fixed.
(This situation "could" be different. There is an oustanding DISAPPROVE vote on the current 802.3 revision project to split off the point-to-multipoint portions of the standard into a separate standard within the 802.3 family.)
Best regards,
Geoff
On 208//12 7:17 PM, Marek Hajduczenia wrote:
Matt,
You’re 101% correct. We will be developing an amendment for EPoC and not a new spec, so some of the discussion, while technically correct, might be moot
Marek
From: Matthew Schmitt [mailto:m.schmitt@xxxxxxxxxxxxx]
Sent: Tuesday, August 21, 2012 03:12
To: STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx<mailto:STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx>
Subject: Re: [802.3_EPOC] EPoC Bandwidth Discussion
Someone will correct me if I'm wrong, but if memory serves we won't be writing an EPoC spec; rather, we'll be generating amendments to the Ethernet spec. If so, rather than having to rewrite how a term is used throughout the entire existing spec (even if we could get away with it), might it be a good idea to use existing terms as they're already used?
Matt
On Aug 20, 2012, at 7:17 PM, "Jain, Rajeev" <rajeevj@xxxxxxxxxxxx<mailto:rajeevj@xxxxxxxxxxxx>> wrote:
Thanks Hal.
I dare say the “Bandwidth” for bit rate is a marketing term that has crept in over the years – salesmen in Best Buy can more easily convince the consumers they need “more bandwidth” to sell a higher speed router than “higher bit rate”. “Bandwidth” has a feel good consumerism sound to it :)
Some of us are also old enough to recall the similar confusion between baud rate and bit rate in the good old dial-up modem days. “Baud” was carried over from days of telegraph and it took a while with internet access to move from baud to bit rate – and then we got that all mixed up with bandwidth.
So EPoC may be a good chance to set the record straight and take the IEEE specs back to real engineering.
Either way I do not have a strong opinion but concur with Marek that we need something unambiguous.
Rajeev
From: Hal Roberts [mailto:Hal.Roberts@xxxxxxxxx]<mailto:[mailto:Hal.Roberts@xxxxxxxxx]>
Sent: Monday, August 20, 2012 6:06 PM
To: Jain, Rajeev; STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx<mailto:STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx>
Subject: RE: [802.3_EPOC] EPoC Bandwidth Discussion
I agree with Rajeev,
Seems one can go with the traditional approach where bandwidth refers to spectrum or adopt the more recent use of bandwidth for data rate.
I have to side with the traditional approach. The term “Band-Width” is pretty clear. There is a “band” (chunk of spectrum) and it has a width associated with it.
It is unfortunate that the digital folks appropriated bandwidth for data rate.
Just my opinion.
From: Jain, Rajeev [mailto:rajeevj@xxxxxxxxxxxx]<mailto:[mailto:rajeevj@xxxxxxxxxxxx]>
Sent: Monday, August 20, 2012 8:02 PM
To: STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx<mailto:STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx>
Subject: Re: [802.3_EPOC] EPoC Bandwidth Discussion
Jorge, Marek,
Sorry to come late to the party but my vote is for “spectrum bandwidth” in Hz when referring to the frequency bands in which we transmit and “data rate” or “bit rate” in bps when referring the rate at which bits are transmitted in that frequency band, assuming that is consistent with EPON. I have also found that bandwidth is used loosely and can be misconstrued when used in the context of digital data transmission – data/bit rate is explicit is in this context. Adding the units to the definition will in any case remove any ambiguity.
Cheers,
Rajeev
From: Salinger, Jorge [mailto:Jorge_Salinger@xxxxxxxxxxxxxxxxx]<mailto:[mailto:Jorge_Salinger@xxxxxxxxxxxxxxxxx]>
Sent: Monday, August 20, 2012 5:43 PM
To: STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx<mailto:STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx>
Subject: Re: [802.3_EPOC] EPoC Bandwidth Discussion
Sounds good. So, I think we are in agreement then. Let's use the term bandwidth to mean data capacity, and not use it to mean RF capacity. Instead, when referring to RF capacity we use the term spectrum.
Does that make sense?
Thanks!
Jorge
From: Marek Hajduczenia [mailto:marek.hajduczenia@xxxxxxxxx]<mailto:[mailto:marek.hajduczenia@xxxxxxxxx]>
Sent: Monday, August 20, 2012 08:27 PM
To: Salinger, Jorge; stds-802-3-epoc@xxxxxxxxxxxxxxxxx<mailto:stds-802-3-epoc@xxxxxxxxxxxxxxxxx> <stds-802-3-epoc@xxxxxxxxxxxxxxxxx<mailto:stds-802-3-epoc@xxxxxxxxxxxxxxxxx>>
Subject: RE: [802.3_EPOC] EPoC Bandwidth Discussion
Jorge,
In EPON, we used term “bandwidth” in the meaning of “data rate” / “channel capacity”, which in the retrospect perhaps was not the best idea. For example, we had statements like “Flexible architecture allowing dynamic allocation of bandwidth”, which does not speak of spectrum allocation, but rather channel capacity allocation. I could produce more examples where it is used in this context, but I think you get the idea where it is going.
In general, EPON we only spoke of wavelength / wavelength band allocation in Clause 75/60, which is effectively spectrum allocation. However, the term used was “wavelength” and ”wavelength band/range”.
Does this help ?
Marek
From: Salinger, Jorge [mailto:Jorge_Salinger@xxxxxxxxxxxxxxxxx]<mailto:[mailto:Jorge_Salinger@xxxxxxxxxxxxxxxxx]>
Sent: Tuesday, August 21, 2012 01:15
To: Marek Hajduczenia; stds-802-3-epoc@xxxxxxxxxxxxxxxxx<mailto:stds-802-3-epoc@xxxxxxxxxxxxxxxxx>
Subject: Re: [802.3_EPOC] EPoC Bandwidth Discussion
Marek,
I understand.
So, how is the term bandwidth used in EPON parlance? I looked down in the thread and can't see an example, but maybe I missed it.
Thanks!
Jorge
From: Marek Hajduczenia <marek.hajduczenia@xxxxxxxxx<mailto:marek.hajduczenia@xxxxxxxxx>>
Date: Monday, August 20, 2012 8:08 PM
To: "Salinger, Jorge" <Jorge_Salinger@xxxxxxxxxxxxxxxxx<mailto:Jorge_Salinger@xxxxxxxxxxxxxxxxx>>, EPoC Study Group <STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx<mailto:STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx>>
Subject: RE: [802.3_EPOC] EPoC Bandwidth Discussion
Jorge,
The only concern that I have with this is that multiple times we speak of “bandwidth” without other denominators, leaving it open to interpretation.
While I can certainly understand where Geoff would like us to go, if we set different terminology from what was used in EPON before, we will create a rift and people who understand EPON, will have to get used to a different terminology for EPoC, something that I’d rather (personally) avoid.
Marek
From: Salinger, Jorge [mailto:Jorge_Salinger@xxxxxxxxxxxxxxxxx]
Sent: Tuesday, August 21, 2012 01:04
To: STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx<mailto:STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx>
Subject: Re: [802.3_EPOC] EPoC Bandwidth Discussion
All,
No quite being an RF guy, but having worked in the cable industry for over 20 years… (for those thinking that I am old… I started when I was 19 years old ;-)
My 2 cents…
I would propose that when speaking of RF capacity we refer to it as spectrum. So, in the case of the discussion referenced by this Email thread, we say "spectrum for EPoC" or "EPoC spectrum".
I think that the term bandwidth can be ambiguous unless it is used in a specific context, such as "the bandwidth of an analog cable channel is 6 MHz" and I also hear "the bandwidth of the highest HSD tier is 300 Mbps".
Jorge
From: Duane Remein <Duane.Remein@xxxxxxxxxx<mailto:Duane.Remein@xxxxxxxxxx>>
Reply-To: Duane Remein <Duane.Remein@xxxxxxxxxx<mailto:Duane.Remein@xxxxxxxxxx>>
Date: Monday, August 20, 2012 1:15 PM
To: EPoC Study Group <STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx<mailto:STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx>>
Subject: Re: [802.3_EPOC] EPoC Bandwidth Discussion
Geoff,
I see your point, could we at least agree to preface there term with “Spectra” when referring to RF (old habits, the only kind I have at this point, die hard)?
Best Regards,
Duane
FutureWei Technologies Inc.
duane.remein@xxxxxxxxxx<mailto:duane.remein@xxxxxxxxxx>
Director, Access R&D
919 418 4741
Raleigh, NC
From: Geoff Thompson [mailto:thompson@xxxxxxxx]
Sent: Monday, August 20, 2012 1:04 PM
To: STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx<mailto:STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx>
Subject: Re: [802.3_EPOC] EPoC Bandwidth Discussion
Colleagues-
Just goes to show the extent to which Wikipedia is not an authoritative source.
The answer to this question is obvious within the word itself. Bandwidth is the width of the spectral band, i.e. it is an analog measurement.
The so called "digital bandwidth" is not a bandwidth it is "channel capacity" or "data rate" or "bit rate".
To illustrate how silly it is to use the spectral term for the digital term just think of it in concrete terms.
You would never use the term "pavement width" when you mean "cars per hour".
I took a brief tour of older clauses of 802.3 and the term is mostly used correctly. There are several instances (which are unfortunately duplicated in several places) where "bit rate" would have been more accurate.
I would strongly prefer that we preserve its analog meaning and use other appropriate terminology for digital rate measurement (even if it means bucking current common terminology).
Geoff Thompson
On 208//12 9:22 AM, Hesham ElBakoury wrote:
Jim,
Wikipedia defines bandwidth as follows:
Bandwidth has several related meanings:
* Bandwidth (signal processing)<http://en.wikipedia.org/wiki/Bandwidth_%28signal_processing%29> or analog bandwidth, frequency bandwidth or radio bandwidth: a measure of the width of a range of frequencies, measured in hertz
* Bandwidth (computing)<http://en.wikipedia.org/wiki/Bandwidth_%28computing%29> or digital bandwidth: a rate of data transfer, bit rate or throughput, measured in bits per second (bps
Hesham
From: Jim Farmer [mailto:jfarmer@xxxxxxxxxx]
Sent: Monday, August 20, 2012 8:50 AM
To: STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx<mailto:STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx>
Subject: Re: [802.3_EPOC] EPoC Bandwidth Discussion
Unfortunately, as if often the case, we have multiple different definitions for the same term – as an RF nerd myself, I had difficulty getting used to “bandwidth” referring to data. But the data usage certainly is a common use of the term. If we need to differentiate the two, I’d suggest the shorter term “RF bandwidth” when we need to differentiate the two.
Thanks,
jim
Jim Farmer, K4BSE
Chief System Architect,
FTTP Solutions
Aurora Networks
1220 Old Alpharetta Rd.
Ste. 370
Alpharetta, GA 30005 USA
678-339-1045 (office)
678-640-0860 (mobile)
jfarmer@xxxxxxxxxx<mailto:jfarmer@xxxxxxxxxx>
From: Duane Remein [mailto:Duane.Remein@xxxxxxxxxx]
Sent: Monday, August 20, 2012 11:32 AM
To: STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx<mailto:STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx>
Subject: Re: [802.3_EPOC] EPoC Bandwidth Discussion
Steve,
Your use ot the term “bandwidth” completely threw me, I’ve always associated this term with data bandwidth.
Can I suggest we use something like Spectrum Channel Width – defined as a portion of RF spectrum dedicated to a transmission channel or sub-channel.
Once we agree on such a term Marek can then add this to the definitions list.
Best Regards,
Duane
FutureWei Technologies Inc.
duane.remein@xxxxxxxxxx<mailto:duane.remein@xxxxxxxxxx>
Director, Access R&D
919 418 4741
Raleigh, NC
From: Shellhammer, Steve [mailto:sshellha@xxxxxxxxxxxx]
Sent: Friday, August 17, 2012 6:42 PM
To: STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx<mailto:STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx>
Subject: [802.3_EPOC] EPoC Bandwidth Discussion
EPoC Group,
Several of us had a good discussion on EPoC bandwidth this morning. I would like to see if there are other who would like to join us for future calls on Friday mornings (10 AM Pacific Time).
If anyone else would like to join us, please send me an email and I will add you to the meeting invite.
Once we have some slides put together we will review them on the Monday AM calls, hosted by Comcast.
Steve
________________________________
________________________________________________________________________
To unsubscribe from the STDS-802-3-EPOC list, click the following link:
https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-3-EPOC&A=1