Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

Re: [STDS-802-3-EPOC] EPoC RF Spectrum Ad Hoc Agenda



Duane, 

 

You're making quite a lot of assumptions in the table below. 

 

First, I was not surprised that the motion on MDIO-based provisioning
failed. I do not feel that the set of PHY registers is where this
information really needs to be stored, once that it is already available at
the higher layers (OAM client, for example). Repeating this information
within the PHY seems to be at least creating a potential point of contention
between both locations (which is the right set of values in case of conflict
?)

 

Second, it is very unclear on how information gets to PLC in the first place
- do we send it via MDIO to some set of registers and then they are polled
with some frequency by PLC and delivered to the other side, or rather we
periodically send this information to PLC itself. Without knowing that,
taking decisions is a bit over the edge in my opinion. 

 

Third, the more I look at PLC, the more this model seems to be broken, where
we offload already existing information channel (OAM) towards PLC. PLC
should be only used to help find the main data channel and not sending
management data through. 

 

Fourth, our decisions on exclusion bands have been (at best) non-technical
so far. Looking at motion 14
(http://www.ieee802.org/3/bn/public/decisions/decisions.html), I wonder how
many motions of this type we need to adopt to really take any technical
decision. The discussions so far have been at best non-committal and that
make me wonder how you can really provide even a rough estimate of what
volume of information should be sent. 

 

Fifth, I do not like the idea of fixed size fields. Sooner or later, someone
is bound to find a use case where they need to send more data than the
allocated size that we found acceptable. I'd move that if we need to send
data, we use TLV-based mechanism already used successfully in OAM.
Type/Length information will not burn a hole in the PLC and will make it so
much more robust and future-proof. 

 

Finally, I do not understand where 10 times per second comes from. Is it
some reference to long-gone OAM frame rate limit? Since it is broadcast
information (as far as I understand), why not make that configurable and let
operator decide how quickly they want this information disseminated in the
network. I would assume you might want to cycle through this information as
quickly as possible to allow new stations join the network as quickly as
possible, and when all known stations are registered, throttle down. Also,
it is not that we will use PLC for subscriber data transfer anyway, so we do
not need to worry about resource sharing and QoS here. Send as quickly as
possible and that is all. 

 

I am not going to even get into the discussion on the PMD parameters - I do
not feel qualified enough to get involved into it.

 

Regards

 

Marek 

 

From: Duane Remein [mailto:Duane.Remein@xxxxxxxxxx] 
Sent: Monday, 28 January, 2013 9:07 PM
To: STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx
Subject: Re: [STDS-802-3-EPOC] EPoC RF Spectrum Ad Hoc Agenda

 

Steve,

I volunteered to quantify the quantity of data needed for the PHY-Link
Ad-Hoc. Part of that effort is to determine what was needed to define the
exclusion bands. Below is what I have so far for this item. I would suggest
we use this as a starting point. 

I was quite surprised that the motion to provision the exclusion bands via
the MDIO did not even garner a second during the meeting. There are only two
interfaces we have to choose from, MDIO and MII, and MII is not appropriate
for provisioning information. I would like to straw poll this during
tomorrows meeting.

 

Another Topic I think we need to settle on is how big is a sub-carrier group
(or whatever we decide to call it).  I believe there are 2-3 options on the
table; 8-12 sub-carriers and 40-50 sub-carriers (i.e., 1 MHz). I'm OK with
defining the Channel center frequency in steps of 1 MHz but have problems
with defining other parameters in this granularity. For example do we really
want a 1 MHz PHY-Link? While we could pick and choose granularity for each
item that needs to be considered I think there is something to be said for
consistency (i.e., use the same unit for everything RF that is less than a
channel). Just my 2 cents, please hold now while I done my flame proof suit
J.

 

Best Regards,

Duane

 


Area

Information to send

Description

Step

Min

Max

Bits

Times sent per Sec.

Data Rate (Kbps)


Global

TDD/FDD

 

 

0

1

1

10

0.01


Global

Power Mngmt

tbd

 

 

 

 

 

0


Global

Number of DS OFDM Channels

Assumed upto 10 DS RF Channels that can be bonded

 

1

10

4

10

0.04


Global

Cyclic Prefix length

0 to 8 us in 0.25 us steps.

0.25

0

8

5

10

0.05


Global

Windowing

Enumerated list, 5 values of Nt for windowing calculation, Alpha =
Nt/204.86e6

 

 

5

3

10

0.03


Global

DS Ch1 Center Freq

tbd to 1800 MHz in 1 MHz steps.

1

tbd

1800

11

10

0.11


Global

DS Ch2-8 Center Freq

as per Ch1

 

 

9

99

10

0.99


Global

Number of US OFDM Channels

Assume upto 4 US RF Channels that can be bonded

 

1

4

2

10

0.02


Global

US Ch1 Center Freq

tbd to 1800 MHz in 1 MHz steps.

1

tbd

1800

11

10

0.11


Global

US Ch2-4 Center Freq

as per Ch1

 

 

3

33

10

0.33


Global

 

 

 

 

 

 

 

 


 

Total Global

 

 

 

 

 

 

1.69


DS CH Desc

DS Exclusion Band Low

Distance in MHz from lower frequency limit of channel; 0 to 96MHz in 1 MHz
steps

1

0

96

7

10

0.07


DS CH Desc

DS Exclusion Band High

Distance in MHz from upper frequency limit of channel; 0 to 96MHz in 1 MHz
steps

1

0

96

7

10

0.07


DS CH Desc

DS Number of mid exclusion bands

0 to 8 (assumes up to 8 internal exclusion bands)

 

0

8

3

10

0.03


DS CH Desc

DS Mid Exclusion band 1 F0

Distance in MHz from lower frequency limit of channel; 0 to 192 MHz in 1 MHz
steps

1

1

192

8

10

0.08


DS CH Desc

DS Mid Exclusion band 1 Fx

Distance in MHz of exclusion band above and below exclusion band F0; 0 to
32MHz in 1 MHz steps

1

1

32

5

10

0.05


DS CH Desc

DS Mid Exclusion band 2-8

as per above

 

 

7

43

10

0.43


DS CH Desc

 

DS Exclusion band is described for this RF Channel only (assumes each
Channel has it's own PHY-Link)

 

 

 

 

 

 

 

FutureWei Technologies Inc.

duane.remein@xxxxxxxxxx

Director, Access R&D

919 418 4741

Raleigh, NC

 

From: Shellhammer, Steve [mailto:sshellha@xxxxxxxxxxxxxxxx] 
Sent: Monday, January 28, 2013 3:10 PM
To: STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx
Subject: [STDS-802-3-EPOC] EPoC RF Spectrum Ad Hoc Agenda

 

All,

 

               This is a reminder that we have an RF Spectrum Ad Hoc call
tomorrow at 11 AM Pacific Time (2 PM Eastern Time).

 

               We discovered in Phoenix that a number of people have not
been attending or following the RF Spectrum Ad Hoc, and that there are some
topics to discuss with those who have not be following the Ad Hoc.

 

               Unless someone has a presentation to give I suggest we select
a topic and discuss it.  We could also craft a straw poll for those on the
call and possibly others.  Those who had been participating in the Ad Hoc
had decided that the exclusion sub-bands are to be configured, but some
others in the Task Force seemed to have other ideas.  I suggest we discuss
the following topic tomorrow,

 

               EXCLUSION SUB-BANDS: CONFIGURED BY MDIO OR SETUP IN SOME
OTHER MANNER?

 

If anyone has a presentation they would like to make please notify me.

 

Agenda

.         Review IEEE-SA Patent Policy

o
https://development.standards.ieee.org/myproject/Public/mytools/mob/slideset
.pdf 

.         Attendance

.         Discuss how exclusion sub-bands are setup

.         Other

 

Steve

 

 

  _____  

<="" p=""> 

 

  _____  

<="" p=""> 


________________________________________________________________________

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