| Thread Links | Date Links | ||||
|---|---|---|---|---|---|
| Thread Prev | Thread Next | Thread Index | Date Prev | Date Next | Date Index |
I'm sorry that I haven't been able to devote much time to this
yet, but Mike's comments below triggered something of an "old chestnut" for me,
which you might want to think about.
"Spectral efficiency" (E) to me is a
parameter related to the air interface in terms of its ability to convert
channel bandwidth into raw or useful throughput (e.g. bps/Hz etc). This is
independent of system or deployment parameters / scenarios but can be linked to
channel models and traffic models etc.
"Spectrum Utilisation Efficiency"
is a completely different measure of the efficiency with which a given base
station or wide area (multicellular) deployment converts its wide area spectrum
allocation (e.g. licence) into useful payload. The result is a combination of
intra-cell frequency reuse (R) (whether by FDMA, CDMA, TDMA, SDMA or various
combinations less any overheads imposed by diversity techniques such as STC,
MIMO etc) and inter-cell frequency reuse (N)( which depends on geographic
sectorisation and interference mitigation techniques). The overall frequency
reuse is then expressed as the ration R/N. So a spectral efficiency of 4
bps/Hz with an internal reuse of 6 (e.g. through adaptive arrays) and a sector
plan of N=3 would give a wide area spectrum utilisation efficiency of 4 * 6 / 3
= 8 Mbps per MHz of assigned spectrum.
For more information, look at
"Proposal 5" in http://grouper.ieee.org/groups/802/16/tg3/contrib/802163c-01_41.pdf
Regards
-----Original Message-----
From:
owner-stds-80220-requirements@majordomo.ieee.org
[mailto:owner-stds-80220-requirements@majordomo.ieee.org]On
Behalf Of
Michael Youssefmir
Sent: 23 July 2003 22:27
To: Joseph
Cleveland
Cc: stds-80220-requirements@ieee.org; Michael
Youssefmir
Subject: Re: stds-80220-requirements: 802.20 Requirements v5
-
C802.20-03-6921
Hi Joseph,
I have a couple
comments on this. The term sustained spectral
efficiency was first used in
the PAR and is defined in the
appendix. The intent I believe was to capture
the need to provide
high network capacity. I think at the requirements level
this is
captured well in the document.
I agree with you that
evaluating spectral efficiency is
dependent on many different factors. This
is more or less
captured in the current evaluation criteria document
that
talks about the need for the group to agree to or at least
present system
parameters when evaluating spectral efficiency.
In particluar the evaluation
criteria talks about measuring
spectral efficiency across load/coverage
operating points.
Given this, I think we are in fairly good position with
the text
in the current document.
Mike
Michael
Youssefmir
ArrayComm, Inc
On Tue, Jul 22, 2003 at
01:25:39PM -0500, Joseph Cleveland wrote:
> Hi All:
>
> I
applaud the effort to define a data rate requirement in Section 41. in
>
terms of a sustained average. However, a sustainable throughput is
a
> function of the peak burst rate, user environment (Ped A, Ped B,
etc.),
> propagation conditions, fading conditions, RF network design,
user
> distribution, etc. For a minimum throughput (i.e. sustained
data rate)
> requirement to have meaning, it is necessary to state
specifically all of
> these conditions - a very difficult
task.
>
> Isn't it more straightforward - and less controversial -
to specify a peak
> burst rate per sector and then state a maximum allowed
degradation in data
> rate (or Eb/Io penalty) for different degradation
factors?
> .
> Joseph Cleveland
> Director, Systems
& Standards
> Wireless Systems Lab
> Samsung Telecommunications
America
> Richardson, TX 75081
> (O) 972-761-7981 (M)
214-336-8446 (F) 972-761-7909
>