stds-80220-requirements: Section 4.1.2 Spectral Efficiency
Hi Dan
Thanks for your comments on section 4.1.2 Spectral Efficiency.
I want to reiterate the joint contribution by Samir, myself, and others on this
section dated 10/16/03.
First, on the subject of sectors, Samir's text read:
"For the purpose of evaluation, proposals should be limited to at most three sectors
to maintain consistency with current deployment practices. For the purpose of evaluation,
proposals should be limited to at most three sectors to maintain consistency with
current deployment practices. "
while your text reads: "Sustained spectral efficiency is computed in a loaded multi-cellular,
three-sector per cell, network setting.... Rationale: The change reflects recent comments and
requests to change the definition to bps/Hz/cell and the need to baseline all proposals on
a three-sector cell configuration..."
I disagree that we should limit ourselves to evaluating only three sectored configurations.
The reason is simply that an omni configuration is a valid deployment model and proposals
should therefore be free to present results in either omni or sectorized deployment modes.
For example, the PHS networks in Japan use omni deployments and are arguably the most
successful wide area network as far as wireless Internet services over laptops and PDAs is
concerned.
Second:
You define spectral efficiency as:
"It is defined as the ratio of the expected aggregate throughput (taking out all
PHY/MAC overhead) to all users in a cell divided by the total spectrum deployed
in that cell."
while Samir's text reads:
"It is defined as the ratio of the expected aggregate throughput (taking out
all PHY/MAC overhead) to all users in an interior cell divided by the system
bandwidth."
My issue concerns the change from "by the system bandwidth" to "the total spectrum
deployed in that cell". The latter would imply that a frequency planned system
would not divide by the system bandwidth but rather by the spectrum in use in only
one cell. Such a system could therefore claim higher spectral efficiency which
I think is not your intention.
My suggestion is to therefore change "the total spectrum in that cell" to "the total
spectrum across the network".
Third:
Regarding the actual numbers, I think I agree with you that the numbers need to be
looked at. The 2 bits/sec/hz downlink and 1 bits/sec/hz uplink deviated from the PAR's
flat 1 bit/sec/Hz and were proposed somewhat arbitrarily and so I think we may need
more discussion rather than just multiplying these numbers by a factor of 3.
Let me know your thoughts.
Mike
On Thu, Oct 23, 2003 at 04:23:58PM -0400, Gal, Dan (Dan) wrote:
> The following are comments and text change proposals for incorporation in the next revision of IEEE 802.20 SRD (System Requirements Document) rev 8b:
>
> Regards,
>
> Dan Gal
> Lucent Technologies O
>
>
>
> ----------------------------------------------------------------------------------
>
> Section 4.1.2 - Spectral Efficiency
>
> Current text: "
>
> Editors Note: Michael Youssefmir to supply definition of expected aggregate throughput for Appendix B.
>
> Sustained spectral efficiency is computed in a loaded multi-cellular network setting. It is defined
>
> as the ratio of the expected aggregate throughput (taking out all PHY/MAC overhead) to all
>
> users in an interior cell divided by the system bandwidth. The sustained spectral efficiency
>
> calculation shall assume that users are distributed uniformly throughout the network and shall
>
> include a specification of the minimum expected data rate/user.
>
> [Downlink > 2 bps/Hz/sector]
>
> [Uplink >1 bps/Hz/sector]
>
>
>
> 1. New text:
>
> Sustained spectral efficiency is computed in a loaded multi-cellular, three-sector per cell, network setting. It is defined
>
> as the ratio of the expected aggregate throughput (taking out all PHY/MAC overhead) to all
>
> users in a cell divided by the total spectrum deployed in that cell. The sustained spectral efficiency
>
> calculation shall assume that users are distributed uniformly throughout the network and shall
>
> include a specification of the minimum expected data rate/user.
>
> Given the above definition, the minimum spectral efficiency requirements shall be:
>
> - Downlink: greater than 6 bps/Hz/cell
>
> - Uplink: greater than 3 bps/Hz/cell
>
> Rationale: The change reflects recent comments and requests to change the definition to bps/Hz/cell and the need to baseline all proposals on a three-sector cell configuration. Thus, the previous figures were factored by 3.
>