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

[802SEC] Material supporting starting 802.15.3c Sponsor Ballot



Colleagues-

Here is material in advance of our Friday EC meeting supporting the agenda item requesting approval to start Sponsor Ballot on 15.3c D06 plus MEC edits. For your convenience, text of the first link is reproduced below. The remaining two links are for the unedited unsatisfied comment file and for the Coexistence Assurance Document.

https://mentor.ieee.org/802.15/dcn/09/15-09-0231-00-0000-3c-sponsor-ballot-submission-summary.doc
https://mentor.ieee.org/802.15/dcn/09/15-09-0224-00-003c-unsatisfied-comments-list.xls
https://mentor.ieee.org/802.15/dcn/09/15-09-0022-09-003c-coexistence-assurance.pdf


Letter Ballot Summary:
Total Voters= 168
Initial1st2nd3rd
Yes 99(78.6%)100(78.1%)100(83.3%)112 (84.2%)
No27282221
Abstain6766

·   There have been no new NO comments or Voters on the last two recirculations
· A total of 96 comments remain unresolved, 48 rejected, 48 accepted or accepted in principal
·   All comments and responses have been recirculated at least once if not more
· Only 4 out of the 21 NO voters attended any 802.15 meetings during the prior 2 sessions
·   Individual Letters have been sent to each NO voter and 3 responses received

Status of the Draft:

Version D06 was balloted in the third recirculation with no new no votes and new comments associated with a no vote. Two voters restated their objections from previous ballots, carrying forward their previous comments (two comments from one no voter and three from the other no voter).

Draft was submitted for Mandatory Editorial Review in October 2008. Results were received on February 23, 2009 which required editorial changes to the draft prior to Sponsor ballot so that it complies with the IEEE 2009 Style Guide (e.g., correcting the copyright statements, putting in the correct disclaimer language).

The editor posted a revised version of the draft D06 with these required editorial changes applied on March 4 and announced its availability to the working group. Also posted were a comparison showing the changes and a summary of changes generated by the comparison process.

A summary of the editorial changes is posted at:
<https://mentor.ieee.org/802.15/dcn/09/15-09-0132-00-003c-d06-d07-mandatory-editorial-coordination-changes.ppt>https://mentor.ieee.org/802.15/dcn/09/15-09-0132-00-003c-d06-d07-mandatory-editorial-coordination-changes.ppt
D06 with MEC edits applied would be submitted to Sponsor Ballot.

Status Coexistence Assurance Document:

CA Doc has been reviewed by 802.19 and all suggested comments have been applied. A copy can found at <https://mentor.ieee.org/802.15/dcn/09/15-09-0022-09-003c-coexistence-assurance.pdf>https://mentor.ieee.org/802.15/dcn/09/15-09-0022-09-003c-coexistence-assurance.pdf

Comment Resolution Response Summary:

For those who are curious the following summarizes the rejected comments and reasons for the rejections. The full file can be found at <https://mentor.ieee.org/802.15/dcn/09/15-09-0224-00-003c-unsatisfied-comments-list.xls>https://mentor.ieee.org/802.15/dcn/09/15-09-0224-00-003c-unsatisfied-comments-list.xls. The 48 comments fell into 11 categories, which were more philosophical in character such as "customized for each market segment", "too many versus normal", "expectation level difference of realization" and such.

Summary by technical category (11):

1.       2 OFDM PHY mode (Number of comments:15, number of no voters: 15)

Essence of comments: It doesn't make sense to have 2 complete OFDM PHY modes that are so similar.

Reasons of rejection: Different PHYs are a result of demands of different market segments, which are stated in the usage models. Usage model 1 requires uncompressed video streaming. Due to the nature of uncompressed video signals a special PHY (AV PHY) is selected to provide high throughput. Usage model 4 requires a Conference ad hoc system in which all devices inside the WPAN will have bidirectional, NLOS high speed, low latency communication which is answered by HSI PHY.. Although AV and HSI OFDM PHYs share the same modulation technique, their frame design and approach to communication is very different..

2.      Common Mode (Number of comments:13, number of no voters:13)

Essence of comments: In order to promote coexistence it is not sufficient that a multi-mode PNC transmit beacons in the CR in addition to it's preferred mode. In order to allow efficient sharing of the spectrum, a multi-mode PNC must also be capable of allocating channel time to DEVs of different modes so that piconets of different modes can operate simultaneously on the same channel. Same-channel multi-mode operation is important because in many regulatory domains there will be only three channels available and some DEVs will only be able to operate on one of these channels (since that is all that some PHYs require).

Reasons of rejection: The SC and HSI PHYs are interoperable since they are both sending CMS beacons for coexistence and interoperability. For other cases, the CMS sync frame shall be used for interference mitigation. Therefore no change is required.

3.      Number of PHY mode (Number of comments: 7, number of no voters: 7)

Essence of comments: Three PHYs are two too many. No evidence was presented that the performance of any of these PHYs is significantly better than any other so the two additional PHYs simply makes coexistence and interoperability more difficult for no good reason.

Reasons of rejection: Different PHYs are a result of demands of different market segments, which are stated in the usage models. For example usage model 5 is for Kiosk applications. It demands 1.5 Gbps in 1 m range. SC-PHY can provide such a data rate in such a short range with less complexity thus lower cost than an OFDM PHY. Whereas usage model 1 requires uncompressed video streaming. Due to the nature of uncompressed video signals a special PHY (AV PHY) is selected to provide high throughput. Usage model 4 requires a Conference ad hoc system in which all devices inside the WPAN will have bidirectional, NLOS high speed, low latency communication which is answered by HSI PHY. Mandatory data rates of all those PHYs are selected according to specific usage models. As with any other standard we want to give implementers more options than the mandatory ones, therefore for SC PHY higher data rates are introduced which require equalization. However it is not appropriate to compare optional features of SC PHY with mandatory features of any OFDM PHY. Although AV and HSI OFDM PHYs share the same modulation technique, their frame design and approach to communication is very different.

4.      Remove DAMI/OOK (Number of comments:3, number of no voters: 3)

Essence of comments: DAMI and OOK are inefficient.

Reasons of rejection: The OOK/DAMI modes are optional and designed for low complexity and low power consumption. The OOK mode can adopt the simplest envelope detection for low complexity, power consumption and cost implementation. DAMI is for high data rate applications, and offers low power consumption and complexity and several implementers are shown interest.

5.      Unified aggregation (Number of comments:3 number of no voters: 2)

Essence of comments: Having multiple aggregaion schemes increases complexity in implementation and makes the spec unecessarily complicated

Reasons of rejection: Different aggregation schemes are result of contradicting requirements of different applications. Therefore group created 3 different aggregation schemes.

6.      Signal measurements (Number of comments:2, number of no voters: 2)

Essence of comments: At what point shall the ACI and AACI be measured (e.g., at the antenna connector)?

Reasons of rejection: To address comment on the lack of measurement point, the following sentence has been added to 12.1.2 RF power measurements in draft D07:"For DEVs without an antenna connector, the measurements shall be interpreted as EIRP and any radiated measurements shall be corrected to compensate for the antenna gain in the implementation."

7.      Skewed constellation (Number of comments:1, number of no voters: 1)

Essence of comment: Skewed constellations increase implementation complexity without providing any benefit over coding schemes for unequal error protection. Since we already have an effective UEP mode using coding why should we include this mode?

Reasons of rejection: The skewed constellation as an option provides different advantages as compared with coding schemes.

8.      Non-PHY specific MAC (Number of comments:1, number of no voters: 1)

Essence of comment: The spec draft defines the MAC features dependent on the PHY. It makes each MAC PHY dependent, and supporting separate set of features. As a result there is no MAC compatibility and the spec effectively becomes set of few separate specs. Such an approach complicates the spec and more important significantly increases complexity of implementation, verification, and certification. There is no limitation seen to unify the MAC features.

Reasons of rejection: There have been efforts trying to unify the MAC functionality as much as possible. In D07, sections 7, 8 and 13 are common to all the PHYs. There are some features specified for particular PHYs, such as AV aggregation. This is a common and reasonable approach which can be also seen in other 802 standards such as 802.16.

9.      AV PHY SIFS time (Number of comments:1, number of no voters: 1)

Essence of comment: A SIFS time of 2us corresponds to over 5000 samples at the nominal sampling rate. Such a large SIFS adds a very high per-frame overhead and results in very low throughput for applications that require transmission of short frames (such as wireless equivalents of wired peripheral buses). Note that the SIFS duration for the 802.11 OFDM PHY is only 200-320 samples. Achieving such low turn-around times is technically challenging but will be necessary if the specification is to support applications other than simple streaming and file transfer.

Reasons of rejection: The SC and HSI PHYs are interoperable since they are both sending CMS beacons for coexistence and interoperability. For other cases, the CMS sync frame shall be use for interference mitigation. Therefore no change is required.

10.   Mandatory beam forming (Number of comments:1, number of no voters: 1)

Essence of comment: 60GHz WPAN has to do beam forming in order to fulfil the link budget. Why is beam forming optional? - Specify a mandatory beam forming scheme.

Reasons of rejection: Very simple, low cost devices will use only a directional antenna and will rely on the user to point the device in the direction of the other device.

11.    SC PHY Preamble (Number of comments:1, number of no voters: 1)

Essence of comment: Two different CES lengths seems awkward, and the number of header rates can also be reduced. SFD2 can also be used for deliminator to gain frame timing reliability.

Reasons of rejection: In SCPHY, long CES and short CES are kept to increase performance or efficiency if necessary.





Bob Heile, Ph.D
Chairman, ZigBee Alliance
Chair, IEEE 802.15 Working Group on Wireless Personal Area Networks
11 Robert Toner Blvd
Suite 5-301
North Attleboro, MA  02763   USA
Mobile: +1-781-929-4832
email:   bheile@ieee.org

----------
This email is sent from the 802 Executive Committee email reflector.  This list is maintained by Listserv.