Copyright 1984-1999 FileMaker, Inc. HBAM2016AUG95HPro 5.0sg€KhDECISION OF GROUP @ B J U_isACE@ ACOAA M N J!I"A#N$O%I&"I''M(,I)1A*C+B,4A-8N.:N/?E0BA1FH2JB3LA4A5OA6PN7UA8XN9[O:^M;`AgN?iA@jOAmBoOCqFDsIEuAFwNGzAH}OI€HJ€EL€ JM€ ON€MO€NP€NQ€NR€#OS€'ETBU€)AV€,NW€-AX€/AY€0NZ€2A[€4O\€6A]A^€9N_€<A`€>Ha€@Ib€EFc€JJd€KAe@K!@.HBAM3016AUG95@i AAzHH(FG(HH(d}HzA('dD0,.$n &R-S/+@/.N$JPOg0HjHnYOUO?./<0??< /< ?/.?*HnN `UOBg/<0>g0*rg.p.? jf4UOHj0H2+HҀ0* HЁ F2(H=A?.?*JGg HjNXO0*rg,v`RC 2 0gp*2Hoz*0HS@Jjg$?*/ 0*r?HjNBgO `/ 0*r?HjNO HnBgJGg\.rgR jf=n`0+H F2(HҀ0* HQ4 1st Quarter 2nd Quarter 3rd Quarter 4th Quarter,  ,  X5RH53^(,RҸV0SunMonTueWedThuFriSatJanFebMarAprMayJunJulAugSepOctNovDec h<    7/18/2001@ BALLOTNUMBERCHANGE COMMENT DATE COMMENT NUMBER COMMENT TYPE COMMENTARY VERSIONCOUNTERDOCNUMBER EXPORTDATE FIRST NAME LAST NAMEMEMBERSHIP TYPEREASONREASON PROPOSEDRECOMMENDATION RESOLUTION BYRESOLUTION OF GROUPRESOLUTION PROPOSEDSECTION STARTING LINE NUMBER STARTING PAGE NUMBER  nK!ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]EPro 5.0F!2(M1,,./: AM PMYyTtNnFfNALL OTABSP^SundayMondayTuesday WednesdayThursdayFridaySaturdayJanuaryFebruaryMarchAprilMayJuneJulyAugust SeptemberOctoberNovemberDecemberQ1Q2Q3Q4 1st Quarter 2nd Quarter 3rd Quarter 4th Quarter,  , K ABCDEFGHI J K L M NOPQRSTUVWXYZ[\]    !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~  !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghi $,.YesNo/,  , : AM PM HelveticaGenevaArial  B GA ExportDateB GA BallotNumberB GA DocNumberB GA First NameB GA Last NameB GAMembership TypeB GAComment NumberB G A Comment DateB G A Comment TypeB G AStarting Page NumberB G AStarting Line NumberB G ASectionB GAChangeB GAReasonB GACounterB G !A1.1D802.16ab-01/01`yout #1A Layout #1B$$$$@:A  $,.YesNo/,  , : AM PMB   C  $,.YesNo/,  , : AM PMD   E  $,.YesNo/,  , : AM PMF   G  $,.YesNo/,  , : AM PMH   I  $,.YesNo/,  , : AM PM             DRACommentary VersionA Resolution byB GARecommendationB GAResolution ProposedB GAReason ProposedB GAResolution of GroupB GADecision of GroupB GAReason for Group DecisionB GA Group's NotesB GAGroup's Action ItemsB GAEditor's ActionsB GA Editor's NoteB GAEditor's QuestionB GAEditor's Action ItemsB G DECISION OF GROUP DOCNUMBEREDITOR'S ACTION ITEMSEDITOR'S ACTIONS EDITOR'S NOTEEDITOR'S QUESTION EXPORTDATE FIRST NAMEGROUP'S ACTION ITEMS GROUP'S NOTES LAST NAMEMEMBERSHIP TYPEREASONREASON FOR GROUP DECISIONREASON PROPOSEDRECOMMENDATION RESOLUTION BYRESOLUTION OF GROUPRESOLUTION PROPOSEDSECTION STARTING LINE NUMBER STARTING PAGE NUMBER  BCommentary VersionD ExportDateF BallotNumberH DocNumberJ First NameL Last NameNMembership TypePComment NumberR Comment DateT Comment TypeVStarting Page NumberXStarting Line NumberZSection\Change^Reason`Counterb Resolution bydRecommendationfResolution ProposedhReason ProposedjResolution of GrouplDecision of GroupnReason for Group Decisionp Group's NotesrGroup's Action ItemstEditor's Actionsv Editor's NotexEditor's QuestionzEditor's Action Items @BBAt  !"#$%&'()*+,-./0123456789tB%468888888 8 8 8 8 8 8 88888 8"8$8&8(88 8"8$8&8( J   K  $,.YesNo/,  , : AM PML   M  $,.YesNo/,  , : AM PMN   O  $,.YesNo/,  , : AM PMP   Q  $,.YesNo/,  , : AM PMR   S  $,.YesNo/,  , : AM PMT    U  $,.YesNo/,  , : AM PMV   W  $,.YesNo/,  , : AM PMX   Y  $,.YesNo/,  , : AM PMZ   [  $,.YesNo/,  , : AM PM\   ]  $,.YesNo/,  , : AM PM^    _  $,.YesNo/,  , : AM PM`   a  $,.YesNo/,  , : AM PMb   c  $,.YesNo/,  , : AM PMd   e  $,.YesNo/,  , : AM PMf   g  $,.YesNo/,  , : AM PMh    i  $,.YesNo/,  , : AM PMj   k  $,.YesNo/,  , : AM PMl   m  $,.YesNo/,  , : AM PMn   o  $,.YesNo/,  , : AM PMp   q  $,.YesNo/,  , : AM PMr    s  $,.YesNo/,  , : AM PMt   u  $,.YesNo/,  , : AM PMv   w  $,.YesNo/,  , : AM PMx   y  $,.YesNo/,  , : AM PMz   (B 2001/07/04EJin-MengFHoGOtherH001I 2001/06/29JTechnical, Non-bindingAThe binary exponential backoff procedure as implied in the 802.16 draft may need to be refined or redefined. CSMA backoff operates in a distributed control fashion, while 802.16 is under central control. Contending stations choose backoff parameters (including the minimum and maximum backoff windown sizes) on their own in a distributed environment, but take the initial backoff based on the Backoff Start value sent by the BS in a centralized environment. Thus, for 802.16, will retransmiBssions double their previous backoff windows and thus ignore the recent Backoff Start values, or use the latest Backoff Start value for the next backoff and hence circumvents the binary exponential procedure?@ Q8*8,8.8082848688 v        "$&(*,.02468888888 8 8 8 8 8 8 88888 8"8$8&8(88 8"8$8&8( E()*+!,-./301F23*45967# 8 9: T 2 C     TU$U#D&6(S5 8H!:6G"JZ#L&Y$\l%^k&n~'p#} A('*:,B9<L>8KN^PB]`p bEo Br tF  $ % 3 2 @@@A Layout #1A Layout #1B$$$$@:A  $,.YesNo/,  , : AM PMB   C  $,.YesNo/,  , : AM PMD   E  $,.YesNo/,  , : AM PMF   G  $,.YesNo/,  , : AM PMH   I  $,.YesNo/,  , : AM PMs@AIf an SS sends a bandwidth request message based on the Request Backoff Start value contained in the most recently received Uplink MAP message, and if the SS determines from the next Uplink MAP message that transmission to have failed, will the SS retransmit the same request by doubling the backoff window and hence ignoring the Request Backoff Start and Request Backoff End values contained in the latest Uplink MAP message, or by using the Request Backoff Start value appearing in the latestB Uplink MAP message? If retransmission is based on the Request Backoff Start and Request Backoff End values that were broadcast at the time the initial transmission was made, the BS basically loses its control over contending stations once they have started to send (contend). However, the following paragraph written for "Contention Resolution" on page 124 of the 802.16/D2-2001 draft seems to indicate otherwise? "The BS has much flexibility in controlling the contention resoluCtion. At one extreme, the BS may choose to set up the Data Backoff Start and End to emulate an Ethernet-style back-off with its associated simplicity and distributed nature, but also its fairness and efficiency issues. This would be done by setting Data Backoff Start = 0 and End = 10 in the UL-MAP. At the other end, the BS may make the Data Backoff Start and End identical and frequently update these values in the UL-MAP so all SS are using the same, and hopefully optimal, back-off window.Dt" Again, insightful feedback on this note--which is by not means intended to be a criticism of the related wo >P57B 2001/07/04EMikaFKasslinGMemberH002JTechnical, Non-bindingN5Change the ARQ mechanism described in 802.16abc-01/07P782B 2001/07/04ERogerFMarksGMemberH003I 2001-07-03J EditorialNAdd the following chapters from 802.16.3c-01/61: 1. "Bandwidth Request Using CDMA Codes in OFDMA(OFDM) Base PHY" 2. "Ranging Enhancement" ng the contention resoluOMost of IEEE 802.16.3c-01/61 was incorporated into 802.16ab-01/01. However, these two sections were left out, possibly due to an editorial oversight. P777B 2001/07/04EJin-MengFHoGOtherH004I 2001/06/29JTechnical, Non-bindingK1NOne more field distinguishing a cell from other cells need be added to the generic MAC header. Such a field may contain the BS address.AA single CID-type MAC address identification is adequate in DOCSIS as the application is carried on a closed environment, but a CID alone may not be sufficient for uniquely identifying a received packet in an open wireless environment. For example, if two or more adjacent cells happen to be operating in the same frequency band, how does a receiving station unambiguously determine the sending station of a received packet to be from the same cell, but not from another cell? P56B 2001/07/04EKenFStanwoodH005I 2001/07/02JTechnical, Non-bindingK1L1AThis document proposes 5+ PHYs (it's hard to be certain of the exact number), ignoring such issues as TDD vs. FDD and channel widths, and 3 ARQ algorithms to solve what appears to be 2 problems. Shouldn't there be a bit more agreement than this before things go further? The SC PHY section should lose all the verbage saying how its better than OFDM. If both are in the spec, they should be chosen by the implementor. If one is clearly better, get rid of the lesser PHY.B 2001/07/04EDavidFTrinkwonH006I 2001/07/03J EditorialK1L49MKeywordsNDelete ", millimeter waves"O"Not applicable to 2 - 11 GHz bandsB 2001/07/04EDavidFTrinkwonH007I 2001/07/03JTechnical, BindingK7L54M 8.3.9.1x & 2xN@Delete sections 8.3.9.1 and 8.3.9.2 (and in the body text areas)AWThese sections relate more to Recommended Practices for deployment purposes than to an air interface specification. Note : Phil Whitehead's PAR for extensions to the 802.16 RP should probably be expanded to include license exempt deployments, based on this material, unless this will be done jointly with the 802.11 / 15 coexistence group(s).WB 2001/07/04EDavidFTrinkwonH008I 2001/07/03J EditorialK8L37M8.3.10.3N2Add abbreviation "(MO) for Multiple Output systemsO-Consistency with Multiple Input (MI) section. B 2001/07/04EDavidFTrinkwonH009I 2001/07/03JTechnical, BindingK14L37MFigure 1N_Figure 1 : Add an (optional) TDD / HFDD switch function(s) at the RF TX/RX side of the diagram.O+Give equal footing to non-FDD technologies. B 2001/07/04EDavidFTrinkwonH010I 2001/07/03J EditorialK15L12M 8.3.1.1.1.2N Define AWGNOAAcronyms should be spelled out the first time that they are used. B 2001/07/04EDavidFTrinkwonH011I 2001/07/03JTechnical, BindingK15L37 ` M 8.3.1.1.1.4A Add a paragraph about the characteristics needed to support non-LOS and near-LOS deployment while maintaining payload throughput and link availability / reliability, and techniques are used in the SC and MC modulation techniques to support vthis critical requirement. AWFor predictable / reliable NLOS operation, most modulation schemes must be limited to 16-QAM (or less). This becomes even more important when considering cheap / simple installation procedures for CPE, or even self-installability and portability for CPE. There is currently nothing in the document which even mentions this critical capability.W B 2001/07/04EDavidFTrinkwonH012I 2001/07/03J EditorialK16L1M 8.3.1.1.1.5N%Spell out FFTs and IFFTs on first use B 2001/07/04EDavidFTrinkwonH013I 2001/07/03JTechnical, BindingK16L43MFigure 3! NFigure 3 Add an optional "beam mapping" function for spatial multiplexing (eg SDMA) systems. Delete note about " exists in OFDM based PHYs only ".OnAdaptive antenna configurations allow for multiple simultaneous frequency reuse, and can also apply to SC PHYsB 2001/07/04EDavidFTrinkwonH014I 2001/07/03JTechnical, BindingK18L7MTable 1NNTable 1 : Add entries for Cellular, PCS, WCS, 3650-3700MHz ,4.9GHZ GWCS bandsOConsistency with PAR.B 2001/07/04EDavidFTrinkwonH015I 2001/07/03JTechnical, BindingK18L7MTable 1A]Table 1 : 2.15 - 2.690 GHz entry : i) Change "Channel Bandwidth Options" entry to "2 to (n x 6) MHz paired with 200kHZ - (n x 6) MHz (undefined separations)" ii) Change "Reference" entry to say "FCC Parts 21 and 74 (MDS). Add MMCS (MCS ?) to Canadian Designation. Add CITEL Rec xxx (currently under consideration, primarily drafted by Canada).]O(Accuracy and to remove FDD connotations.B 2001/07/04EDavidFTrinkwonH016I 2001/07/03 "JTechnical, BindingK18L7MTable 1ATable 1 : 3.5 GHz entry : i) Change "Channel Bandwidth Options" entry to "1.75 - 28 MHz paired with 250kHz - 28 MHz (50MHz or 100MHz separations)" ii) Change "Reference" entry to include the appropriate Canadian and CITEL recommendations / referencesO(Accuracy and to remove FDD connotations.B 2001/07/04EDavidFTrinkwonH017I 2001/07/03J EditorialK18L7MTable 1NTable 1 : 5 GHz entry : i) Change "Channel Bandwidth Options" entry to describe 100 MHz or 5/10/20 MHz options ii) Change "Reference" entry to include the appropriate FCC (e.g. FCC Part 15...) and Canadian referencesB 2001/07/04EDavidFTrinkwonH018I 2001/07/03J EditorialK20L1MFigure 6NFigure 6 : Add Editors Note : This diagram might change as a result of the current FCC proceeding for Terrestrial DARS repeaters which are seeking higher transmit power allowances.O1Insert reminder to revies / revisit in due courseB 2001/07/04EDavidFTrinkwonH019!#I 2001/07/03J EditorialK20L33MFigure 7NeFigure 7 : Add Masks for Cellular, PCS, 3650-3700 (when available) and (Canadian) 3400-3700 MHz etcO CompletenessB 2001/07/04EDavidFTrinkwonH020I 2001/07/03JTechnical, Non-bindingK20L62M 8.3.2.3.1NPDelete ""...applicable for all license exempt bands between 2 and 11GHz, but .."OThis specification does not include the Direct Sequence Spread Spectrum / frequency hopping constraints of the US ISM bands / applications (eg at 2.4GHz) which are also applied in other countries.B 2001/07/04EDavidFTrinkwonH021I 2001/07/03JTechnical, BindingK22L1M 8.3.3.1.1NDelete "...and hence is capable of efficiently allocating the available traffic transport capacity" Add (after "vary with time") : ", or in adjacent sectors / cells or channels" Insert "symettric or" between "fixed" and "asymmetric"OTechnical accuracyB 2001/07/04EDavidFTrinkwonH022I 2001/07/03JTechnical, BindingK22L12M 8.3.3.1.2"$wA Change first sentence to : "In a point to multipoint (P-MP) system the base station resources can be more economically shared between multiple customer locations or connections." Capitalise the "T" for "The downlink (DL) .." which now becomes the second sentence. OAccuracyB 2001/07/04EDavidFTrinkwonH023I 2001/07/03JTechnical, BindingK22L20M 8.3.3.1.3AChange "is problematic" to "can be a limitation" Change "because of the limitations associated with a broadcast mechanism" to "which can significantly increase cell payload capacity by using simultaneous (non-broadcast) transmissions to multiple customers."OAccuracyB 2001/07/04EKenFStanwoodH024I 2001/07/02J EditorialK24L35N3Change "encoder directly" to "encoder are directly"OgrammarB 2001/07/04ENicoFvan WaesH025I 2001/07/03JTechnical, Non-bindingK25L8M 8.3.3.1.5.2#%NHarmonize puncturing over modes, resulting in two lines with generic puncturing pattern, with applicable I/Q, XY patterns underneath.O3Pointless to specify different puncturing patterns.B 2001/07/04EDavidFTrinkwonH026I 2001/07/03JTechnical, BindingK29L4M 8.3.3.1.5.4NAInsert 32-QAM (Optional) Modulation Scheme and constellation mapsAzNLOS operation is only likely to be assured with 16-QAM or lower modulation schemes, especially when cheap / simple / self install procedures are required at CPE. 32-QAM offers significant payload capacity improvements for customers who are near enough to experience better fade margins, even in NLOS deployments. 64-QAM and higher is unlikely to satisfy ANY NLOS applications.zB 2001/07/04EDavidFTrinkwonH027I 2001/07/03JTechnical, BindingK29L4M 8.3.3.1.5.4NoDelete 64-QAM and 256-QAM modes / constellations. Note : 64-QAM should in any case be Optional, not Mandatory.$&OFRD requires NLOS operation (Mandatory). Unless / until someone can demonstrate NLOS operation at modulation rates higher than 16 or 32-QAM (with field deployment experience to back it up) higher modes should NOT be included in the 802.16a Standard.B 2001/07/04EGregorioFNunezGOtherH028I 2001/07/02J EditorialK31L7M8.3.3.2NRTo replace the table 6 title by "Downlink Modulation Schemes for data transmissionOTo clarify that the pilot carrier will be always BPSK modulated in the case, at least, of License-OFDM. Therefore, BPSK will be used for pilot transmission, at least.B 2001/07/04EKirkFGriffinH029I 2001/07/02J EditorialK31L30M 8.3.3.3.1NFTable 7 title should be changed to: "Uplink ....", from "Downlink...."OTypographical error.B 2001/07/04EGregorioFNunezGOtherH030I 2001/07/02J EditorialK31L30M 8.3.3.3.1N=To replace the table 7 title by Upolink Modulation Schemes...O*This table refers to the UL, not to the DLB 2001/07/04EGregorioFNunezGOtherH031%'I 2001/07/02J EditorialK31L30M 8.3.3.3.1NTTo replace the table 7 title by "Uplink modulation schemes for data transmission..."O$Same reason as the previous comments B 2001/07/04EKirkFGriffinH032I 2001/07/02JTechnical, Non-bindingK31L35M 8.3.3.3.1NCRecommend changing 64QAM for the uplink from mandatory to optional.ODue to assymetry of data, 64 QAM uplink is superflouous. Making it optional will support broader band uplink for some applications, but does not make it mandatory where it needn't be.!B 2001/07/04ELeiFWangH033I 2001/07/04JTechnical, Non-bindingK32L6M8.3.4.1Njuse different names than Mode A / Mode B for continuous transmission and burst transmission, respectively.OIn the same document, the terms, Mode A and Mode B, are defined differently three times in: Section 8.3.4 (MAC and PHY Interfaces); Section 8.3.6.3 (OFDM PHY Layer for Licensed Bands); and Section 8.3.6.4 (OFDM PHY Layer for License-exempt bands)."B 2001/07/04EGregorioFNunezGOtherH034&("I 2001/07/02JTechnical, Non-bindingK32L8M8.3.4.1N_To review whether Mode A has been already removed from the last version of IEEE 802.16 standardO/As fa as we know, this mode has been discarded.#B 2001/07/04EKenFStanwoodH035I 2001/07/02JTechnical, Non-bindingK32L8NAre both mode A and mode B used for SC and OFDM? Are all types used for both point to multipoint and mesh? This document is very confusing regarding exactly what PHYs exist.$B 2001/07/04EDavidFTrinkwonH036I 2001/07/03JTechnical, BindingK32L9M8.3.4.1N=Add to Mode B : " (single or multiple simultaneous streams)"OAllows for Spatial Multiplexing%B 2001/07/04EGregorioFNunezGOtherH037I 2001/07/02JTechnical, Non-bindingK32L26M8.3.4.2N&To review whether mode A keeps on use.O'The same reason as the previous comment&B 2001/07/04EDavidFTrinkwonH038I 2001/07/03J EditorialK32L30M8.3.4.2NChange TDD to burst-TDDO Consistency.'B 2001/07/04EDavidFTrinkwonH039I 2001/07/03JTechnical, BindingK32L34')'M8.3.4.2NExpand on CPE requirement. Must standards compliant CPE initially be compatible with Mode A and B until handshaking is complete ? What happens if someone (eg a customer) tries to instal / register a non-supported Mode CPE ?O-Need to address real world deployment issues.(B 2001/07/04EKenFStanwoodH040I 2001/07/02JTechnical, Non-bindingK32L62NWIs QPSK used for control even in systems that allow BPSK? If so, why bother with BPSK?)B 2001/07/04EKenFStanwoodH041I 2001/07/02JTechnical, Non-bindingK34L4NAMode A should do this, too, like the old above 11 GHz mode A did.*B 2001/07/04ELeiFWangH042I 2001/07/04J EditorialK34L5M 8.3.4.2.2Neither describe more about the burst mode transmission convergence (TC) packet format or reference to the TG1 corresponding section/figure.+B 2001/07/04ELeiFWangH043I 2001/07/04JTechnical, Non-bindingK34L24M8.3.4.3Nadd the definition of DAMA.O'what is Demand Assigned Multiple Access,B 2001/07/04EDavidFTrinkwonH044(*,I 2001/07/03JTechnical, BindingK34L35M8.3.4.3N-Add OFDMA and SDMA (SDMA also applies to SC)O Completeness-B 2001/07/04ELeiFWangH045I 2001/07/04JTechnical, BindingK34L50M 8.3.4.3.1.1NCexplain why multiple burst types are needed in continuous downlink.OTG1's continuous downlink only support one burst type. If TG3/TG4 needs multiple burst type in continuous downlink, then what is the difference between it and the burst downlink?.B 2001/07/04ELeiFWangH046I 2001/07/04J EditorialK35L11M 8.3.4.3.1.2N!change "Figure 15" to "Figure 14"O>the Figure 15 is for TDD Frame, not the for the FDD operation./B 2001/07/04EKenFStanwoodH047I 2001/07/02J EditorialK35L12NWrong figure reference.0B 2001/07/04EKenFStanwoodH048I 2001/07/02J EditorialK35L24NCThe "Guardband" should be "TTG". The RTG should be in the picture.O!Consistency with everything else.1B 2001/07/04ERobertFNelsonH049I 2001/07/02JTechnical, Non-bindingK35L43M 8.3.4.3.1.3)+1AReplace lines 43 - 49 with the following: In the case of TDD, the uplink and downlink transmissions share the same frequency, but are separated in time (Figure 15). A TDD frame has a fixed duration and contains one downlink and one uplink subframe. Allocation of bandwidth within a frame is performed in terms of mini-slots (MS). The size of a mini-slot is configured for a specific system implementation and is dependent on the underlying PHY in use. The split between uplink and downlink isB a system parameter, expressed as a number of mini-slots, occurs at a mini-slot boundary within the frame, and is controlled at higher layers within the system. 8.3.4.3.1.3.1 OFDM Time Base For OFDM based systems, mini-slots are defined in terms of a finer unit called the "tick" which has a duration of 1 microsecond . The mini-slot size is always chosen so that its duration is an integer number of microseconds and so that an integer number of mini-slots fills the entire TDD frame. 8.3.4.*,1C3.1.3.2 SC Time Base For single carrier based systems, mini-slots are defined in terms of physical slots (PS). (A physical slot consists of 4 symbols). The mini-slot size must be selected to consist of N physical slots where N = 2^m where m = 0, 1, 2, 3, 4, 5, 6, or 7. This allows the size of a mini-slot to range from the size of one PS (4 symbols) to that of the SC grant request packet (128 symbols). s@A*  33)  33*ABurst packets for the SC PHY are fixed at either 128 or 256 symbols. Scheduling in terms of an arbitrary 1 microsecond time base will result in wasted bandwidth that is unnecessary. Using a time base related to the PHY symbol rate circumvents the problem. The revised text in this and succeeding comments, identifies the mini-slot definition as a PHY-dependent entity. Note: Requiring a mini-slot to be an evenly divisible fraction of a frame may be too restrictive a requirement in all cases.+-X1B In the case of a 1 millisecond frame. the only valid mini-slot values would be 1, 2, 4, 5, 8, 10, 20, 25, 40, 50, 100, 125, 200, 250, 500, and 1000 2B 2001/07/04EDavidFTrinkwonH050I 2001/07/03J EditorialK35L57M 8.3.4.3.1.3.1NDelete "clause".O inappropriate3B 2001/07/04ERobertFNelsonH051I 2001/07/02J EditorialK35L59M8.3.4.3.1.3.1 AReplace "variable" with "configurable" in line 59 The TTG has a configurable duration, which is an integer number of mini-slots. The TTG starts on a mini-slot boundary.@AR  331  33A  33M  33ROnConfigurable implies the value is a system setting under control of the equipment operator. Variable does not.4B 2001/07/04ERobertFNelsonH052I 2001/07/02JTechnical, Non-bindingK36L8M 8.3.4.3.1.3.2,.4AReplace the last two sentences of the paragraph with the following: The RTG starts on a mini-slot boundary and extends to the end of the frame. For scheduling purposes, the start of the RTG indicates the end of the TDD frame@A*  33C  33*OWith the mini-slot no longer required to be a quantity that is an even fraction of a full frame for SC, it is unlikely that the RTG will occupy an integer number of mini-slots5B 2001/07/04EDavidFTrinkwonH053I 2001/07/03J EditorialK36L8M 8.3.4.3.1.3.2N.Typo - insert space between "an" and "integer"6B 2001/07/04EKenFStanwoodH054I 2001/07/02J EditorialK36L60NChange "DL_MAP" to "DL-MAP".O(That's the official name of the message.7B 2001/07/04EKenFStanwoodH055I 2001/07/02JTechnical, Non-bindingK36L63N#Change "increasing" to "decreasing"8B 2001/07/04ELeiFWangH056I 2001/07/04JTechnical, Non-bindingK37L3M Figure 17-/8Ndefine the uplink contention intervals without distinguishing registration contention interval and BW request contention interval, so that an uplink contention interal can be used to send either a registration pkt or a BW request pkt. Awhat are the differences between the Registration Contention uplink interval and the BW Request Contention uplink interval? If the both initial registration pkt and the bandwidth request pkt can fit into the smallest possible uplink interval, why don't we just schedule an integer number of smallest uplink intervals for contention. These contention uplink intervals can be used by the CPEs for registration submissions and BW request submissions.9B 2001/07/04EKirkFGriffinH057I 2001/07/02J EditorialK37L41M 8.33.4.3.2.1N5Correct "Inteval" to "Interval" near end of the line.OTypgraphical error.:B 2001/07/04ELeiFWangH058I 2001/07/04J EditorialK38L35M 8.3.4.3.4.1.0:Nkin Table 8, switch the values of B(FDD) and B(TDD), that is, PHY type=0 is B(TDD), and PHY type=1 is B(FDD)O9keep consistence with TG1. avoid unnecessary confusions.;B 2001/07/04EKenFStanwoodH059I 2001/07/02JTechnical, Non-bindingK38L43NfAre we letting the physics of the channel differentiate between SC and OFDM? What about PMP vs. mesh?B 2001/07/04ELeiFWangH062I 2001/07/04JTechnical, Non-bindingK39L41M 8.3.4.3.4.3.3NJchange the jitter description in the line 49 to be PHY symbol independent.O.TG3/TG4's timestamp is PHY symbol independent.?B 2001/07/04/1?ELeiFWangH063I 2001/07/04J EditorialK40L27M 8.3.4.3.4.3.4N#Change "PHY Type=5" to "PHY Type=2"@B 2001/07/04EKenFStanwoodH064I 2001/07/02JTechnical, Non-bindingK40L49NThe Ack Time was removed from the UL map in March as part of an efficiency improvement of the GPC mode. This applies to page 44, line 27, too.AB 2001/07/04ELeiFWangH065I 2001/07/04JTechnical, Non-bindingK41L1M 8.3.4.3.4.4.1NAdd the descriptions to answer the following questions: 1. does a SS need to know a downlink interval is in the TDM portion or in the TDMA portion of a DL transmission? Why? 2. if yes, how does a SS know?Oit is unclear why we need the TDMA portion in the burst downlink. Since downlink has only one transmisster, i.e., the base station, it is multiplexing, not multiple access at all.BB 2001/07/04ERobertFNelsonH066I 2001/07/02JTechnical, Non-bindingK41L10M8.3.4.3.4.4.1 02BAReplace lines 10 - 14 The available bandwidth in both directions is defined with a granularity of one mini-slot (MS). The size of a mini-slot is a configurable value that must confirm to the requirements of the underlying PHY. @A*  33  33*O1Consistency with definitions in earlier sections.CB 2001/07/04ELeiFWangH067I 2001/07/04J EditorialK42L1M Figure 21N.Change the "TDM" to "TDMA" in the TDMA portionO4keep consistance with TG1 (there is a same diagram).DB 2001/07/04EGregorioFNnezH068I 2001/07/04J EditorialK42L12M 8.3.4.3.4.4.1NTo change the name of the TDMA bursts included in figure 21. To replace the name "TDM DIUC..." by "TDMA DIUC..."in each burst preceeded by a preamble.OFAll the burst preceeded by a preamble are TDMA bursts not TDM bursts. EB 2001/07/04ERobertFNelsonH069I 2001/07/02J EditorialK42L32M8.3.4.3.4.4.2 NDIUC=0O1Typo - UIUC was incorrectly used instead of DIUC.FB 2001/07/04ELeiFWang13FH070I 2001/07/04JTechnical, Non-bindingK43L1M 8.3.4.3.4.4.6NLAdd more Frame Length Code to the Table 9 to allow MAC frame size up to 10msOfor more flexibilityGB 2001/07/04ELeiFWangH071I 2001/07/04JTechnical, BindingK43L1M 8.3.4.3.4.4.6NFadd the description how the MAC frame size info is deliveried to SS's.OIn TG1, it uses the PHY synchronization field in the DL-MAP for the burst mode to delivery the MAC frame size info to SS's. In TG3/TG4, burst mode DL-MAP PHY synchronization field is used to carry the uplink time stamp. HB 2001/07/04ELeiFWangH072I 2001/07/04JTechnical, BindingK43L1M 8.3.4.3.4.4.6N:discuss why we need to define the MAC frame and its size? OIf the system initialization time is used as the reference point of offset fields in the MAP messages, it seems not necessary to specify the MAC frame size.IB 2001/07/04EKenFStanwoodH073I 2001/07/02JTechnical, Non-bindingK43L36NLThis is the mode A description. Use the description from the 10-66 GHz PHY.J24JB 2001/07/04ERobertFNelsonH074I 2001/07/02JTechnical, Non-bindingK43L50M8.3.4.3.4.4.7 A|Replaces lines 50 - 63 The format of the PHY Synchronization field is a 32-bit unsigned integer value. The value is incremented for each frame transmitted and wraps-around when the maximum value (2^32 -1) is reached. PHY control logic on subscriber units deals with actual timing synchronization with the base station using the periodic burst transmissions from the base station.|@A*  33  33*OSince TDD operations are based on discrete frames, using them as the time unit provides a reliable and straightforward method for synchronizing time references betwen the base station and subscribers. KB 2001/07/04ELeiFWangH075I 2001/07/04J EditorialK43L50M 8.3.4.3.4.4.7Nremove this paragraphO5it is identical to the paragraph on page 39, line 41.LB 2001/07/04ERobertFNelsonH076I 2001/07/02JTechnical, Non-bindingK44L3M8.3.4.3.4.4.8 35LAReplaces lines 3 - 25 The field Allocation Start Time appears in both the downlink (DL_MAP) and uplink (UL_MAP) map messages. The value specifies a mini-slot boundary that is the reference point for all mini-slot offset values appearing in the same map message. The value is expressed as the least significant 16-bits of the full frame count value (DL_MAP synchronization field value, section 8.3.4.3.4.4.7) corresponding to the target frame and a 16-bit mini-slot offset form the start of theB frame. When computing the size of a mapped element whose successor has an offset beyond the end of the frame, the start of the RTG gap is used in place of the successor's offset. For computing the start of the successor element, the portion of the TDD frame between the RTG and the end of the frame is ignored. (i.e. for scheduling purposes, the start of the RTG and the start of the next frame are considered coincident.) @A*  33  33*46LAReflects the change to using frame count as the alloc start time. Preserves ability included in current proposal to map across frame boundaries. Removes discussion on compatability ot tick mechanism for SC and OFDM which was removed for SC in earlier sections.MB 2001/07/04ELeiFWangH077I 2001/07/04J EditorialK44L5M 8.3.4.3.4.4.8N#Change "PHY Type=5" to "PHY Type=2"NB 2001/07/04ERussellFMckownH078I 2001/07/03JTechnical, Non-bindingK44L8M 8.3.4.3.4.4.8Nqdelete text from line 8 to line 25. Replace with: "The mini-slot is an integer number of symbol intervals." AThe definition of the "mini-slots" in section 8.3.4.3 .4.4.8 is defective. Mini-slots should be defined in terms of an integer number of symbols. This allows the MAC to efficiently allocate bandwidth. The symbol duration should be a free design parameter.OB 2001/07/04ERobertFNelsonH079I 2001/07/02JTechnical, Non-bindingK44L30M8.3.4.3.4.4.9 57OA!Replace lines 30 & 31 with the following The Ack Time is the frame counter value (DL_MAP synchronization field value, section 8.3.4.3.4.4.7) of the last frame whose uplink transmissions have been processed by the base station. This time is used by the SS for collision detection purposes.!@A*  33(  33*AReflects change from mini-slot counter to frame counter for synchronization field. For finer resolution, the format of the alloc start time (frame count plus mini-slot offset) could be used. Note: The current (D3/2001) TG1 draft has removed the field from the UL_MAP message. The only reference remaining in the draft is text describing bandwidth request operations. As a result, if TG1 is truly deleting the field, the appropriate action here is to delete the section as well.PB 2001/07/04EVladimirFYanoverH080I 2001/07/03JTechnical, Non-bindingK44L33M8.3.4.468PNAInsert the text from the document "802.16abc-01/03" by V.Yanover O2This supplies the relevant material to the sectionQB 2001/07/04EVladimirFYanoverH081I 2001/07/03JTechnical, Non-bindingK44L33M8.3.4.4NXInsert in the section 8.3.4.4 the text from the document "802.16abc-01/02" by V.Yanover.O2This supplies the relevant material to the sectionRB 2001/07/04EKenFStanwoodH082I 2001/07/02JTechnical, Non-bindingK45L54N.Show the protocol for what is learned and how.SB 2001/07/04EKenFStanwoodH083I 2001/07/02J EditorialK49L63N(Change "frame.Figure" to "frame. Figure"TB 2001/07/04ELeiFWangH084I 2001/07/04J EditorialK49L63M 8.3.4.5.2.3N"Change "Figure 297" to "Figure 29"UB 2001/07/04EKirkFGriffinH085I 2001/07/02JTechnical, Non-bindingK51LNAM8.3.5.3NRecommend adding additional information to the single carrier section of the document to include specific requirements on the PHY signal path and equipment. Refer to the following document: IEEE 802.16abc-01/06.79UA5Important environmental and performance parameters are left out of the document for the SC section. Basic requirements must be identified to insure standards compliance at the PHY level. Equivalent parameters may be included for the OFDM section as needed. OFDM is not addressed in the referenced document.5VB 2001/07/04EDavidFTrinkwonH086I 2001/07/03JTechnical, BindingK56L1M 8.3.5.4.2NDelete section 8.3.5.4.2OeNot relevant to a Standards Specification. We should not be arguing SC vs OFDM PHYs in this document.WB 2001/07/04EGregorioFNnezH087I 2001/07/04J EditorialK56L54M 8.3.5.4.2N$To replace the parameter "M" by "N".OIt is understood that it refers to the OFDM symbol duration without considering the cyclix prefix inclusion. This duration is N according to figure 37XB 2001/07/04EBrianFEidsonH088I 2001/07/03JTechnical, Non-bindingK57L31M 8.3.5.5.2.18:XNhComplete rewrite of the Concatenated Reed-Solomon and Convolutional Code section for Single Carrier modeAPrevious description is very sketchy, and does not even reflect some of the description differences between OFDM and SCM Concatenated FEC which were referenced in the common area. In particular, the SCM-specific concatenated FEC section does not indicate how codes are convolutional codes are generated, the bit-to-constellation map, or what process is used to determine the shortening factor for the Reed Solomon. All of these shortcomings are addressed in the contribution. YB 2001/07/04EKenFStanwoodH089I 2001/07/02J EditorialK57L43NMissing figure reference.ZB 2001/07/04EKenFStanwoodH090I 2001/07/02J EditorialK58L19N+Change "set that are set" to "that are set"[B 2001/07/04EDavidFTrinkwonH091I 2001/07/03JTechnical, BindingK59L20M 8.3.5.5.2.2.1N&Delete 64-QAM modes / constellations.9;[OFRD requires NLOS operation (Mandatory). Unless / until someone can demonstrate NLOS operation at modulation rates higher than 16 or 32-QAM (with field deployment experience to back it up) higher modes should NOT be included in the 802.16a Standard.\B 2001/07/04EDavidFTrinkwonH092I 2001/07/03JTechnical, BindingK59L20M 8.3.5.5.2.2.2NAInsert 32-QAM (Optional) Modulation Scheme and constellation mapsAzNLOS operation is only likely to be assured with 16-QAM or lower modulation schemes, especially when cheap / simple / self install procedures are required at CPE. 32-QAM offers significant payload capacity improvements for customers who are near enough to experience better fade margins, even in NLOS deployments. 64-QAM and higher is unlikely to satisfy ANY NLOS applications.z]B 2001/07/04EKenFStanwoodH093I 2001/07/02JTechnical, Non-bindingK62L59N2Line 59 and line 65 seem to contradict each other.^B 2001/07/04EBrianFEidsonH094I 2001/07/03J EditorialK63L28:<^M8.3.5.8N3"Sc-FDE" to "SC-FDE" and eliminate " and OFDM" OTypo on "Sc-FDE". Ensuing sub-sections now dedicated solely to Single Carrier mode; their description really do not apply to the OFDM modes any more._B 2001/07/04EBrianFEidsonH095I 2001/07/03J EditorialK64L34M 8.3.5.10.1.1NChange from: The interval between Unique Words is typically chosen ... To: The interval between Unique Words may be chosen ...ONo reason to presuppose/recommend a typical choice, since other equalization options, such as overlap save equalization, are available. Leaving this as 'may' seems more even-handed.`B 2001/07/04EBrianFEidsonH096I 2001/07/03J EditorialK64L37M 8.3.5.10.1.1;=`A>Change from: However, this is not a necessary requirements if an overlap save algorithm is used for equalization. To: However, FFT sizes for frequency domain equalizers are not required to equal unique word intervals, since, for example, frequency domain equalizers can be implemented using overlap save techniques.>OThe original statement was not precisely worded, and seems to imply that FFT sizes may be different from unique word intervals only if an overlap save technique is used. The intent was to provide one example where this is not necessary aB 2001/07/04EBrianFEidsonH097I 2001/07/03J EditorialK64L40M 8.3.5.10.1.1NChange from: To reduce computational requirements of FFTs, the length F = N + U can preferably be ... To: The length F = N + U can be chosen as ... (Also, append this to the previous paragraph)OStatement of a fact is fine. Flow fits better with preceding paragraph, so this sentence should be merged with previous paragraph.bB 2001/07/04EBrian<>IbFEidsonH098I 2001/07/03J EditorialK65L41M 8.3.5.10.1.1NModify sentence and complete its end to include the desired chapter reference Instead of "preferred lengths", use "lengths" ... Unique Words, U, and lengths, U, may be found in Chapter 8.3.5.12.2.OReference at end of sentence must have been dropped during editing. "preferred" is eliminated, because it is not an accurate descriptor. Supported Unique Word lengths are mandated in 8.3.5.12.2; other unspecified lengths are not allowed.cB 2001/07/04EKenFStanwoodH099I 2001/07/02JTechnical, Non-bindingK65L61=?cAThe use of "MAC Header" is inconsistent here. This is really the control portion of the channel which typically contains teh DL-MAP and UL-MAP as opposed to a "MAC header" on a PDU. We tried earlier to define logicla channels such as a Broadcast Control Channel (BCCh), but no one wanted to bother. This might be a good time to reintroduce that concept to add clarity to the discussion. Alternativesly, "Frame Control MAC messages" on page 66, line 36 is a better term.dB 2001/07/04EKenFStanwoodH100I 2001/07/02JTechnical, Non-bindingK66L56N#Change "increasing" to "decreasing"eB 2001/07/04EBrianFEidsonH101I 2001/07/03J EditorialK67L43M 8.3.5.10.3.6Nremove "fixed"O>a-priori known is good enough; "fixed" may be too restrictive.fB 2001/07/04EKenFStanwoodH102I 2001/07/02J EditorialK68L1N"Change "that it the" to "that the"gB 2001/07/04EBrianFEidsonH103I 2001/07/03J EditorialK68L61M 8.3.5.10.5>@gNHEliminate sentence Additional details related to accepted lengths ... ODoes not add descriptive content to section, and no binding requirements are placed on F = N + U, anyway. Plus, the section reference at the end of the sentence would have to updated, anyway.hB 2001/07/04EBrianFEidsonH104I 2001/07/03J EditorialK70L38M 8.3.5.11.2.1N"Note that the patterned 'middle portion' of the block" ... To: "Note that the patterned 'middle portion' of a Unique Word-Payload ... Unique Word-Payload block" OThis makes it clearer that the reference is to a UW-patterned block. As sentence 33 on this same page implies, this fundamental pattern of UWs may or may not be used. iB 2001/07/04EBrianFEidsonH105I 2001/07/03J EditorialK70L39M 8.3.5.11.2.1?AiA# ... "that is spanned by several FFTs, and is processed by an equalizer using the overlap save method". To: "that is spanned by, potentially, several FFTs, and is processed by an equalization technique which does not necessarily rely on cyclic prefixes, such as overlap save processing."#OThe original text is worded as if the overlap save processing must be used. The receiver implementation should be up to the implementer.jB 2001/07/04EBrianFEidsonH106I 2001/07/03J EditorialK70L41M 8.3.5.11.2.1ANote that the end of a block patterned ... to clear the receiver. To: Note that the end of any burst, repetitively Unique Word patterned or not, is not required to conclude with a Unique Word. However, the burst must be followed by a blanking interval, where nothing is transmitted, to allow channel delay spread to clear the receiver. This blanking interval shall be the length of a Unique Word.@BjOFThe above more accurate expresses the intent of the original sentence.kB 2001/07/04EBrianFEidsonH107I 2001/07/03J EditorialK70L46M 8.3.5.11.2.1N@"F = N + U should preferably be chosen" To: "may be chosen" OCDo not provide/bias preferences when choice is entirely up to user.lB 2001/07/04EBrianFEidsonH108I 2001/07/03J EditorialK70L60M 8.3.5.11.2.1NEliminate last sentenceOoF interval reference does not add to discussion, and the lengths are, in the unpatterned case, almost anything.mB 2001/07/04EBrianFEidsonH109I 2001/07/03J EditorialK71L12M8.3.5.11.2.1.2A... "the final payload block to can be shortened any length" ... To: ... "the final payload block to can be shortened to any length" ... AND change "process part of the last block of a frame, using the overlap save method for FFTs" To: "process part of an already-FFT processed data segment ,as well as the remainder of the data, together, as a single FFT, using, for example, the overlap save method".AC*mOw1st comment: "to" added to make sentence grammatically correct. 2nd comment: to make the last statement more explicit.nB 2001/07/04EBrianFEidsonH110I 2001/07/03J EditorialK71L15M8.3.5.11.2.1.2A"the final block size (not including the blanking interal at the end to accommodate delay spread) is 2^n - U + 1. Nend + U = 2n" To: "the final payload block size (not including the blanking interal at the end to accommodate delay spread) is Nend = FFT size - 2U + 1. "O"payload block size" makes "block size" reference clearer. 'Equation appears to have been 'splattered' during editing. This returns it to a cogent form.oB 2001/07/04EBrianFEidsonH111I 2001/07/03J EditorialK72L51M 8.3.5.11.3NFEliminate "Additional details related to accepted lengths" statement. BDoOThis was eliminated in continuous section, for the same reason: that F is an optional length---and it confuses the reader to refer to options that somewhat depend on the context in which they are used.pB 2001/07/04EBrianFEidsonH112I 2001/07/03JTechnical, Non-bindingK73L73M 8.3.5.11.4NtWill be made in the form of a contribution, by commentor on section 8.3.5.11.4. (Also involves replacing Figure 49)ADue to editing confusion, Figure 49 is a replica of Figure 48, and the text describing what was supposed to be Figure 49 did not find its way into the draft document. A host of other descriptions seem to have disappeared as well. The new description (and Figure) for Fig 49 should eliminate the final unique word, and demonstrate a case where periodic Unique Words are not used. What's more, the concluding blanking interval, which was described in previous sections, did not find itself into B9this section---even in Figure 48. This will be included. 'qB 2001/07/04EBrianCEqFEidsonH113I 2001/07/03JTechnical, Non-bindingK74L52M8.3.5.12AAdd new section with examples and illustrations of (a) continuous/FDD downstream -burst upstream (b) TDD downstream/upstream with framing details broken out, and some MAC interface structures illustrated. Entended change submitted as a contribution on the part of commentor. ANo place in the document illustrates how the burst elements (TDMA and TDM frames) and continuous elements fit into an FDD or TDD scheme---for Single Carrier. As well, there is no illustrate on how these are tied to MAC structures, such as access structures and the location of MAPs. Such material, in a slightly less detailed format, appeared in the March 2001 submission (of a single carrier contribution), but somehow did not find its way into the draft document.rB 2001/07/04EKenFStanwoodH114I 2001/07/02J EditorialK78L3N4Change "although the although the" to "although the"sB 2001/07/04EKenFStanwoodH115DFsI 2001/07/02J EditorialK78L36N Change "asFigure" to "as Figure"tB 2001/07/04EBrianFEidsonH116I 2001/07/03J EditorialK79L32M8.3.5.16N=All references of 64 to 63, and all references of 192 to 193.OAccepted UWs (prefixes) for the single carrier mode are 63, rather than 64. Use of 64 is inconsistent with the UW definition. 256-63 = 193, hence the use of 193 in lieu of 192. uB 2001/07/04EBrianFEidsonH117I 2001/07/03JTechnical, Non-bindingK79L32M8.3.5.16NExtended comment that is intended as an overall rewrite of this section, with changes focusing on the second half of this section. Will be made as a contribution to section 8.3.5.16 by commentor.EG7uADetails on shortened messages have been reworked to improve bandwidth efficiency (i.e., remove transmission overhead), while simultaneously simplifying the decision-making required of the FEC to determine the amount of shortening (on Reed Solomon codewords, for example), that it must use. An additional example is also provided. Also, the rates of convolutional codes being used in this section are made consistent with the rates of convolutional codes specified elsewhere in the document (B[rate 5/6 16-QAM and rate 7/8 64-QAM, for example, are specified elsewhere in the document).IvB 2001/07/04EKenFStanwoodH118I 2001/07/02JTechnical, Non-bindingK80L51N+All variants of the PHY or just TDMA burst?wB 2001/07/04EKirkFGriffinH119I 2001/07/02JTechnical, Non-bindingK81L43M 8.3.5.18.0.1FHwNDelete the % signs in front of 0.15 and 0.25 in reference to the rolloff factors for the pulse shaping filter. Also, an alpha for .15 will lead to a very high degree of ISI, so it is recommended that this value be deleted.OXTypographical. Alpha of 0.15 will be practically too small for robust ISI performance.xB 2001/07/04EGregorioFNnezH120I 2001/07/04J EditorialK85L54M 8.3.6.1.2NTo replace "(s" by "microsec."O Typo erroryB 2001/07/04ELeiFWangH121I 2001/07/04J EditorialK85L54M 8.3.6.1.2Nchange/remove "(s"zB 2001/07/04ENicoFvan WaesH122I 2001/07/03JTechnical, Non-bindingK87L19M 8.3.6.1.4N;Delete FFT-sizes 128,512, 1024, 4096 (throughout document).GIzADue to their optional status, CPE's must start up in a mandatory FFT size, then switch to an optional one. A BS supporting this must then be capable of switching from the mandatory to the optional FFT sizes on the fly to support these CPE's while not prohibiting CPE's running the mandatory modes only. This all creates complexity without serving any purpose (it would be different if different FFT sizes were mandatory for different channel widths, but this is currently not the case). {B 2001/07/04EGregorioFNnezH123I 2001/07/04J EditorialK93L2M 8.3.6.2.1NTo replace "(s" by "microsec."O Typo error|B 2001/07/04EKenFStanwoodH124I 2001/07/02J EditorialK93L2NWhat is "2(s" supposed to be?}B 2001/07/04EGregorioFNnezH125I 2001/07/04J EditorialK96L59M 8.3.6.3.2.1.1NTo replace ". The downlink initialization of the randomizer for OFDMA...", by ". The uplink initialization of the randomizer for OFDMA...".HJ}ODIt is refered to the randomizer initialization for the uplink period~B 2001/07/04ELeiFWangH126I 2001/07/04JTechnical, Non-bindingK100L38M 8.3.6.3.2.4.1NGadd a paragraph to explain why the network synchronization is required.O6what are the benefits to have network synchronization?B 2001/07/04ELeiFWangH127I 2001/07/04J EditorialK100L54M8.3.6.3.2.4.1.2NAdd reference to the table listing the receiver sensitivities.€B 2001/07/04EJoriF ArrakoskiGMemberH155I 2001/07/02J EditorialK130L45M 8.3.6.4.2.5.7NRemove sectionONot part of the standard.€B 2001/07/04ENicoFvan WaesH156I 2001/07/03JTechnical, Non-bindingK131L11M 8.3.6.4.2.6N4Significantly reduce number of coding possibilities.OLThis many possibilities is expensive to implement and technically pointless.€B 2001/07/04EKenFStanwoodH157I 2001/07/02JTechnical, Non-bindingK131L42QS€N:LSB first, MSB last is the opposite of the MAC definition.€B 2001/07/04ENicoFvan WaesH158I 2001/07/03JTechnical, Non-bindingK132L38M8.3.6.4.2.6.2.2NdDelete figure 91 through 94, and adjust line 38-46 to refer to 8.3.3.1.5.2.2 and its figure instead.O"Harmonization of methods and text.€B 2001/07/04EKenFStanwoodH159I 2001/07/02J EditorialK140L1N5The bibliographic references should be in section 13.€ B 2001/07/04EDavidFTrinkwonH160I 2001/07/03JTechnical, BindingK144L1MAnnexN Delete AnnexOJThis should be part of a Recommended Practise for coexistence purposes etc€!B 2001/07/04EKenFStanwoodH161I 2001/07/02J EditorialK144L26NRChange "interference Wireless HUMAN interference" to "Wireless HUMAN interference"€"B 2001/07/04EKenFStanwoodH162I 2001/07/02J EditorialK146L50N2Change "calculations the" to "calculations is the"€#B 2001/07/04ENicoFvan WaesH163I 2001/07/03JTechnical, Non-bindingK148L18M8.3.9.3NDelete section 8.3.9.3.RT€#O_Bad polling idea, and power control parts should be rolled into the MAC in one form or another.€$B 2001/07/04EKenFStanwoodH164I 2001/07/02J EditorialK149L36Nchange ") ." to ")."€%B 2001/07/04EKenFStanwoodH165I 2001/07/02J EditorialK157L27N&Change "paths send" to "paths to send"€&B 2001/07/04EKenFStanwoodH166I 2001/07/02J EditorialK166L41M1.1AWhile breaking the scope section up into problem domain areas is a good idea, I feel that the proposed structure is not the best. The paragraph on page 166 lines 52-56 is applicable to all 802.16 implementations and should be moved up to line 38. Section 1.2.2 should be multiple sections. You have defined at least 5 different systems below 11 GHz - SC continuous, SC burst, OFDM continuous, OFDM burst, and mesh. The rationale for so many variants and should be here. By the same token,Bg the paragraph for 10-66 should be improved to distinguish its different environment form below 11 GHz.U€'B 2001/07/04SU€'EJin-MengFHoGOtherH167I 2001/06/29JTechnical, Non-bindingK167M1.4.1NChange the "optional mesh topogloy" to a mandatory "sidelink communication" capability, by which an SS can send a packet directly to another SS, with channel access still coordinated by a BS.AThe current definition of CID--a unidirectional mapping between a given SS and the BS--is not adequate for the MAC address filtering of an SS to SS transmission in a mesh topology, yet it appears in the generic MAC header (and ARQ Feedback IE) which is also used for the mesh topology. It is true that most subscriber units are idle most of the time, but a few active units at any given time (or at peak times) will occupy the whole shared channel (otherwise we would not need so much bandwidtTVC€'Bh in the first place). Hopping through idle stations in a mesh network bites into the channel capacity--in an inefficient way--as it leads to collisions which in turn would destroy the QoS expectations due to the uncertain process of collision resolution--especially in an ad hoc (i.e., mesh) topology. As far as power consumption and interference are concerned, they are not always related. Interference results from transmission, be it from a BS or SS. P55€(B 2001/07/04EKenFStanwoodH168I 2001/07/02J EditorialK167L3M1.1.2NAre we using terminology like "802.16a"? What about "802.16.3" on line 8? Is "IEEE 802.16.3-00/02r4" going to generally available to the public? If not it shouldn't be reference in the spec.€)B 2001/07/04EKenFStanwoodH169I 2001/07/02J EditorialK167L14M1.1.2UW€)A'The last sentence of the paragraph should be deleted. The change to the specification should not need to justify itself. Justification should be outside the specification. The need for non-line of sight, etc. is part of the problem domain for below 11 GHz and should be stated in those terms.'€*B 2001/07/04EKenFStanwoodH170I 2001/07/02J EditorialK168L9M1.4.1NmThe "sentence" that starts out "Likewise," is not a sentence and makes the paragraph difficult to understand.€+B 2001/07/04EKenFStanwoodH171I 2001/07/02J EditorialK168L21M4AThis acronym list appears to be an outdated version from an old copy of the spec. What are you trying to add? It's completely unclear. This is a CHANGE document. You can simply list the new acronyms that need to be added, thus avoiding a big CM problem.€,B 2001/07/04EDavidFTrinkwonH172I 2001/07/03J EditorialK168L25MAcronyms & AbbreviationsVX€,NAdd AWGN, MI, MO, SC, MC, OFDM, OFDMA, SDMA, burst-FDD, burst-TDD and other missing acronyms in appropriate alphabetical sequence.€-B 2001/07/04EVladimirFYanoverH173I 2001/07/03JTechnical, Non-bindingK170L57M6.2.2A1. Rename subsections of the current 6.2.2 the way that reflects that they specify the formats for SC PHY only For example, "Sub-11 GHz Downlink Map (DL_MAP) Message" shpuld be reanamed to "Sub-11 GHz Downlink Map (DL_MAP) Message for Single Carrier PHY" 2. Arrange a subsection under the section 6.2.2 with the name "Single Carrier PHY Management Messages Formats" and place there the current content of 6.2.2. with new 3. rename section 6.2.2. to "Sub-11 GHz Management Messages Formats"B WY'€-A$The formats of these management messages are taken obviously from the document #61 (or #61r1) where they appear under the title "MAC-PHY Interface for Single Carrier PHY". They never were represented or discussed or accepted as universal formats suitable for all the PHY options of 802.16a/b$€.B 2001/07/04EVladimirFYanoverH174I 2001/07/03JTechnical, Non-bindingK170L57M6.2.2NAdd one more subsection under 6.2.2. to describe OFDM/OFDMA PHY sepecific message formats with division as needed into OFDM/OFDMA optionsOThere is no section to specify the named formats though the formats for OFDM/A obviously needs another approach because of e.g. two-dimensional structure of the frame €/B 2001/07/04EKenFStanwoodH175I 2001/07/02JTechnical, Non-bindingK171L1M 6.2.2.4.3.1XZ@€/AYou don't need a separate DL-MAP message for below 11 GHz. Based on requests from TG3 and TG4, PHY specific elements of the MAPs are supposed to be defined in the applicable PHY section. Looking at your DL-MAP message on page 172, lines 5-30 are identical to what's currently in the spec. If you need an "Allocation Start Time" (page 172, line 31-37 and page 173, lines 17-19) for the downlink, there are 2 options: 1) add it to the main message and have it apply to all versions of 802.16 B2) move it to the PHY specific section. It was already agreed between TG1, TG3, and TG4 that the Map Information Elements (page 172, lines 38-55) would be moved to the appropriate PHY section. This includes page 173, lines 23-28.€0B 2001/07/04ELeiFWangH176I 2001/07/04JTechnical, BindingK172L1M 6.2.2.4.3.1Y[*€0NRemove the Connection ID field in the Information Elements (IEs) of the DL-MAP in the figure shown on page 172; Also, remove the corresponding description given on page 173, line 25.O1. Connection ID will be given in the MPDUs in the data playload of an DL interval; 2. If connection ID is shown in an IE of DL-MAP, multiple IEs are required for multiple CIDs with the same DIUCs, even for the same SS. €1B 2001/07/04ELeiFWangH177I 2001/07/04JTechnical, Non-bindingK172L1MFigure 1NRemove the allocation start time field in the DL-MAP shown on page 172; Also, remove the corresponding description on page 173, lin 17 to line 19.O?The uplink "Alloc start time" is given in the UL-MAP already.€2B 2001/07/04ERobertFNelsonH178I 2001/07/02JTechnical, Non-bindingK173L10M 6.2.2.4.3.1 Z\€2AInsert the following before PHY Synchronization. Also insert the appropriate field entry to the graphic on page 172 DCD Count Matches the value of the Configuration Change Count of the DCD which describes the burst parameters that apply to this map. @A*  33s  33*OcThe field is missing but is required for proper coordination of changes to the contents of the DCD.€3B 2001/07/04EKenFStanwoodH179I 2001/07/02J EditorialK173L18M 6.2.2.4.3.1Nchange "uplink" to "downlink"€4B 2001/07/04ERobertFNelsonH180I 2001/07/02JTechnical, Non-bindingK173L25M 6.2.2.4.3.1AAppend following notation to the end of the Connection Identifier entry of the Information element list. Also make an appropriate modification to the graphic on page 172 This field is only included in the information element if the host system employs TDMA for downlink transmissions.@A*  33  33*[]€4OCID requires 2 bytes of bandwidth for each information element appearing in the map and is wasted space for TDM-based systems which do not require this information. €5B 2001/07/04EKenFStanwoodH181I 2001/07/02JTechnical, Non-bindingK173L43AYou don't need a separate UL-MAP message for below 11 GHz. It was agreed in February, at TG3 and TG4's request, that the PHY specific portions be moved to the appropriate PHY section. This includes page 175, lines 42-63; page 176, lines 38-44; page 177line 1 through page 178, line 50.€6B 2001/07/04ELeiFWangH182I 2001/07/04JTechnical, BindingK175L1M 6.2.2.4.4\^>€6A1. Change the "Connection ID" to "the basic uplink connection ID of a SS" in the IEs of the UL-MAP shown in the figure on page 175. 2. change the "connection identifier" to " the connection identifier of the ss's basic uplink connection" on page 176, line 41. 2. change the "connection ID" to "the basic uplink connection ID of a SS" in Figure 3 on page 177. 3. change corresponding description on page 177 like this: The connection ID field in the UL-MAP IE shall be the basic uplink conneB6ction ID of a SS. The uplink intervals specified in the UL-MAP are assigned on the SS basis, not on the uplink connection basis. In the UL-MAP, a SS is identified by its basic uplink connection ID. It is the SS's responsibility to distribute the scheduled uplink bandwidth among its active uplink connections.$]_€6A41. UL interval (UL transmission opportunity) should be assigned on a SS basis, not on uplink connection basis, because the SS does the transmission; 2. if UL interval is assigned on connection basis, multiple IEs are required in the same UL-MAP for the assignments to the multiple connections of the same SS.4€7B 2001/07/04EKenFStanwoodH183I 2001/07/02JTechnical, Non-bindingK175L28NThe "Acknowledgement Time" is no longer in the UL MAP. This was eliminated in a comment from the March meeting that resulted in improving the bandwidth efficiency of the GPC mode. This includes the deletion of page 176, lines 19-23.€8B 2001/07/04EKenFStanwoodH184I 2001/07/02JTechnical, Non-bindingK175L35NThe backoffs fields were already moved to the UCD message since they don't need to be sent every time there's a UL-MAP. This applies to page 176, lines 24-37 also.€9B 2001/07/04ELeiFWangH185I 2001/07/04JTechnical, Non-bindingK176L15M 6.2.2.4.4.1^`€9Noremove " the start of a frame in which DL-MAP message is transmitted (PHY Type ={0,1}) or" and "(PHY Type=2)"OgThe PHY synchronization field in the MAP messages has been changed to uplink time stamp in TG3/TG4 doc.€:B 2001/07/04ERobertFNelsonH186I 2001/07/02JTechnical, Non-bindingK176L20M 6.2.2.4.4.1 NDelete definition of Ack TIme Delete Ack Time field from the graphic on page 175 Delete pending notfication list from graphic on page 177 (lines 21-30)OuIn D3/2001 the Ack Time field has been deleted except for one text reference describing bandwidth request operations.€;B 2001/07/04EKenFStanwoodH187I 2001/07/02JTechnical, Non-bindingK177L20NDelete lines 20-30. The GPC grant/request protocol was made much more efficient in March, and no longer needs the data grant pending map entries.€B 2001/07/04EKenFStanwood`b€>H190I 2001/07/02J EditorialK179L44NIf the map time relevence is not going to be kept generic to TDD or FDD, it all needs to be moved to the appropriate PHY sections.€?B 2001/07/04EKenFStanwoodH191I 2001/07/02JTechnical, Non-bindingK180L1AThe drawings are confusing. They show a "maximum" time relevance, with no minimum. Are you assuming 0 processing time for the minimum. Why is no relevance shown for the Uplink? Do you really mean that this applies to both the UL and the DL, but between the 2 arrows? Is this for use with the combined map message only? The combined map message could be a useful above 10 GHz, but not with these arbitrary restrictions. Given that the maps are variable length, and you have the domain of B1 map including the transmit time of the next, you have a race condition. You don't really know where the first burst profile that's differenc from the well known control burst profile will start.€@B 2001/07/04EKenFStanwoodH192ac€@I 2001/07/02J EditorialK181L11N6Lines 11-26 need to be in teh appropriate PHY section.€AB 2001/07/04ELeiFWangH193I 2001/07/04JTechnical, Non-bindingK181L11M 6.2.8.0.0.1N\change the "connection identifier" to " basic uplink connection ID" in the power control IE.OATx power should be on the station basis, not on connection basis.€BB 2001/07/04ELeiFWangH194I 2001/07/04JTechnical, Non-bindingK181L23M 6.2.8.0.0.1N^consider to use the normal ranging messages to perform power control, not use the MAP message.OQRanging messages are designed for the SS to acquire the correct parameter values.€CB 2001/07/04EKenFStanwoodH195I 2001/07/02J EditorialK181L31NIf special maps are needed for Mesh systems, they need to be added to table 11. The Request and Grant IEs should be moved to the appropriate PHY section.€DB 2001/07/04EKenFStanwoodH196I 2001/07/02JTechnical, Non-bindingK183L1NDThis looks identical to the MSH-SCH message. What's the difference?€EB 2001/07/04EKenbd€EFStanwoodH197I 2001/07/02J EditorialK184L42N*6.2.7.7.4 should be a subsection of 6.2.9.€FB 2001/07/04EKenFStanwoodH198I 2001/07/02J EditorialK185L5NThe message definitions on page 185, lines 5-65; page 186, lines 7-14; and page 186, line 30 through 188, line 62 should be a subsection of 6.2.2.4 similar to 6.2.2.4.10. The TLV definitions should be in section 11.1.€GB 2001/07/04EKenFStanwoodH199I 2001/07/02J EditorialK186L4Nchange "basis on" to "basis of"€HB 2001/07/04EKenFStanwoodH200I 2001/07/02JTechnical, Non-bindingK186L7NIn reference to block 14, does blocking occur before or after fragmentation. If after, as is shown here, then there are scenarios that can cause ambiguity about the length in the receiving end unless extra care is taken.€IB 2001/07/04ELeiFWangH201I 2001/07/04J EditorialK186L7M 6.2.7.8.1NGchange one of the two identical section titles: 6.2.7.8.1 and 6.2.7.8.3€JB 2001/07/04EJin-MengFHoGOtherH202I 2001/06/29ce€JJTechnical, Non-bindingK189NReduce the ARQ options defined on pages 189-208. Block based numbering is too complex, short sequence field is not adequate, seperating ACK from bitmap based CACK is not worth.AThe ARQ mechanism as defined on pages 189-208 introduces a fundamental MAC-level acknowledgment mechanism into 802.16 and is an essential adaptation of the DOCSIS MAC for a wireless network. However, it specifies too many features or options, and may result in substantial complexity that will inhibit 802.16 from being as successful as DOCSIS. The fate of 802.14 should be avoided.P53€KB 2001/07/04EJin-MengFHoGOtherH203I 2001/06/29JTechnical, Non-bindingK189df€KAA few lines need be added to make the ARQ operation normative. (a) The expiration of "retransmission retries" of a given packet used in defining the ARQ Timeout parameter on page 192 need be referenced to a specific time point. This reference may be chosen to be the point when the next out-of-sequence packet is received at the MAC, so that a received packet is passed up to the higher layer within an ARQ Timeout of its reception at the receiving MAC, regardless of the reception status ofB packets of lower sequence number. (b) There need be a Retransmission Timeout parameter at the transmitting MAC entity, so that the transmitting MAC will not retransmit a frame for which the retries are considered to have expired by the receiving MAC. Such a Retransmission Timeout may be equal to the ARQ Timeout, and is referenced to the time when the packet is first transmitted from the transmitting MAC. Thus, if a lost packet has not been successfully retransmitted within an ARQ Timeoei€KCut of its initial transmission, it is discarded by the transmitting MAC, but is considered to have been positively acknowledged for the ARQ Window Size rule. (c) Whether a packet received after another packet of a higher sequence number has been passed up to the higher layer will be discarded by the receiving MAC or continued to be passed up need be explicitly stated. (d) The definition of the ARQ Window Size should not be expressed in terms of receipt of a NACK, as a NACK may be returneDd by the recipient for a lost packet, by itself lost on the way to the sender and hence not received by the sending MAC. It should be stated in terms of the sliding window concept. Thus, the ARQ Window Size is defined to be the number of packets the transmitting MAC may send ahead of, but including, the lowest-numbered packet for which a positive ACK has not been received. (e) There need be a rule specifying whether a packet received before another packet with a sequence number at leashEXDECISION OF GROUP @ B J U_isACE@ @ACOAA M N J!I"A#N$O%I&"I''M(,I)1A*C+B,4A-8N.:N/?E0BA1FH2JB3LA4A5OA6PN7UA8XN9[O:^M;`AgN?iA@jOAmBoOCqFDsIEuAFwNGzAH}OI€HJ€EL€ JM€ ON€MO€NP€NQ€NR€#OS€'ETBU€)AV€,NW€-AX€/AY€0NZ€2A[€4O\€6A]A^€9N_€<A`€>Ha€@Ib€EFc€JJd€KAeCfEiGjj@K!@g€K€NNk€PNl€TNm€XNn€YBo€ZOp€^Lq€bFr€fNs@K!@fj€KEt the ARQ Window Size smaller is pending retransmission will be discarded by the receiving MAC, or whether any lost packet with a sequence number at least the ARQ Window Size smaller than a recently received packet will be considered by the receiving MAC to have been discarded by the transmitting MAC due to the Retransmission Timeout limit and hence to have exceeded the ARQ Timeout limit at the receiving MAC. In the case where a received packet of a higher sequence number is discarded dueF to receive buffer overflow, further specification is required of whether the corresponding acknowledgment will be a positive, a negative, or no ACK at all. (f) In a case where a station (BS or SS) has the right to access the channel but has reached its ARQ Window Size limit--a case likely to occur when bitmap-type acknowledgments covering ranges of transmitted packets are lost--it need be specific whether that station will be using the access right to retransmit those packets not yet posik[€KGitively acknowledged from the sender's standpoint. (g) The range of sequence numbers required for ARQ purposes need be larger than estimated by the bandwidth-delay product, since the recipient usually cannot acknowledge a packet immediately upon its reception. If fact, the sender may be allowed to have a long transmit opportunity before any other station has its turn for transmission.  OaThe ARQ algorithms as defined on pages 189-208 do not lead to unambigious operational procedures.P54€LB 2001/07/04EKenFStanwoodH204I 2001/07/02J EditorialK189L1NChange "HBS-RFMM" to "HBS-DRFM"€MB 2001/07/04EKenFStanwoodH205I 2001/07/02J EditorialK189L30NUWhy put this in a section 6.2.12 when there's already a section 6.2.4 waiting for it?€NB 2001/07/04ERobertFNelsonH206I 2001/07/02JTechnical, Non-bindingK189L33M6.2.12jl€NNA revised/enhanced ARQ has been submitted for Session #14. If accepted many portions of this and other ARQ-related sections will be affected. The submission is document IEEE 802.16abc-01/01.AThe ARQ material in the current working document provides a packing and serialization algorithm and a collection of data structures with the intent of supporting multiple algorithms. The intent to support multiple algorithms dilutes efforts to develop the standard and ultimately diminishes the effectiveness of the standard in practice. Also, some constructs included in the ARQ material appear to conflict with the design philosophy of the current MAC definition and may impact the flexibilitBoy available to designers implementing the standard. Finally, general algorithms for ARQ operation are outlined.]€OB 2001/07/04EKenFStanwoodH207I 2001/07/02J EditorialK189L40NChange "it peer" to "its peer"€PB 2001/07/04ELeiFWangH208I 2001/07/04JTechnical, Non-bindingK190L1M6.2.12.1km€PNYin all ARQ TLV tables on pages from p190 to p193, well define the Type and Length fields.O1. different parameters should have different type values. 2. the length field should be 2 bytes if it is of 16 bits, not 4 bytes.€QB 2001/07/04EKenFStanwoodH209I 2001/07/02J EditorialK190L3N6.2.12.2 through 6.2.12.6 should be further indented subsections under 6.2.12.1. The TLV definitions protion should be a subsection of 11.1. the message definition portions should be a subsection under 6.2.2.4.€RB 2001/07/04ELeiFWangH210I 2001/07/04J EditorialK190L12M6.2.12.2Nadd the definition of SFES?€SB 2001/07/04EKenFStanwoodH211I 2001/07/02J EditorialK190L14N"SFES" is not in the acronym list. It apparently stands for "Service Flow Error Parameter Set", which is not obvious from teh acronym.€TB 2001/07/04EKenFStanwoodH212I 2001/07/02JTechnical, Non-bindingK190L36ln€TNThe type codes in sections 6.2.12.2 through 6.2.12.6 should be properly defined. Appropriate values should be obvious once the TLVs are moved to sectio 11.1.€UB 2001/07/04EKenFStanwoodH213I 2001/07/02JTechnical, Non-bindingK190L36NIn sections 6.2.12.2 through 6.2.12.6, the length field defines the TLVs to be 32 bits, but the value field defines them to be 16 bits, yet uses at most 8 bits (except for 6.2.12.5 which uses 16 bits). Which are they?€VB 2001/07/04EKenFStanwoodH214I 2001/07/02JTechnical, Non-bindingK190L53NzThe Request/transmission Policy TLV in section 11.4.11.17 should have bit #7 used to indicate ARQ on/off for a connection.€WB 2001/07/04ELeiFWangH215I 2001/07/04JTechnical, Non-bindingK190L59M6.2.12.3N?define the term "packet", which is the unit of ARQ window size.O2The unit of ARQ window size shall be well defined.€XB 2001/07/04EKenFStanwoodH216I 2001/07/02J EditorialK193L50mo€XNyThis section should be part of your document's section 6.2.12, which really should be section 6.2.4 in the main document.€YB 2001/07/04EVladimirFYanoverH217I 2001/07/03J EditorialK193L50M6.3ARearrange the TOC for ARQ section the following way: 6.2.4 ARQ Mechanism 6.2.4.1. Formats and Data Structures 6.2.4.1.1. General 6.2.4.1.2. Block Numbering Based Retransmission Scheme 6.2.4.1.2.1. ARQ Blocks 6.2.4.1.2.2. Transmitter Operations: MAC Message Creation and Numbering 6.2.4.1.2.3. Receiver Operations 6.2.4.1.3. MPDU Numbering Base Retransmission Scheme 6.2.4.1.3.1. Transmitter Operations: MAC Message Creation and Numbering 6.2.4.1.3.2. Receiver Operations 6.2.4.1.4. Formats of Anp€YBRQ Related Signaling 6.2.4.1.4.1. Encoding of the Sequential Number 6.2.4.1.4.2. Encoding of the ARQ Feedback in Sub-headers and Partial Payloads 6.2.4.1.4.3. ARQ Feedback Information Elements used in Sub-headers and Partial Payloads 6.2.4.1.4.4. Encoding of the ARQ Feedback in MAC Messages 6.2.4.1.5. Additional Operations 6.2.4.1.5.1. Format of Discard and Reset Messages 6.2.4.1.5.2. Discard 6.2.4.1.5.3. Reset 6.2.4.1.6. Retransmission Algorithms 6.2.4.1.7. Examples of ARQ Schemes 6.2.4.1C.7.1. MSDU/Fragment- oriented Scheme 6.2.4.1.7.2. Byte pointer-based Scheme 6.2.4.1.7.3. MAC Message-oriented Scheme 6.2.4.1.8. ARQ Related Operations at the Connection Creation/Change 6.2.4.2. ARQ Algorithm O1. The structure of ARQ section in the considered document is erroneous . 2. The location of ARQ at section 6.2.4. is defined by the original 802.16 document (D3-2001) €ZB 2001/07/04ELeiFWangH218I 2001/07/04JTechnical, BindingK193L57M6.3.1N!change "obligatory" to "optional"oq€ZO*allow more flexibility in implementation. €[B 2001/07/04EKenFStanwoodH219I 2001/07/02JTechnical, Non-bindingK195L14NThe word "Message" should be changed to "PDU" or whatever is correct. "Message" is ambiguous and will lead to confusion. This applies throughout this section.€\B 2001/07/04EKenFStanwoodH220I 2001/07/02JTechnical, Non-bindingK198L1NThis figure doesn't make sense. If fragments 1 and 2 are sent in the same PDU, then they shouldn't have been fragmented relative to each other. The choice of FC codes won't properly rebuild teh original SDU. €]B 2001/07/04ELeiFWangH221I 2001/07/04J EditorialK198L1MFigure 4N1. Change "partial payload Frag 1 of MSDU #1 FC=01" to "partial payload Frag 1 of MSDU #1 FC=11" 1. Change "partial payload Frag 2 of MSDU #1 FC=11" to "partial payload Frag 1 of MSDU #1 FC=01"OHbased on TG1 document, FC=11 : continuing fragment FC=01 : last fragment€^B 2001/07/04EKenFStanwoodH222I 2001/07/02JTechnical, Non-bindingK201pr€^L9NThe ARQ sub-header is not sufficient to replace the Fragmentation sub-header. The Packing subheader and the fragmentation subheader have the same size FSN field, allowing the continuous flow of sequence numbers whether packed or not.€_B 2001/07/04EKenFStanwoodH223I 2001/07/02JTechnical, Non-bindingK201L62NThe packing subheader should not be corrupted, causing it to be inefficient for non-ARQ connections simply because someone wants to implement ARQ using layer violations.€`B 2001/07/04EKenFStanwoodH224I 2001/07/02JTechnical, Non-bindingK202L19N%Where's the TLV for this negotiation?€aB 2001/07/04EKenFStanwoodH225I 2001/07/02JTechnical, Non-bindingK202L31AThe fact that multiple IEs are in the ARQ "sub-header" shows that it's really an ARQ message. Forcing it as a subheader of some unrelated connection's PDU rather than sending it as a messages on the CPE's Basic or Primary management connection is a blatant layer violation.€bB 2001/07/04EKenqs€bFStanwoodH226I 2001/07/02JTechnical, Non-bindingK204L37N&Figure 6 i sa blatant layer violation.€cB 2001/07/04EKenFStanwoodH227I 2001/07/02JTechnical, Non-bindingK205L12AnThe shown "Packing Subheader" breaks the above 11 GHz system, as well as breakinf packing on connections that don't have ARQ. If you step back from insistence on solving ARQ through layer violations, you'll find that this is not necessary. If you insist on layer violations, at least have the courtesy to use another type code to indicate your own sub-header type.n€dB 2001/07/04EKenFStanwoodH228I 2001/07/02J EditorialK209L7NWThe basics of section 6.5 should be part of 6.2.7. 6.5.2 should really be in section 3.€eB 2001/07/04EKenFStanwoodH229I 2001/07/02J EditorialK210L6NChange "then" to "than"€fB 2001/07/04EKenFStanwoodH230I 2001/07/02JTechnical, Non-bindingK210L11r;€fNrBased on the described usage, there is no need to make special messages, just special usage for existing messages.€gB 2001/07/04EKenFStanwoodH231I 2001/07/02JTechnical, Non-bindingK210L18N_If an SS missed the map messages, how does it know when to listen for these other map messages?€hB 2001/07/04EKenFStanwoodH232I 2001/07/02JTechnical, Non-bindingK211L12AThis is less efficient than the way DL burst profiles are handled. For DL burst profiles, rather than eating up bandwidth sending statistics to the BS, the SS monitors the DL itself, and only sends a message when a change needs to be made. Can the same paradigm be applied here?€iB 2001/07/04EKenFStanwoodH233I 2001/07/02J EditorialK211L12NQMessage definitions should be in section 6.2.2.4, TLVs should be in section 11.1.