WPCh 2BJ ZCourier#|d,+BoldTimes Roman@K@Apple postscriptHPLAIIP0.PRSd6X@8;hhhhm@2, < #|d,+t IIIon LPT1HPLASIII.WRSd6X@8;,,,/@CourierCourier Bold@^DNkttϴNNNtNNNNttttttttttNNgϨNZӬ|ۨNNNtttgtgtgNttAAtAttttVZAttttggtgtDD'ttt`tttgggggϛgggggNANANANAtttttttttDtgtttttttttggttttttttttttttttZtA`tttDDV`|ZDDNDkttNg__ttDtttRtttRNNDDDDDDDDDDDDDNRtNπ捍:t@t__CzsffzDNkttϴNNNtNNNNttttttttttNNgϨNZӬ|ۨNNNtttgtgtgNttAAtAttttVZAttttggtgtDDDDDDDDDDDDDDDDDNNDDDDDDDDDDDDDDNttttttttt_NtttRRttktRt_gϟNNNNϨtgggggggggggAAAAtttttttttttttt2 <LR2  CourierCourier BoldTimes RomanTimes Roman BoldHelveticas4ddd"d `K@N:{p\  PCP@N:!lp4  p(ACDDDDDDDDDDNRtNπ捍:t@tccCzsffzDNkttNNNtNNNNttttttttttNNtϨZpۨ騨NNNttttggRtANAtgZNtttgttttDDDDDDDDDDDDDDDDDNNDDDDDDDDDDDDDDNtttttttttcNtttRRtttRtct񨛛ZZZZϴttttttgggggAAAAtttttttt"S^:N_uu¶NNNu:N:AuuuuuuuuuuAAhרN[ϨܨNANmuNhuhuhNuuAAuAuuuuN[Auuuuhp/p~N:NN'NNNNNNNNNNuAhhhhhϜhhhhhNANANANAuuuuuuuuuuhuuuuuhhhhhhuhhhhuuuuuuNuNuNNNu[NuANNAuuuuuϨNNN[[[[NNuuuuܨuhhNR:juNhuuuuu@HuFuF*hhuNNuuNhFt捍:]@s:`1uxxxxxxxx;C\ccCCCcCCCCccccccccccCCDZY}}vCM}rk}CCCcccYcYcYCcc77c7ccccJM7ccccYYcYc;;;;;;;;;;;;;;;;;CC;;;;;;;;;;;;;;CcccccccccQCcccFFcoc\cFcQY}}}}CCCCrcYYYYYYYYYYY7777ccccccccccccccCourierCourier BoldTimes RomanTimes Roman Bolds (WN)CG Times (WN) (Bold)Univers (WN)Univers (WN) (Bold)CG Times (WN) (Italic)HeadingChapter HeadingJ d  ) I. ׃  2 t vRight ParRight-Aligned Paragraph Numbers>a݅@  I.   X(# BibliogrphyBibliography݅X  (# SubheadingSubheading0\ E A.  a8DocumentgDocument Style StyleXX` `  ` 29pNkk) a4DocumentgDocument Style Style . a6DocumentgDocument Style Style GX  a5DocumentgDocument Style Style }X(# a2DocumentgDocument Style Style <o   ?  A.  2 vk  { a7DocumentgDocument Style Style yXX` ` (#` a1Right ParRight-Aligned Paragraph Numbers :`S@ I.  X(# a2Right ParRight-Aligned Paragraph Numbers C @` A. ` ` (#` a3DocumentgDocument Style Style B b  ?  1.  2Xa3Right ParRight-Aligned Paragraph NumbersL! ` ` @P 1. ` `  (# a4Right ParRight-Aligned Paragraph NumbersUj` `  @ a. ` (# a5Right ParRight-Aligned Paragraph Numbers_o` `  @h(1)  hh#(#h a6Right ParRight-Aligned Paragraph Numbersh` `  hh#@$(a) hh#((# 2a7Right ParRight-Aligned Paragraph NumberspfJ` `  hh#(@*i) (h-(# a8Right ParRight-Aligned Paragraph NumbersyW"3!` `  hh#(-@p/a) -pp2(#p Tech InitInitialize Technical Style. k I. A. 1. a.(1)(a) i) a) 1 .1 .1 .1 .1 .1 .1 .1 Technicala1DocumentgDocument Style Style\s0  zN8F I. ׃  2Cxa5TechnicalTechnical Document Style)WD (1) . a6TechnicalTechnical Document Style)D (a) . a2TechnicalTechnical Document Style<6  ?  A.   a3TechnicalTechnical Document Style9Wg  2  1.   2Q?a4TechnicalTechnical Document Style8bv{ 2  a.   a1TechnicalTechnical Document StyleF!<  ?  I.   a7TechnicalTechnical Document Style(@D i) . a8TechnicalTechnical Document Style(D a) . 2&3*! e.%!%Doc InitInitialize Document Stylez   0*0*0*  I. A. 1. a.(1)(a) i) a) I. 1. A. a.(1)(a) i) a)DocumentgPleadingHeader for Numbered Pleading PaperE!n    X X` hp x (#%'0*,.8135@8:@.SSxSSJJSJS1111111111111111188111111111111118SSSSSSSSSD8SSS::S\SLS:SDJxxxxxxrhhhh8888{{xxxxxx{{{{x_SJJJJJJoJJJJJ....SSSSSSSSSSSSSS"S^2CRddCCCdq2C28dddddddddd88qqqYzoCNzoozzC8C^dCYdYdYCdd88d8ddddCN8ddddY`(`lC2CC!CCCCCCCCCCd8YYYYYYzYzYzYzYC8C8C8C8ddddddddddYddddodYYYYYYdzYzYzYzYddddddCdCdCCCdNCdz8zCzCz8dddddCCCoNoNoNoNzCzCdddddzYzYNF2[dCYddddd7>d<d<$YYdCCddooCYd<d<+oodCCddddCo&&&s&&&&&&&&&&: S3S3S3S3S3fMM3F3F3F3F3& & & & S:S:S:S:S:S:S:S:S:S:S3S:S:S:S:@:S3S3M3M3M3M3S:F3F3F3F3S:S:S:S:S:S:&:&:&&&:-&S:F F&F&F S:S:S:S:S:fSM&M&M&@-@-@-@-F&F&S:S:S:S:mSS:F3F3N(4:&3::::: $VV:#WW:V#33:s&&::q@@s&3s#tAA??AArEEx.A _R9/rxXR_XXXLc?1FS?O?:SS?[O 6NA<p 2p}wCL 1sC82:s2PkCXP  /xC8< \:x2p}wCX7tC2!vt4  p(ACX<5nC2K%n*f9 xCXX5j@0j\  PC?P2(bL\SZG"S^.=f\\===\i.=.3\\\\\\\\\\==iii\zzpG\zpfzz=3=k\=\fQfQ=\f3=f3f\ffQG=f\\\QH(H_=.============f3\\\\\QzQzQzQzQG3G3G3G3f\\\\ffff\\\\\\pf\\QQQQfzQzQzQzQ\\\\\fGfGfG=Gf\=fz3z=z=z3fff\\QQQfGfGfGfGz=z=ffff\zQzQN@.c\=\\\\\\7=\7\73\\\==\\ff=\7tiieeiioo.Ii2[-Kye1pe~efe~ 6NA<p 2p}wCL 1sC82:s2PkCXP  /xC8< \:x2p}wCX7tC2!vt4  p(ACX<5nC2K%n*f9 xCXX5j@0j\  PC?P 5o@0!o4  p(AC?<3i@0K"i*f9 xC?XS @&&4@\  PCYP2bZb_ZhTr5ddd#d6X@K@s4ddd"d `K@N:{p\  PCP@N:!lp4  p(AC7oC2bo\  PCXP  y.]8*]\  PCPL 9NA22PkCP 6NA<p 2p}wCL 1sC82:s2PkCXP  /xC8< \:x2p}wCX7tC2!vt4  p(ACX<5nC2K%n*f9 xCXX5j@0j\  PC?P 5o@0!o4  p(AC?<3i@0K"i*f9 xC?XS @&&4@\  PCYP2f=.Zf\  PC&P2j=.!j4  p(AC&S?[O W , a  # P7{P#September 1993`(#Bdoc: IEEE 802.1193/152     yxdddy KY  #Xw P7bXP##x PjQP#yxdddy   Submission2؃`(#61phat39g C. Rypinski LACE, Inc.K    U. #xP7P#IEEE 802.11 ă 802 LAN Access Method for Wireless Physical Medium   yx dddy #X~xP7XP#  R  DATE: ` ` D3 1, 4D   yxk dddjy  R  AUTHOR: ` ` Chandos A. Rypinski, ` `  Chief Technical Officer ` ` LACE, Inc.`&(#ETel: +01 415 389 6659 ` ` 655 Redwood Highway #340`(#EFax: +01 415 389 6746 ` ` Mill Valley, CA 94954 USA`(#BEm: rypinski@netcom.com  l yxdddy  Pm  TITLE:` ` #XxPʖQXP#NECESSARY PHY LAYER ALTERNATIVES(#` ` ` FOR ET BAND AND FOR "HIPERLAN" Ċ  ( yxddd>y   P)  Z  SUMMARY AND CONCLUSIONS  Xl #X PjQbXP#The present PHYs under development by the PHY group address the needs of and optimizes for spontaneous, autonomous groups. The result is unlikely to be useful for any public network compatible connection type service or for LAN comparable interactive communication. It is necessary and desirable to have parallel effort on two additional sets of premises:  X  ETSI HIPERLAN and 5.85 GHz ISM bands  X Premise: To provide high transfer rate and capacity/hectare services supporting bandwidthondemand connectiontype and connectionless services consistent with business inbuilding needs.  X5  ET band 1.851.99 GHz  Xx  Premise: 19101990 and 18501910+19301990 all with fully utilized 5/10/20 MHz bandwidth, and an attempt at common airinterface for licensed and unlicensed services consistent with voice and data service inbuilding or in semipublic or public places. ,''xx  XC  # #XxPʖQXP#NECESSARY PHY LAYER ALTERNATIVES  P  $FOR ET BAND AND FOR "HIPERLAN" ă  X   #X PjQbXP# Table of Contents`a !SPage ă    X44PURPOSEƤ4p!!T 1 X44X 44TABLE I POSSIBLE LAN FREQUENCY BANDSƤ p!!T 2 X44DESCRIPTION OF PROPOSED NEW PHY LAYERSƤ4p!!T 3 X44X 44Proposed PHY Description ETSI "HIPERLAN" and 5.85 GHzƤ 44 ISM bandsp!!T 3 X44X 44Proposed PHY Description ET bandƤ p!!T 3 X44X 44Comparison of Capacity and Premises for Three PHYsƤ p!!T 4 X44X TABLE II CAPACITY/HECTARE FOR VARIOUS PHYSƤ p!!T 4 X44X X ! 802.11 Proposed Frequency Hopping PHYƤ p!!T 5 X44X X ! 802.11 Proposed LBT Direct Sequence Spread SpectrumƤ p!!T 5 X44X X ! 802.11 Proposed Time and Code Division Spread SpectrumƤ p!!T 5 X44X X ! 802.11 Proposed Short Symbol Direct Sequence SpreadƤ 44  !Spectrump!!T 6 X44X X ! Optimization for a Fixed Narrow Bandwidthe.g.; 10 MHzƤ p!!T 6 X44CONCLUSIONSƤ4p!!T 7 X44REFERENCESƤ4p!!T 7  ,''xx  X      XX  # #XxPʖQXP#NECESSARY PHY LAYER ALTERNATIVES  PK  $FOR ET BAND AND FOR "HIPERLAN" ă  V (#҇ ZT X` hp x (#%'0*,.8135@8:E yO  ** #x PjQP#э,"FQPSK: A modulationpower efficient RF amplification proposal for increased spectral efficiency and capacity GMSK and !/4QPSK compatible PHY standard," Kamilio Feher, University of California at Davis, IEEE 802.1193/97, 7/13/93I suggest that a signaling rate of at least 12 Mbps may be obtained in the 10 MHz channel. Sufficient improvement on the basic channel might be developed by some combination of the following techniques: a),Multiple narrowly directive antennas at the accesspoint with combining type diversity (not selection). b),Embedded strong forward error correction channel coding combined with short segments implementing cell relay. .&*&*LL c),DD9Redundant coverage of station transmitters, and repeated or sequential transmissions to stations from different sources.(#D d),DD9Smart and timely retransmission of failed segment transfers.(#D  VB  Comparison of Capacity and Premises for  V# Three PHYs An approximate comparison of system capacity for the proposed PHYs and a few others is shown in the Table below. The capacities presented are very rough approximations for the voicebased systems since they are normalized for a 10 MHz frequency space. They do not include antifade measures beyond those in the prototype system. The effect of the cell size on this measure of capacity is much more significant than fine points of transmission technology. Inherently, all large area cover telephone systems show up poorly. 3:t 88.&*&*LLCA-&*&*LL8ԯ V N  Table II Capacity/hectare for Various PHYs ă System type $hh*-06CapacityMbps/hectare ,ETSI HIPERLAN:$hh*-065001000! 5.8 GHz ISM band (936):hh*-062 x 125250  V 5.8 GHz ISM band (9176):hh*-062 x 4 2.4 GHz FH (40 m reach):hh*-060.13 2.4 GHz DS (40 m reach):hh*-060.52 1.91.92 DQPSK NB+diversity (20 m reach):613  V 1.9293 GSM1 variant (70 m reach):-06.0022.009  V Qualcomm CDMA SS1 (.5 mi reach):06<.01 Note: The methodology used for the capacity estimate was given in IEEE 802.1193/101."#.&*&*LL  T ԇ802.11 Proposed Frequency Hopping PHY  V5 The proposed FH SS PHY2 now before 802.11 will not come within an order of magnitude of the necessary capacity and transfer rate. This due to some combination of causes including the following: a),Excessive emphasis on ad hoc autonomous mode losing the radio propagation advantages in reach and reduction in multipath available from use of a shared privileged antenna. b),Use of frequency hopping preserving the disadvantages of narrow band systems with respect to Rayleigh fading. c),Use of frequency hopping which does not provide in the limit as much capacity as if the available channels were used parallel without hopping. c),Use of a frequency band where interference is a certainty in urban areas making a significant performance loss inevitable. The modulation is optimized against unrelated interferers where high capacity systems must be optimized for resistance to interference from other stations in the same system. d),The presumption that overpowering transmitters will assure adequate performance on obstructed radio paths thereby assuring large interference range and low reuse factor. An example of the effort required to make digital communication work in the 2.4 GHz band in a recent, excellent and credible  V. contribution&>E yO  ; ** #x PjQP#э,"An RF Data Transport ProtocolThe RF Adaptation sublayer and RF Physical Layer Specifications for Slow Hopping Spread Spectrum Radio LAN," Ed Geiger, Apple compute, Cupertino CA, IEEE 802.1193/104& by E. Geiger of Apple. The technique combines the use of frequency hopping in a way that minimizes the effect of many spectrally narrow interferers, segmented transmissions and XY forward error correction of segments to greatly reduce the incidence of unusable transfers. The need for this type of function in narrowband systems has been overlooked in all other 2.4 GHz PHY proposals.(.&*&*LLԌ,DD9This work is a warning that a simple narrow band channel cannot produce reliable communication without the artful use of a considerable fraction of the transfer capacity for redundancy and error correction.!D  T 802.11 Proposed LBT Direct Sequence Spread  T Spectrum  V The proposed DS SS2 plan is reasonable, but divides the capacity and spectrum used into three parts to support parallel independent systems. The peak transfer rate is then limited to onethird of what it might be with time rather than frequency division sharing. The presumption of a listenbeforetalk access control makes inevitable a poorer use of channel time than is possible with a central manager. The reuse factor for the assumed autonomous groups requires a much high reuse factor (1625) than would be needed with privileged antennas at access points and an organized infrastructure.  T 802.11 Proposed Time and Code Division  T Spread Spectrum The code division SS plan of Dr. J. Y. C.  V Cheah yO  [#** #x PjQP#э,"A Proposed IEEE 802.11 Radio LAN Architecture," Jonathon Y. C. Cheah, Hughes Network Systems, San Diego, IEEE 802.11/917 (Pages 1318 show 31bit 12vector correlation diagrams) has been offered but is not currently advocated. He proposes that with a 31 chip symbol, there are at least 12 codes available that are sufficiently orthogonal to enable overlapping coverage to be resolved by code discrimination. In the 2.4 GHz band, this modulation would allow at least a 2 Mbps rate (without use of quadrature phase) and possibly double that. There would be far less loss from self interference with code discrimination than with protocol recovery from detected collisions used in the LBT uncoded DS SS. This plan was originally intended for 2.4 GHz, but in this discussion has been normalized to the bandwidth and other assumptions for use of the 5.85 GHz ISM band.8'.&*&*LL(+*&*&*LL8  T 802.11 Proposed Short Symbol Direct  T Sequence Spread Spectrum A direct sequence plan proposed by this contributor uses a short symbol length to attain a high megabit data transfer rate. The spreading is used solely to mitigate multipath propagation and Rayleigh fading further extending the advantage inherent in wider bandwidth. As compared with the three plans above, this is the only PHY which has set out with high transfer rate and capacity per hectare as a primary goal. Others have focused on minimizing infrastructure cost, or on using existing technology; and then arguing that the capacity provided is adequate. While the offered technology is arguable (though there has been little discussion of the technical detail), the service and functional goal is both fundamentally different from that of any present 802.11 2.45 GHz PHY proposal. Arguments against use of 5.85 Ghz by reason of cost or power drain are simply not valid in the context of short reach. This is less of a cost obstacle than dealing with narrowband and microwave oven interference at 2.5. While the final choice might be considerably different than this proposal, it may also be the best available starting point. The high capacity/function capability is both needed and feasible.  T Optimization for a Fixed Narrow Bandwidth T e.g.; 10 MHz Regardless of technical desirability, maximized use of a 10 MHz bandwidth is and will be a reality. This constraint may be the one that lead the development of common technology for licensed and unlicensed operation. It is imaginable, that on private premises infrastructure would be provided (financially and administratively) by the operator of the premise; and in public places (airports, convention centers, hotels, hospitals), by a service provider. Rights to use of air spectrum(.&*&*LL space would be defined in the same patterns as property lines. The channel switching feature is not necessarily part of the access method, but rather a system service access selector. It would be desirable to have the technique upwardly and downward scalable 5 and 20 MHz with proportional changes in capacity. The same protocol might be used independently of rate. A further need is a common air interface within the common bandwidth. This contributor believes that the commonality between voice and data channels is best achieved with a common cell relay definition (as in ATM). However, it is also possible to have a common chipping rate and different symbol definitions for data and voice. Long symbols would maximize range at low bandwidths, and short symbols would maximize transfer rate for sequential multiplexing. Much of the work and many of the contributions on modulation types done in support of frequency hopping is equally applicable to this problem. ,DD9The computer data communication community must accept that service provider compatibility is desirable to users. This achievement will widen market and usage.!D Users offpremises may be willing to pay for service. Given that opportunistic pricing will exist as it does for hotel telephone calls, the data service may still be cheap if the charge is based on data transferred rather than open connection time. The dogma that they won't must be abandoned, and replaced by an effort to make the cost reasonable. 8%.&*&*LL)YU-&*&*LL8  V  CONCLUSIONS The present PHYs under development by the PHY group address the needs of and optimizes for spontaneous, autonomous groups. The result is unlikely to be useful for any public network compatible connection type service or for LAN comparable interactive communication. It is necessary and desirable to have parallel effort on two additional sets of premises:  .&*&*LL  V  ETSI HIPERLAN and 5.85 GHz ISM bands Premise: To provide high transfer rate and capacity/hectare services supporting bandwidthondemand connectiontype and connectionless services consistent with business inbuilding needs. Capacity objective: 100 Mbps/ha minimum Transfer rate objective: 16 Mbps minimum.  V  ET band 1.851.99 GHz  V  Premise: 19101990 and 18501910+19301990 all with fully utilized 5/10/20 MHz bandwidth, and an attempt at common airinterface for licensed and unlicensed services consistent with voice and data service inbuilding or in semipublic or public places.8J .&*&*LL e i[&*&*LL8ԯ  V  References: