脉冲位置调制光谱切片传输中英文翻译

Pulse position modulation for spectrum-sliced transmission

Proponents of ultra-wideband (UWB) technology promise todeliver large amounts of data with very lowpower spectral density. The ultra-wideband radio concept is very attractive as it promises to open large amounts of spectrum to a variety of user sand at the same time it claims little interfere nice between users. Unlike conven tional wire less communications systems that are carrier based,UWB-based communications is baseband .It use sase ries of short pulses that spread thee nergy of thes ignal from near DC to a few GHz. One typical technique is to as sign a window in time and shif the position of the pulse with in that window. This is classical pulse position modulation. With band width restrictions effectively removed, UWB promises to speed upwire less data trans ferrates. Some pub lished work has addressed the issue of how many users can the UWB channel support[2][3].This work considered one type of capacity-the capacity interms on the number of users. The capacity was derived from signal-to-no is eratio (SNR)consi derations. Acer tain SNR is required to achieve a specified biterr or probability(BEP).The no is eflo or raises with the number of users.

A more fund amentalty peofcapacity is the well known Shannon capacity in bits/s. Our go a list odevelopanunderstanding of thero leofvar iouspara me terso nthe Shannoncapac it yo fUW

B commun ications.The well known Shannon!s form ulC=Wlog(l+SNR),where SNR is the 2 signal-to-noiseratio and Wist hechannel band width, predicts the channel capacity Cfor an additive white Gaussiannoise (AWGN) channel with continuousal uedinput sand outputs. This express iondoesnot apply to communications with PPM signals. PP mis a modul ation with discret e-valuedi nputsan dcon tinu ous-val uedoutput s.Further more,P Pmsignalsa reorthogo nal,whichi mposesa naddition alconst raintont hecapacityc alcul ations. Newcapacitycal culations arerequi red,which htakeint oaccountthes econstrain ts.PPMcap acitywas studiedint hecontex tofwireles sinfra reda nd optic alcha nnels[4].The capacityo fPPM modul ationservesas thestarti ngpointtow hichUWB- speci ficcons traintsar eadded.Thec onstraints arethepow erspectr umdensityl imitation underF CCPar t15r ulesand thes preadingra tiocons train t.Sincegiga hertzunoc cupiedslice sofbandwidth arenotavailable atmic rowavef requencies,underFCCreg ulations,U Wbradiomust betreat eda ss p uriousint erferenc etoallother communicat I

onsystems.

The FCC is currently assessing whether to allow UWB emission son an unlicensed basis. Un in tend edout of bande missionsa regov ernedb yFCCPart15ru les.Inthisp aper,weassumetha tUWBtransmiss ionscomplywi ththisFCCrul e.Thespreading ratioconst raintisinte ndedtosatisfyi nterferences uppressionre quirements[5 ][6].Thispape risorgani zedasfollows. Sectionintrodu cesthesi gnalmodel.The linkbudgetcalculati onforUWBunde rFCCPart 15rules isdevel opedinsectio n.AnalysisofU WBM-ary PPM capacity is carried out insection IV.Finally,conclusionsarep rovidedinse ctionV..SIGNALMODEL Considera single-user UWB pulse position modulated communication ssystem. The M-ary PPM signal setis {S1(t),S2(t),SM(t)}, where Sm(t), (mM) Canbewrit tenas: (t)=p(t-j). Inthisexpr ession, p(t)is the UWB puls eofduration Tp (normalizedsuchthat), Episthee nergyperp ulse,Tis the puls erepetition time, thesequencecj frepresents thet imehoppi ngcode,withcj Tanadditional times hiftto the jthpul seof the ccomm unication burst. Fora fixedT ,the symbolrate R=1/(N,Tf) determinesN, the fnumbero fpulsestha tformasymbol. The symbolduration is then T,=N,Tf. The parameter is the additional times hift corresponding to the pulseposition modulation, mwith T=0, T100.

Theband widthrequ irementso fM-aryPPMca nbeestim atedfromth eLandau- Pollak theorem[8], which hstates that the band width of as ignalt imelimitedto T,(symboldu ration)andof dimensionali tyMisappro ximatelyB = .Thisval uecanbeint erpretedasthe minimumbandw idthrequire dtotransmi tuncodedM -aryPPM.UWB using Mary PPM is spread spect rumin thesense that theUWBbandwidthW>>B.From (2),W=c/Tp,andT,=NT,the spread spectrumcondition Wsf>>Bismet for aspread ingratio=>>M.Themot ivationfor spreadspe ctrumUWBPPMis twofold:(1)atta inspecifie dleveloftr ansmittedpower subjecttopower densityconstra intssetb yFCCPart1 5regulat ion,(2)attai nspecif iedlev elofresist anceto othe remissi onsintheb and(jamresistance).The transmitted powerre quirementis deter m ine dbythedis tanceb etweenthe transmitte randthere ceiver.Thi sissueisdi scusse dinthenextse ction.Jamre sistance of U WB isanalyzed .

UWB LINKB UDGET InMay 2000,the FCC published aNot ice of Proposed RuleMaking(NPRM) to include UWB emissionsunder Part15 rule.In the notice,the FCC tentatively proposed that UWB emissions above 2GHz,shoul dcomply with the existing Part 15limits,and that UWBemissionsbelow 2GHz should beat tenuate datleast12dB below the Part 15 limits.Inthissec tionofthe paper,weassess the range of communication using UWB by carrying out linkbudget calculation sassuming compliance with FCC Part 15 rules.The commonlyused link budgete quation canbeexpressedas:,3 where Ptistran smitted UWB power,Pisreceived power at distanced.Gtand Garean tennagain sfor transmitt erandrec eiverre spectiv ely,bot hassumed as0dB.Thee xpression ofpathlo ssPLtobeus edin(3)depen dsonthepr opagation model.Currently,there is nocommonly accepted model for UWB.In[9],it was shown that in an environment dense with obscuring objects,there ceivedsignalstreng thapproxi mately-4followsd-dependency with distanced.At short range suchasseveral meters,the-2trans missioni sdominatedby the direct signal path(ddependency).Thegeneral UWB path lossmode lcanbe written:PL(dB)=10log ,where nis the power at tenuation exponent and Xis the wave leng thcorre sponding to the working frequ encyfc.The working frequency is the center of band width,andit is assumed to be2.5GHz for a UWB signal of bandwidth W=1Ghzoccupying the frequency range from2GHzto3GHz.Acco rding to FCC Part 15 rules, em is sionsab ove900Mhzshallnot exceed field strength levels of E=50 0micro-volts/meter/Mhz measure data distanceof3elation ship between Pta nd Ecand epen donanu mberofaddi tionalfac tors,acom monly-usede xpressiontoap proxim

atetheirre lationship is[10]:.5From(5)andafterso mecalculat ions,weobtain thetransmitte dpowercons traintf oraUWBtra nsmission overa1GHzb andwidthi s.6Toproper lydetectan ddecodethere ceive dsigna l,there ceiverreq uiresa certain minim umS NR.Ifw econs ideronlyt hermalnoisea stheprimar ysourceo fin terfer ence,therece ivednois epowerNca nbecalcu latedf rom:N=Ktwf,7whe rekistheBo ltzmann! sconstan1.3 8xJoules /K,Tisth eroomtem perature( typicallytake nas300K) ,andFisth enoisefigure (optimisti cally)assum edF=5dB.F oraUWBrec eiverwithband widthW=1GHz ,thecalcul atednois eflooris- 89dBm.Unl ikeotherc ommunica tionsmetho ds,which transmitc ontinuouss ignals(100%dutycycle)and thepeakpower equalsthea veragepower,with UWB,the pulseduration Tpismuc hshorter thanthepul serepeti tiontim eTf.Thesp readingr atioP,which isdefined asth eratioTf/Tp,I stypicallyl argerthan100,resu ltinginapu lsepeakpo werhundred softimelar gerthant heaverage power.TheSNRp ersymbolatt heinputofth ecorrelation receiverisex pressed=2E,/No,wher eE,istheener gypersymbol andNoisthenoi sepowerdensity.Thes ymbolSNRisrelat edtotheaver ageSNRasfollows.Theav eragesigna lpowerisgiv enbyE,/T,Wit hW=and= ,weha vethefollowing expressi onfortheav erageSNRatthe receiver:SNR= ,8wher eg=2cN.Thepar amete rgplayst heroleo fproces sing gain(ratioofthesymbolsSN Ravail ableatdet ectiontot heaverageSNR atther eceiv erinput).Itfollow sthatthere quiredsen sitivity ofthere ceiveri sgivenby:S=N+F+SN ,9whereSNR is there quired signal tonoisera tioan dallquant itiesa reexp ress edindBmD ordB.Inord ertogetthe desiredperf ormancea tadistanced, theminimum receivedsig nalpowershou ldbeequalorl argertha nthesensiti vity,viz.,d S.From(3)a nd(9),wecangetr elationsbetw eensystempar ameterssuc hasthespre adingratio pandthe maximumcom munication srange &ax.Fig.2demo nstrates thetrade-off betweendatar atea ndra ngeofanuncode d,1GHzUWBcomm unicationsyst emutilizin g2-PPMmodul ation.Thefigu reshowsth emaximumrang e(subjec ttoFCCPart15 emissionres trictions)atwhi chthereq uiredsign altonoisera tioSNRisstil lmetasafunct ionofD-6thed atarate.TheS Nrrequ irementisco mputedforBEP=1 0Thecurvescorre spondDtodif ferentpa thlossfa ctors.Notet hatinor dertomainta inth esameBEPas therangeincr eases,th eprocessingg ainghastoi ncrease.Sinc eg=2cNandfor afixedSband width,thislead stolowerd atarate s(either through higher frame duration Tftoincrease== ,or through increased Nsnumber of UWB pulsesper symbol).For example,utilizingan UWB band widt hof1GHz,arat eof25Mbit/sis availabl eat20mforap oweratt enuationf actorofn=4oratar

angeof490mforaf actorofn=2.

While this sectiondeal twiththel inkbud get foruncoded communications,the nextsection discuss esthe capacity of UWB communications.IV.UWBCHANNECLAPA CITY The well known Shannon!s for mulC=Wlog(l+SNR),where C is the capacity in 2bits/s and Wis the band width,applies tot he AWGN channel withcon tinuous-valued input sandoutputs.This is not thecase forPPM-UWB,which has adiscrete-valued input and acontinuous-valueoutput.The capacity of PPM or thogonal signal sover the AWGN channel was computedin.

There feren ceserves asastarting point in the computation of the capacity of UWB signals.58 For the discrete-valued input and continuous-value dout put channel shown in Fig.3,akk-bit in forma tion source U=(U ,U ,U ,!-,ismapped to aM=2PPMsignalsetX=123kX ,X ,!-,,wheree very signalX ,can be represented asan M-dimensional vector 12Mmx ,.The signal constellation of Xisin terpret edasacollection of points in mM-dimensional signal space with one point located one achcoor dinatea xisatadi stance of a from the origin.The vector representation of X,is given by x=(0,!-,0 ,0,!-,0.mThis model holds foranyse to fM-aryor thogonal signals,with energy Eperdimension alssymbol.Capacity is the maximumamoun to finformation that canbetransmitted reliably and is given by C=maxI(Y,U),where I(Y;U) is the mutualinformation between the pUchannelout put Yandthechannel inputU,p(U)is the M-arysource probability distribution.Since Xis aninvert ible function of U,the capacity can be expressed C=maxpI(Y,U).Inaddition,since M-ary PPM consists of symmetrical or thogonal signals,X capacity is achieved with auni form sourced is tri bution,pX=x=1/M ,form=1, ..,M.m When such signals are transmitted over the AWGN channel with two-sided no is espectral density= ,the capacity Casa function of the channel symbol SNRM-PPM pisgivenby[4]][bits/symbol],where the randomvariab lesvm=1!Mhave the following distri but ion conditional mon the tran smitted signal x1(duetothesymmetryofM-aryPPM,anysignal can serve tocondition the output):The symbolN(m,1)de notes the Gauss iandistri but ionwi thmeanman dvariance1.

The channel symbol SNRp thusserves thero leof after-demodulation SNR.Tomake(10)applicable to UWB,it is necessary tocust omize the expression of pfor UWB.From(3),(4)and(9),under FCC Part 15 rules (constrained transmitted power asper the previoussection)the SNR at distanced from the UWB transmitter can be

written:SNR,11where ktisaconstant that dependsonthetran smitted power constraint,receiver antennagain,and center frequency.Tomeet there quired per formanceat distanced,SNRSNR .DThesymbolSNRp,can be expressed as a function of max imumcommun ication sdistanceby letting the signal to no is eratiobe the smallest allowed,i.e.,SNR=SNRD.It follows from(8),=2c ,12 where all quantities were defined previously.Fig.4presents the capacity inbitsper PPM symbol of UWB as a function of the symbol channel SNR as per(10)for various number of levels M.The curves were obtained by Monte Car lorunsof(10).For reference,the point so funcoded M-ary PPM corresponding to abiterror probability of areal so shown.From the figure it can be bserved that the gap between uncoded M-ary PPM and capacity is 3-6dB.

This gap is the improvement that can be obtained with coding.These result sarevalid for the AWGN.The spectralefficiency inbits/s/ Hzcanbeob tained from(10)as follows:59==[bits/s/Hz].13Note that increasing the spreading ratiop has the effect of reducing the spectral efficiency.Fig.5and6were generatedusing(13)incon junction with(12)and show the upper boundon the spectral efficiency as a function of range for a UWB subject to FCC Part 15 emissionscon straints.The system as sumptions used in the linkbudge tof the previoussection apply here as well.Fig.5 was generated with a power at tenuation exponent n=4,while in Fig.6n=2.From(12)and(13),it can be observed that capacity is dependent on the product g=c.Both figures were generated withg=100.These figures can beusedt opredict the upper boundon the performance of UWB systems(meetingFCCPart15requirement)over the AWGN.With apower atte nuation exponent of n=4(Fig.5),a-2 spectral efficiency of 5x l0bits/s/Hz can be obtaine datad is tance of 1 0 meters using 32-PPM signalling.With apower at tenuation of n=2(Fig.6),the same spectral efficiency extends to about 75 meters.As a not her example,from Fig.6,for a 1GHz channel,the range of communi cations is up to 75 meters for a data rate of 50Mbits/ sand up to 325 meters for a data rate of 10Mbits/s.These figures provideone more ill ustration of the trade-off between data rate and range of f20f6602842458fb770bf78a6529647d26283479parison of UWB and DS-SS an Channel Capacity Communication sutilizing UWB and direct-sequence spread spectrum(DS-SS) signal sares imilar in the sense that both usea short pulse(PNchipinDS-SS)to get the spread spectrum effect.But,there are fundamental differences between the two systems.

The UWB system is working without acarrier and the pulsesemitted by the transmitter are discontinuous.Conversely,DS-SS signals are continuousand with an

inform at ion wave for mmodulated by aspreadspect rum wave for mandacarrier frequency.With tradi tional DS-SS,the wide bandwid this achieved by modulating the data messagewi thapseudo noise(PN) sequence.The processing gain is obtaine dasaresul tof the PN property and the narrow chip of the modulating sequence.Un like DS-SS,the spread band width of the UWB wave form is generat eddirectly an dnotby modulation with as eparat espreading sequence.

The processing gain of UWB is duet ot heext remely short pulse,which generate save rywide instant aneous bandwidth signal,and isach ievedat thereceiverby time-gatin gmatched to the pulseduration.Fur the rmore,the modulation for matsused in the two system sare different.M-ary PSKin DS-SS systemisasing leort wodimen sionalmodulation,while M-ary PPM in UWB system is M-dimensional.By definition,spread spectrum is in efficient from the point of view of the spectral efficiency of asing leuser.As previously mentioned,its application in UWB is motivated by the need to keep the power spectral density lowand to limit the effect of inter ference.Similarly,in DS-SS,spread spectrum is necessary for inter ference suppression(which leads to the ability of multipleusers to share the band width).It is instructive to compare the capacities of UWB and DS-SS,even though the twosystem haved if ferent mechanism of spread spectrum.For a UWB systemutilizing M-aryPPM and a DS-Sssystem with M-aryPSK,the difference incapacity is due to the different modulation form ats.The capacity of adiscrete-valued in put and acontinuous-valuedout put AWGN channel with Mary PSK modulation was an alyzedin[11].Fig.(7)shows the compar is on between the capacity of UWB as per(10)for M-ary PPM and the capacity of DS-SSasper[11,(5)]for M-aryPSK.Capacities of both systems are plotted as afunction of SNR p,e.g.,UWB and DS-SS have the same power constraint.It is observed that at high SNRp,both systems haves imilarcapacities.For low SNR,and with the exception of M=2 and 60M=4,UWB has a higher capacity than DS-SS.For example for M=32,toachieve acapacit yof 5bits/symbol,the minimum required SNR for DS-SS is about 30dB.The same capacity can be obtained by UWB for only=15dB,again of 15dB.For M=2 and M=4,the situation is reversed with DS-SS having as light advant age of about2.5dB.

脉冲位置调制光谱切片传输

超宽带（UWB）技术的承诺下付出大量的数据非常低功率谱密度的支持者。超宽带无线电的概念是非常有吸引力的，因为它承诺各种用户砂打开大量的频谱，同时它声称小用户之间的干扰不错。不像传统线通信系统，舰载，基于UWB 的通信基带。使用SASE一系列短脉冲传播你NERGY信令从接近DC到几GHz 的成分股。一个典型的技术是，签署一个窗口，在时间和SHIF的脉冲，在该窗口中的位置。这是经典的脉冲位置调制。随着带宽限制，有效地去除，UWB有望加快电子线数据传输高铁酸盐。有些酒吧已发表的工作已经解决的问题，有多少用户可以UWB信道支持。这项工作视为一种类型的能力的能力interms数量的用户容量是来自信号没有ERAT IO（SNR）的注意事项。宏基包含SNR须达到指定的biterr或概率（BEP）。

本课题主要完成PPM的调制与解调。课题的研究PPM 调制的基本原理及数学模型。阐述了PPM 的基本原理并建立其数学模型。其中介绍了单脉冲位置调制（PPM）﹑差分脉冲位置调制（DPPM）和多脉冲位置调制（MPPM）的基本原理,并对几种调制方式进行比较,最终应用单脉冲位置调制（PPM）进行调制电路的设计。简单的介绍了一下MATLAB 与Simulink,并利用其进行PPM 调制电路的设计,进而进行仿真,得到仿真波形。

提供访问网络传输容量利用波分复用技术已经成为核心网络的速度增加了越来越重要的。低成本的方法是可取的,有一种可能性是以一个使用光学滤波器的宽带噪声源窄片,方法通常被称为谱或谱切片（SS）。这提供了一个具有成本效益的替代激光二极管源源不连贯了但强度过量的鼻子。

一个原始的分析提出了频谱分割（SS）采用数字脉冲位置调制（PPM）和传输的最大似然检测。治疗是放置在低成本接入网络的背景下,在系统参数允许的高斯近似的就业。ss-ppm相比,通断键控（ss-ook）在分散的存在和发现是一些典型的2-4分贝更敏感。此外,与激光基于OOK,最佳ss-ppm配置减少或消除SS 功率损失进行了比较。

调制与解调是大气激光通信中的一项关键技术，目前的大气激光通信系统大多设计为开关键控(OOK) 和曼彻斯特编码等强度调制/ 直接检测方式(IM/DD)，这种调制解调方式虽然实现简单，但其抗干扰能力差，为了进一步提高传输通道抗干扰能力，可以采用脉冲位置调制(PPM)方式，该调制方式相对于OOK 等其他调制方式具有低的平均功率，较高的峰值功率，具有编码简单，能量传输效率（每光子传输的信息量）高的优点。兼备安全隐蔽和信噪比高的特性，

因此PPM调制技术在自由空间通信(FSO)中被广泛采用。

本文主要介绍了PPM 调制技术在激光通信中的应用、原理及其Matlab与Quartus II 的仿真实现。其中首先阐述了此种调制方式发展的现状及其在空间激光通信领域中的应用前景，分析了其主要的技术性能和特点，介绍了PPM调制方式的基本概念及其在激光通信中的应用。然后具体对3 种PPM 调制方式进行了分析，并将他们与OOK 调制作了比较，对比了他们在激光通信中的优缺点：虽然PPM 提高了对频带宽度的要求，但他的能量利用率比较高，抗干扰能力比较强。文中介绍了MATLAB 的发展应用及SIMULINK 动态仿真的基本知识，给出了任意2 n 进制PPM 调制的设计和基于MATLAB 的仿真。在Verilog HDL 部分的设计中首先对FPGA 进行了介绍并对Quartus II 软件和Verilog HDL 语言也做了一定的介绍，之后对所做的调制系统进行介绍并对仿真波形进行解释。最后，对PPM 这种调制方式作了展望并且分析了其中有待改进的部分。

所谓激光通信，是指利用激光束作为载体进行语音、数据、图像信息双向传送的一种技术，它采用信道编码技术以改善通信的质量。激光具有扩散小，相干性和方向性好，光束功率密度大等优点。因而适合于保密通信和航天通信，与无线电微波通信相比，激光通信由于其通信容量大，发射天线体积小，抗射频，抗电磁脉冲干扰以及反窃听性能好，特别是抗核破坏能力等优点备受军事和航天领域的青睐。

利用光波作为信息载体进行光通信的历史由来已久。早在一百多年前贝尔就获得了光通信的专利。在两次世界大战中也先后出现过光通信机，但采用的都是普通光源，光的单色性、方向性和相干性都很差，调制和接收也很困难，从而限制了光通信的发展。激光通信依据传输技术的不同，又分为四种：光纤通信、大气通信、空间通信、水下通信。

目前，大气激光信系统多采用强度调制/直接检测IM/DM方式。应用于IM/DD 系统的调制方式有很多种，常用的有OOK(开关键控)、PPM(脉冲位置调制)、DPIM[10] (数字脉冲间隔调制)和DPPM(差分脉冲位置调制)等。最一般的形式是开关键控(OOK)和曼彻斯特编码。在OOK 系统中，通过在每一个比特间隔内使光源脉冲开或关对每个比特进行发送。这是调制光信号最基本的形式，只需通过光源闪烁即可完成编码调制。开关键控OOK 调制方式具体地分为非归零NRZ(Not Return Zero)码与归零RZ(Return Zero)码两种编码格式。OOK调制方式的NRZ 码是在“1”比特时间间隔内发送光脉冲，在“0”比特时间间隔内不发送光脉冲；RZ 码则是在“1”比特的前半个时间间隔内发送光脉冲，在“0”比特时间间隔内不发送光脉冲。因此，NRZ 码与RZ 码的比特速率是相同的，但是RZ 码的激光器调制速率高，较NRZ 码节省一半的功率。在曼彻斯特编码中，序列中每一

比特由2 个开关脉冲组成。通常，光源由编码脉冲波形进行强度调制，同时直接检测接收机对强度调制后信号进行解码，是大气激光通信系统调制方式中最简单、最一般的方式。但是它的功率效率很低，受背景光的影响较大，信噪比很难提高，在光信号经过长距离的大气衰减后己经变的很微弱的情况下，这种调制方式无法保证通信的可靠性与全天候，且通信速率很难提高，不能很好的发挥光通信的频宽优势。

DPIM调制方式中每个符号所包含的时隙数是变化的而不是固定的，并可分为无保护时隙和有保护时隙两种。有保护时隙的DPIM调制方式大多采用一个保护时隙，这样能有效地减少码间串扰的影响，该调制方式的符号S K(k 为符号所表示的十进制数)的时隙个数为k+2，脉冲在每个符号的起始时隙上，后加一个保护空时隙，再加上k 个空时隙表示信息。当接收端解调时，在判断接收到脉冲时隙后，只需要数脉冲时隙后的空时隙个数，再减1 就可以了。因此DPIM 在接收端只需要时钟同步而不需要符号同步，大大简化了系统的实现。

在PPM(Pulse Position Modulation)系统中，采用断续的周期性光脉冲作为载波，载波受到调制信号的控制，脉冲时间位置随之发生变化而传递信息。简单来说，它是一种使激光器发射的激光脉冲的调制方式。在激光通信中，采用PPM 调制方式可以在给定的激光脉冲重复频率下，用最小的光平均功率达到最高的数据传输率，理论上可达1000Mb/s。这大大降低了对激光器发射功率的要求。当然PPM 也存在一定的缺点，比如提高抗干扰能力的同时，付出的代价是增加了对带宽的需求。

毕业设计(论文)外文资料翻译

系别：电子信息系

专业：通信工程

班级：B090310

姓名：田家丰

学号：B09031017

外文出处：Hindawi

附件： 1.原文； 2.译文

2013年03月

本文来源：https://www.bwwdw.com/article/aqje.html