Adaptive Channel Estimation Using Cyclic Prefix

8/2/2019 Adaptive Channel Estimation Using Cyclic Prefix

1/4

Ad a p t i v eChannel E s t i m a t i o nusing Cyclic Pref ixfo r Single Carrier Wireless System with FDE

Wa h y u lAmien SYA F E I * ,Kun i t o s h iN I SH I J O * ,Yu h e iNAGAO* ,Ma s a y u k iKUROSAKI*a n dHi r o s h iOCHIt*D e p a r tm e n to f C om p u t e rS c i e n c ea n d

E l e c t r o n i c sKy u s h uI n s t i t u t eo f Te c hno l o gy,J i z u k aC am p u s ,Kawazu 6 8 0 - 4 ,l i z u k a ,J a p a n8 2 0 - 8 5 0 2Ema i l :( w a s y a f e i , n i s h i j o , n a g a o , k u r o s a k i , o c h i )@ds p . c s e . k y u t e c h . a c j p

t R a d r i x c o . l t d .Ema i l :s u p p o r t@ r a d r i x . c o m

A b s t r a c t -I t ha s b e e nv e r i f i e dr e c e nt l y t h ats i n g l ec a r r i e rc y c l i cp r e f i x e d

( S C - C P )w i t hf r e q u e n c yd o m a i nequ a l i z e r( F D E )i sa p r om i s i n gs y s t emf o rmob i l ec o mmu ni c a t i on a n dr o bu s tc h a n n e le s t ima t i o ni s n e e d e dt o bu i ld t he e q u a l i z e r ' swe i gh t .Conv e n t i o n a lSC-CPw i r e l e s ss y t e mu s e s t r a i n i n gs e q u e n c e st h a t p u t i n e v e r yp a c k e tt o g e t t he c h a n n e li n f o rma t i o n .Th i s i n c r e a s e st h e o v e r h e a doft he s y s t em .H e r ewe p r o p o s ea new c h a n n e le s t ima t i o nt e c h -n i q u eb y e m pl oy in gc y c l i cp r e f i x( C P )t om a in t a i n b ot hs y s t emp e r f o r m a n c ea n d t h r o u g h p u tf o rt h i sSC w i r e l e s ss y s t em ,w i t h3 0 0kHz b an dw i d th a n d1 . 6Mbps t h r o u g h p u t .T h eCP ,wh i c hi si n s e r t e db e t w e e nb lo c ke d d a tat o a v o i di n t e r - b l o c k i n t e r fe r e n c e( I B I ) ,c a nb e v i e w e da s a s our c e ofc h a n n e li n f o rma t i o n .I fwe c a ne s t ima t et he t r a n sm i t t e dCP b yconven t i o n a lw a y,we c a n c o nt i nu et o e s t ima t et he c h a n n e la d a p t i v e l y .The s imu l a t i o nr e s u l t ss h o wt ha t t hep r o p o s e dc h a n n e le s t ima t i o nt e c h n i q u ec a n ma i n t a i nt he s y s t emp e r f o r m a n c ea n d g i v e s1 4 , 9% h i g h e rt h r o u g h p u tc o m p a r e dw i t ht he conven t i o n a lo n ewhen t he number o f p a c k e ti s i n c r e a s e d .I n d e xTe r m s - c h a n n e le s t i m a t i o n ,c y c l i cp r e f i x ,f r e q u e n c y

d o m a i ne q u a l i z a t i o n ,s i ng l e c a rr i er ,w i r e l e s s .

I . I N TRODUCT ION

Or t hogona lF r e q u e n c yDiv i s i o nMul t i p l e x i n g(OFDM)ha sr e c e i v e da l o to f a t t e n t i o n .By i m p l em e n t i n gI nve r s eFas tF o u r i e rTr a n s f o rm( I F FT )a t t h e t r an s mi t te ra ndFFT a t t h er e c e i v e r ,OFDM c o n v e r t sa n i n t e r - s ym b o li n t e r f e r e n c e( I S I )ch ann e li n t opa r a l l e lI S I - f r e esubchanne l sw i t hg a i n se q u a lt ot h ech ann e lf r e q u e n c yr e s p o n s e( C FR )v a l u e so n t h eFFT g r i d .A t t h er e c e i v e r ,e a c h s ub c ha nn e lc a nbe e a s i l ye q u a l i z e db ya s i n g l e - t a pe q u a l i z e ru s i n gs c a l a rd i v i s i o n .T oe l i m i n a t eIB Ib e t w e e ns u c c e s s i v eI F F Tp r o c e s s e db l o c k s ,a CP of l e n g t hn ol e s st h a nt h eCIR o r d e ri s i n s e r t e dp e r t r a n sm i t t e db l o c k ,a n dd i s c a r d e da t t h er e c e i v e r.I na d d i t i o nt o IB I s u p p r e s s i o n ,t h eCP a l s oc o n v e r t st h el i n e a rch ann e lconvo lu t i oni n t oc i r c u l a rc o n v o l u t i o n ,wh i c hf a c i l i t a t e sd i a g o n a l i z a t i o no f t h ea s s o c i a t e dch ann e lma t r i xa n dl e a dt og e tt h ee q u a l i z e rwe i g h tea s i l y.On t he o th e rs i d e ,a n OFDM s i g n a lc o n s i s t sof N s i n u s o i d a l

wav e s ,s o a p e a k - t o - a v e r a g ep o w e rr a t i o( PA P R )b e c om e sh i g h e ra sN i s i n c r e a s e d[ 1 ] ,[ 2 ] .I t a l s os u f f e r sf r o mi n t e r -c a r r i e ri n t e r f e r e n c e( I C I )d ue t o f r e q u e n c yo f f s e tb e t w e e nt r a n sm i t t e ra n dr e c e i v e ra n dDopp l e rs h i f t .A l lof t h e s ed r aw -b a c k scome f r o mt h eI F F Ta t t h et r a n sm i t t e r.[ 3 ] ,[ 4 ] .I t a l s oha s b e e nv er if ie d t ha tc a r e f u l l yd e s i g n e dl i n e a r

p r e c o d i n gOFDM (LP-OFDM) i s m o r ee f f e c t i v ei n d e a l i n gw i t h f r e q u e n c y - s e l e c t i v i t y.One v e r y g o o dl i n e a rp r e c o d e r ,

a c t u a l l ya n n i h i l a t e st h eI F F Ta tt h et r a n sm i t t e r ,a ndl e n d si t s e l ft o a s i n g l e - c a r r i e r( SC )wh e t h e ri t i sc y c l i cp r e f i x e d( SC -CP )o rz e r o - p a d d e d( SC - Z P )[ 5 ] .Th i sl e a dt o s i m p l e rt r a n sm i t ta ndb e t t e rp e r f o rm a n c ec o m p a r e dt oOFDM . [ 4 ] ,[ 6 ] ,[ 7 ] .Co r r e c tch ann e le s t i m a t i o ni s v e r y impo r t a n tt o t h ei m-

p l emen t a t i o nof a nyc ommun i c a t i o nsy s t em .I n mo s tsy s t em ,a k n o w nt r a i n i n gs e q u e n c ei s s e n tb y t h et r a n sm i t t e ra nd

a t r a i n i n ga l g o r i t hmi s p e r f o rm e db y t he r ec ei ve ro n t h eo bs er v ed c ha nn elou tpu ta nd t h ek n o w ni n p u tt o e s t i m a tt h ec h a n n e l .T h ed e t e rm i n i c t i c sl e a s ts q u a r e( D L S )ch ann e li d e n t i f i c a t i o na l g o r i t hmi s s u ch a s imp l ea nd w i d e l yu s edt r a i n i n ga p p r o a c h .H o w e v e ri t i s no t s u i t e df o rt imev a r i a n( TV )c h a n n e l .I n p r a c t i c e ,we e i t h e ra s s um et h a tt h ech ann e li s t ime i n v a r i a n t( T IV )a nd use t h ei n i t i a lt r a i n i n gt o g e tCIR i n f o rma t i o no rp e r i o d i c a l l ys e n t t hi st r a i n i n gs e q u e n c et ot r a ck t hech ann e lv a r i a t i o n .Th e s es o l u t i o n sobv ious lyy i e l di np e r f o rm a n c el o s so rd e c r e a s i n gt h et h r o u g h p u t .O t h e rs o l u t i ow a sp r o p o s e di n [ 8 ] ,u s i n gCP t o a d a p t i v e l ye s t i m a t et h ech ann e li nMu l t i c a r r i e rmodu l a t i o nsy s t em .T h ea l g o r i t hmc a nt r a ck t hech ann e lv a r i a t i o na n dt h ep e r f o rm a n c ei sc omp a r a b lt o s y s t emw i t hr e t r a i n i n g .H o w e v e rt h i ss c h em eemp l o y sr e a lv a l u eo ft h et r a n sm i t t e dt i m e - d om a i ns i g n a lb y u s i n gI F F Ts i z et w i c eb i g g e rt h a nd a t as i z e .I n t h i sp a p e rwe p r o p o s ea d a p t i v ech ann e le s t i m a t i

emp l o y i n gCP i n SC modu l a t i o nw i t h f r e q u e n c yd om a i ne q u a l i z a t i o n( FDE )t omai nt a in b ot hs y s t emp e r f o rm a n c ea ndt h r o u g h p u t .CP b r i n g sm o r ech ann e li nf or ma t io n t ha nP i l o ts ymbo l sa ndCP s i z ei s s h o r t e rt h a n t r a i n i n gs e q u e n c et h a tu s u a l l yo c c u p i e so n e b l o cko fd a t a .We l e t t h et r a n sm i t tt i m e - d om a i ns i g n a ls t a y si nc omp l e xv a l u ea nduse t h es a m es i z eof FFT a nd t h ed a t ab l o c k .T h ep a p e ri s o r g a n i z e da sf o l l ow s .I nS e c t i o nI Iwe e x p l a i nt h eSC-CP w i t hFDE sy s t em ,o b e r v a t i o no n CP,t h ep r o p o s e ds y s t ema nda d a p t i v ech ann e le s t i m a t i o nt e c hn iq u e u s in gCP . S e c t i o nI I I d e a l sw i t h t h ec ompa r a t i o nb e t w e e nt h ep r o p o s e ds y s t emw i t hc onv en t i o n aa ndi d e a ls y s t ems h ow nby s im u l a t i o n .F i n a l l y,Wew r i t esomeconc lu s i oni nS e c t i o nI V .

I I . THEP R O P O SE D ADA P T IV ECHANNE LEST IMAT IONTECHN IQU

Nota t i on :Up p e r - c a s el e t t e r sdeno t ef r e q u e n c yd om a i ns i g -n a l s ;L ow e r - c a s el e t t e r sd eno t e t imed om a i ns i g n a l s ;Bo l dl e t t e r sdeno t ema t r i c e s ;( . ) Ta n d ( . ) Hdeno t et r a n s p o s ea ndHe rm i t i a nma t r i x ;[ . ] i , , jd eno t e st h e( i ,j )- th e n t ryof a ma t r i x

Feb .1 7 - 2 0 ,2 0 0 8I C A C T2 0 0 8SBN9 7 8 - 8 9 - 5 5 1 9 - 1 3 6 - 3-

1 0 3 2 -


2/4

F i g .1 . T h et r a n s c e i v e ro f SC-CP s ys t e m w i t hFDE

wh e r eDH = d i a g [ H ( e ' 0 )H ( e i 2 k. . . H(e i 2N ) ) ]i s a N x Nd i ag on al m at r i xw i t hi t sd i a g o n a le n t r i e sa r eCF R wh i c hc a nbe ob t a i n e db y Fh .T h eFFT ou tpu to f Eq . l i s

F y ( k )= F h x ( k )+ F w(k ) ( 3 )

b y c o n s i d e r i n gE q . 2 ,we o b t a i nF y ( k )= DHF x ( k )+ F w( k )

F i g .2 . T h ep r o p o s e da d a p t i v ec h ann e le s t i m a t o r o rY(k )= DHX(k )+W( k )

O i jd e n o t e szero ma tr ix w i t hs i z ed e f i n e db y i n d e x( ix j )d i a g ( x )i s a d i a g o n a lma t r i x w i t hx on i t sd i a g o n a l .

A .SC-CP w i t hFD E Sy s t emSC-CP as o ne k i n d o fL P - O FDMe l im i n a t e st he n e e do f

I F F Ta t t h et r a n sm i t t e r .C om b i n i n gw i t h f r e q u e n c yd o m a i ne q u a l i z e r( F DE )a t t he r e c e i v e r ,make i t g i v e sb e t t e rp e r f o r -mance t h a nOFDM,e s p e c i a l l yu n d e rmob i l eenv i r onmen t . [ 5 ] ,[ 9 ] ,[ 10 ] , [ 11 ] .F i g u r e1 s h ow st h e t r an s ce i v e ro f c o n v e n t i o n a lSC-CP w i t hFDE s y s t em .In o ur p r o p o s e ds y s t em ,we e n h a n c et h ed a sh ed - li ne b r a c ke t e dp a r t t o bu i l d t he a d a p t i v ec h a n n e le s t im a t i o nb y em p l o y i n gCP ,as i l l u s t r a t e di nF i g .2 .I ft h el e ngh to f CP ( v )i se q u a lt o or g r e a t e rt h a nt h el e ng t h

o f c h a n n e limpu l s eresponse ( C I R )( 1 ) ,i . e . v> 1 ,we o b t a i nt her e c e i v e ds i g n a la f t e rr e m o v i n gt he CP a s :

y ( k )= h x ( k )+ w( k ) ( 1 )

w h e r ex ( k )=

LXO , k ,X I , k ,* *

* ,Xn l , k ] Ti s t r a ns mi t t ed d a t ab l o c k ,w( k )= [WO , k ,W I , k ,* ,Wn_ , k ] Ti s a d i t i v ewh i t eG a u s s i a nn o i s e(AWGN), a n dh i s a N x N c i r c u l a n tma t r i x w i thf i r s tc o l u m ni s [ h O1n - I ] T

From c i r c u l a n tma t r i xp r o p e r t i e s ,we can d i a g o n a l i z es u c hma t r i xb y pr e a n dp o s t - m u l i t p l i c a t i o nw i t hF a n dFH ,w h e r eF i s a N x N FFT ma t r i xw h i c hi t s( n ,k ) - t he n t r y i s d e f i n e db y [ F ] n , k= N- e x p ( - j 2 7 " n k /N ) ,a s :

FhFH = DH ( 2 )

Z e r of o r c i n g( ZF ) FDE i s d o n eb y mul t i p l y i ngt h eFFTo u t p u ta b o v ew i t ht h ei n v e r s eo f DH .

X ( k ) = DH 'Y (k )= X(k )+ DH 'W( k ) ( 6 )

We g e t t he e s t i m a t e dd a t aa f t e rIFFT as

x ( k )= FHX( k )= x ( k )+ DH1w(k )

H e r ewe ca n se e t h a tFDE can suppres c h a n n e le f f e c tf r o mt h er e c e i v e dd a t a .In t h e a b s e n c eo f n o i s e ,t h et r a n sm i t t e dd a t ai sr e c o v e r e dp e r f e c t l y.

I n e a c hp a c k e t ,c o nv e n t i o n a ls y s t e muses 3 b lo ck s o ft r a i n -i n gsequence i n t h ep r e am b l ef o rc h a n n e le s t im a t i o npu r po s ef o l l ow e db y 2 0 b l o c k so f t he d a t a ,as i l l u s t r a t e di nF i g . 3 .

B . Ob s e r v a t i o non Cy c l i cP r e f i xT h eCF R i n f o rm a t i o ni nEq .2 i s usua l lyo b t a i n e db y t r a i n -

i ngprocess. Fo ra t i m ei n v a r i a n tchanne l ,on lyi n i t i a lt r a i n i n gi sn e e d e d .How e v e r ,f o ra t i m ev a r y i n gc h a n n e l ,r e t r a i n i n gmustb e d o n ep e r i o d i c a l l yt o k e e pon w i t ht he c h a n n e lv a r i a t i o no t h e r w i s e ,t he s y s t e mp e r f o r m a n c ed e g r a d e s .Obv iou s l y,s u c ha s c h em ei n cr e a se s t heo v e r h e a do f t he s y s t em .I n t h i ss e c t i o nwe see t h a tb y u s i n gc y c l i cp r e f i x ,r e t r a i n i n gi s n o t neces-sary t o t ra ck t hec h a n n e lv a r i a t i o n s .Le t ' s f i r s tc o n s i d e rt her e c e i v e dp r e f i xp a r t ycp (k )w h i c hi s o r i g i n a l l yd i s c a r d e d .T h er e l a t i o n s h i pb e t w e e nycp (k )a n d t h et r a n sm i t t e dp r e f i xxcp (kis

y, p ( k ) = Tkh + w, p ( k ) ( 8 )

F e b .1 7 - 2 0 ,2 0 0 8ICACT 2 0 0 8

( 4 )

( 5 )

( 7 )

I S BN9 7 8 - 8 9 - 5 5 1 9 - 1 3 6 - 3-

1 0 3 3 -


3/4

wh e r eT k i s a f a tTo e p l i t zma tr ix w i t ht h el owe rt r i a n g l ei sc om p o s e db y t r a n sm i t t e dCP [ x , p ( k ) ] Ta ndt h eu p p e rt r i a n g l ei sc o m p o s e db y t h ep r e v i o u st r a n sm i t t e dCP wh i c hi sa c t u a l l yt h el a s tv p a r tof t h ep r e v i o u ss ymbo l[ x , p ( k -1 ) ] T, o r

2P r e am b l eB P r e am b l eC l P r e am b l eC2 DATA14 0 s a mp l e s 4 0s am p l e s 4 0s a m p l e s 4 0s a m p l e s

3 7 . 5us 11 2 . 5us 3 7 . 5us

8 6 2 . 5us

DATA' OSC s ymbo l s

DATA2 04 0s a m p l e s

2 0x 3 7 . 5us = 7 5 0us~ ,

X - p , kXm - l , k - I. . . Xm-p , k - I

T k ( 9 )

X _ 1 , k ... X - p , k Xm - 1 , k - 1

a ndw,p ( k )= [W- p , k. . . w l , k ] Ti s t h eAWGN c o r r e s p o n d e n c et ot h eCP .No t e :Xm - p , k - I1 X - p , k - I-H e n c e i f a l l t h e p r e f i x p a r t s c o n c a t e n a t e

t o g e t h e rt o f o r m a p a i r o f s e q u e n c e s{ x p l =P . . X_ p , k I... X l, k - l X - p , k X - l , k. . } a n dI Y c pI I . . .Y- p , k - I . . . Y- 1 , k - 1Y- p , k ... Y- T, k * I th er e l a t i o n s h i pb e t w e e nt h e s etwo s a t i s f i e s

Y cp= x c p* h + wc p ( 1 0 )

E q . 1 0s h ow st h a ti f we s e nd{ x c p }t o t h ech ann e la s t h et r a i n i n gs e q u e n c e ,t h ech ann e lou tpu ti se x a c t l y{ y c p } .We c a nuse t h i st r a i n i n gs e q u e n c e st oe s t i m a t et h ech ann e li f we c a nr e c o v e rt h et r a n sm i t t e dCP { x c p jc o r r e c t l y.

C . T heP r o p o s e dSy s t emO u rp r o p o s e da d a p t i v ech ann e le s t i m a t o ra ss h ow ni nF i g .2

ha s 3 d i f f e r e n tp a r t sc om p a r e dt oconven t i ona lone .T h e ya r eQu a n t i z e r ,t im ed om a i nch ann e le s t i m a t o ra ndF FT.1 ) T im eDomain Ch a n n e lEs t ima t o r :Quan t i z e ri su s e dt o

t o r e d u c et h ep r o b a b i l i t yof e r r o rp r o p a g a t i o nby q u a n t i z i n g( r oundup ) t h ee s t ima t e dd a t aa f t e rI F F Tx ( k )t o t h eo r i g i n a l

Quad r a t u r eAmp l i t u d oMod u l a t i o n(QAM) s ymbo lv a l u e .Fo rch ann e le s t i m a t i o np u r p o s ewe j u s tt a k et h el a s tv s amp l e sofQu a n t i z e rou tpu t[ xn_v, k- -x _ , k ]a s t h ee s t ima t e dC P ,( n o t et h a tXn-v,k= X - v, ka n dXn_ l , k= X _, k ) .A f t e ra r r a n g i n gt h e s ee s t ima t e dCP t o Eq . 9 we c a n e s t i m a t et h ech ann e li n t im ed om a i na s

h ( k )= T - 1y c p ( k ) ( 1By t h ep r o p e r t i e so fTo e p l i t zma t r i xwe c a no b t a i nt h eT- 1 ( k )byu s i n gT t ( k ) ,wh e r e[ . ] ti s t h ep s e u d oi n v e r sma t r i x .S i n c ewe d ot h ech ann e le s t i m a t i o ni nt imed om a i nwh i l et h e

e q u a l i z a t i o ni s i nf r e q u e n c yd oma i n ,we n e e dFFT t o c o n v e r tt h ee s t ima t e dCIR of Eq .11t oCF R v a l u et h e nb u i ld t hez e r o

f o r c i n ge q u a l i z e rc o e f f i c i e n tDH.2 )Adap t i v eChann e lEs t ima t i o n :T o m a k et h i sch ann e le s t i m a t o ra d a p t i fwe u s e t he r e c u r s i v el e a s t - s q u a r e( R L S )a l g o r i t hmt og e tt h ech ann e le s t i m a t i o nv a l u ef r omEq . 1 0d uet oi t sg o o dt r a c k i n gp r o p e r t y[ 1 2 ] .No t i c i n gt h a tt h ed a t ai nE q .1 a nd8 a r r i v eb l o ckby b l o c k ,we us e a b l o ckRLS me t h o dwh i c hu p d a t e st h ech ann e le s t i m a t i o nv a l u eby b l o c k s .T h ee s t ima t e dch ann e la t t im ek i s

h ( k )= r - l ( k )z ( k ) ( 1 2 )

F i g .3 . P a c k e tF o rm a t

wh e r e pr ( k )= X i r ( k- 1 )+ X2E x l ( k )xH( k )

1 = 1( 1 3 )

i st h ea p p r o x im a t i o nof t h e c o rr e la t io nma tr ix oft h e t r a ns mt e dCP,wh i l e

p

z ( k )= X 1 z ( k- 1 )+X 2> x l ( k )y 7 ( k )1 = 1

( 1 4 )

i s t h ea p p r o x im a t i o no f t h ec r o s s - c o r r e l a t i o nv e c t o rb e t w e e nt h et r a n sm i t t e dCP a nd t h er e c e i v e dCP, wh i c hi s f o r m e das x l ( k )= l X - I , k... X- , kX - v , k - I X - v + l , k - I] T and y 1 ( k )=[ Y- l , k. . . Y- ) , kY-v, k - 1. . . Y- v+ l , k -] T ,r es p ec t i ve ly , f or1 =v , v- 1 , . . . , 1 .Wh i l eX ,a n dX 2a r ef o r g e t t i n gf ac to rs ft h ed a t ab e t w e e nb locksa ndw i t h i ns a m eb l o c k ,u s u a l l yweuseX 2= 1 a nd) \ < 1 .T hech ann e le s t i m a t i o nv a l u ei su p d a t e de v e r yk t i m e .

I I I . S IMULAT IONRESULTS

P e r f e c ts y n c h r o n i z a t i o ni sa s s umme d .T h em a i np a r am e t e r so f t h es y s t ems p e c i f i c a t i o na n ds i m u l a t i o na r es h ow ni nTab l eI a n dTab l eI I .

A .Sy s t emP e r f o rm a n c eT h ep e r f o rm a n c ec om p a r a t i o nb e t w e e nt h ei d e a l ,t h ec o n -

v e n t i o n a la ndt h ep r o p o s e dsy s t em ,i na s t a t i cmu l t i p a t hc h an -n e li ss h ow ni nF i g . 4 .I d e a ls y s t emm e a n st h a tt h ei n f o rma t i o nof CIR a l r e a d yp r o v i d e da t t he r ec e i ve rwh i le c on v en t io ns y s t emu s e sr e t r a i n i n gs e q u e n c ei ne v e r yp a c k e ti . e .p r e am bt oe s t i m a t et h eC IR .We c a ns e e t h a to n l owSNRconven t i ona ls y s t ems h ow sg o o dp e r f o r m a n c e .H o w e v e ra f t e r2 3 d B o fS N R ,t h ep r o p o s e ds y s t emg i v e sb e t t e rp e r f o rman c ec om p a r e dt ot h econven t i ona lo n ea st h eSNR i s i n c r e a s e d .

B .Sy s t emTh r o u g h p u tWe c a nc a l c u l a t et h et h r o u g h p u tf o rt h econven t i ona ls y s t em

f r omTab l eI a ndI Ia s

Th roughpu tm , xNDx NB P S Cx RT sYM

30x4x 1

3 7 . 5x 1 0 - 61 . 6Mb p s

( 1 5 )

wh e r eNBPSCi s n um b e rof c o d e db i twh i c hi s d e f i n e dbymodu l a t i o nt y p e .S i n c ewe us e 1 6 - Q AM ,NBPSC= 4 .

By c o n s i d e r i n gt h enon - d a t ap a r t si n t h ep r e am b l e ,weo b t a i nt h ed a t ar a t eb e c om e s

Th r o u g hpu t= 2 0 x 1 . 6Mb p s= 1 . 3 9 1 3Mb p s ( 1 6 )

Feb .1 7 - 2 0 ,2 0 0 8I C A C T2 0 0 8

*c ,.

I S BN9 7 8 - 8 9 - 5 5 1 9 - 1 3 6 - 3 - 1 0 3 4 -


4/4

TABLEIS P E C I F I C AT I ONPARAMETERPARAMETER SC

F r e q u e n c yC a r r i e r[ M H z ] 2 6 5 . 8 5B a n d w i d t h[ kHz ] 3 0 0

Number o f d a t a( ND ) 3 0Number o f p i l o t 2

FFT s amp l i n g[MH z ] 1 . 0 6 6FFT Po in t 3 2

FFT p e r i o d( T F F T )[ P S ] 3 0CP l e n g t h( T G I )[ u s ] 7 . 5S ymb o lp e r i o d( T SYM )[ u s ] 3 7 . 5

TABLEI IS IMULAT IONPA RAMET E R

Modu l a t i o n 1 6QAMCo d i n gr a t e( R ) 1 / 2Mu l t i p a t hC I R [ 1 . 00 . 8 1 30 . 6 7 60 . 5 50 . 4 4 70 . 3 7 2 ]

De l a ys p r e a d[ u s ] 5Number o f t r a n sm i t t e dpa ck e t( P ) 5 0 0 0

1 0

1 0 1 5 2 0 2 5 3 0SN R ( d B )

F i g .4 . B i ter ror r a t e c om p a r a t i o ni nmu l t i p a t hc h a n n e l

A f t e ra pp l y i n gt h ea d a p t i v ec h a n n e le s t im a t o rt h a tem p l o y s

CP ,we

a nn h i l at e t hen e e do f t r a i n i n gsequences s t a r tf r o mt h es e c o n dt r a n sm i t t e dp a c k e ta n dg e t t h eh i g h e rd a t ar a t e as

Th r o u g h p u t p r o2 0 + 2 0 x ( P - 1 )

x 1.6Mbps

2 3 + 2 0 x ( P - 1 ) ( 1 7 )

w h e r eP i s t he n u m b e ro f t r a n sm i t t e dp a c k e t .As P i si n c r e a s e dwe a lmo s ta t t a i nt hemax imum d a t ar a t e .In t h i ss imu l a t i onwes e n d5 0 0 0p a c k e t sa n ds h o wt h et h r o u ghpu tf o rt he f i r s t2 0 0t r a n sm i t t e dp a c k e t si nF i g .5 .F u r t h e rmo r e ,c e r t a i n l yt h es y s t e mt h r o u g h p u tca n b e i n c r e a s e db y u si n g h i g he rmodu l a t i o n ,s u c has 6 4 QAM .

1 . 6 0 1 . .-

1 . 5 5

> 1 . 5 0

Q.

1 . 4 5

1 . 4 0F

MaximumD a t ar a t e

P r o p o s e dS y s t e m

Re t r a i n i ngS y s t e m

I V .CONCLUS IONI nt h i spaper, we h a v ep r e s e n t e dan a d a p t i v ec h a n n e le s t i -

ma t i o nt e c h n i q u eus i ngc yc l ic p r ef i xf o rSC w i r e l e s ss y s t e mw i t h FDE. T h e p r o p o s e ds y s t e mcan ma i n t a i nt he s y s t e mp e r f o r m a n c ea n d g i v e s1 4 . 9 % h i g h e r t h r o u ghpu tas t hen u m b e ro f t r a n sm i t t e dp a c k e ti s i n c r e a s e d .F o rf u t u r ew o r kwew i l le n h a n c e dt he a l g o r i t hma n de x a m i n et he p r o p o s e ds y s t e mu n d e rd y n am i cc h a nn e l t he nus e h i g h e rm o d u l a t i o ns u c has 6 4QAM t o a c h i e v eh i g he r t h r ou g hp u t.

REFERENCES[ 1 ]D F a l c o n e r ,S .L .A r i y a v i s i t a k u l ,A . B e n y am i n - S e e y a r ,a n dB . E i d s o n ,

F r e q u e n c yDomain Equa l i z a t i o nf o r S i n g l e -C a r r i e rB r o a d b a n dWire l e s sSys t ems , IEEE C omm u n i c a t i o n sMa g a z i n e ,Vo l .4 0 ,No .4 , p p . 5 8 - 6 6 ,2 0 0 2 .

[ 2 ]X . Hu a n g ,J . Lu, a n d J . Zh e n g ,Gen e r a l i z e dC i r c u l a rT r a ns f or m f orOFDM Transmi s s i on , i n P r o c . Ve h i c u l a rTe c h n o l o g yC o n f . ,v o l . 1 ,p p . 4 7 4 - 4 7 8 ,2 0 0 4

[ 3 ]S .Ohno ,P e r f o r m a n c eo f S i n g l e -C a r r i e rB l o c kTr a n sm i s s i o nO v e rMu l -t i p a t hF ad in g F ad in gCh a n n e l sWi t hL i n e a rE q u a l i z a t i o n ,IEEETr a n s a c t i o n so n S i g n a lP r o c e ss i n g . Vo l . 5 4 ,No . 1 0 ,p p . 3 6 7 8 - 3 6 8 7 ,O c t o b e r2 0 0 6 .

[ 4 ]Z . Wa n g ,X . Ma ,a n dG . B .G i a n n a k i s ,OFDM o r S i n g l eC a r r i e rB l o c kTr an sm i s s i o n ? ,IEEE Tr a n s a c t i o n son Commun i c a t i o n s ,V 6 l . 5 2 ,No . 3 ,p p . 3 8 0 - 3 9 4 ,March 2 0 0 4 .

[ 5 ]Z . Wang a n d G . B . G i an na k is , L i ne a r lyP r e c o d e do r Coded OFDMAg a i n s tWi r e l e s sC h a n n e lF a d e s ? , i n P r o c . o f IEEE Wo r k s h o po nS i g n a lP r o c .A d n a v c e si nWi r e l e s sComm . Ta o yu a n, Ta i wa n ,p p . 2 6 7 -270 ,March 20 - 2 3 ,2 0 0 1 .

1 . 3 50 5 0 1 0 0 1 5 0

Number o fTr a n sm i t t e dP a c k e t( P )2 0 0

F i g .5 . T h r o u g h p u tc om p a r a t i o nf o rt he f i r s t2 0 0o f P

[ 6 ]Z .Wa n g ,X . Ma , a n d G . B .G i a nn a ki s , O p t im a li t yo f S i n g l eC a r r i e rZ e r o - P a d d e dB l o c kTr a n sm i s s i o n ? ,i nP r o c .Wi r e l e s sC ommu n i c a t i o nNe tw o r k i n gCon f .Or l a ndo ,F l o r i d a ,Vo l . 2 ,p p . 6 6 0 - 6 6 4 ,March 1 7 - 2 12 0 0 2 .

[ 7 ]Y P .L i na n dS . M . P h oo ng ,BERMi n im i z e dOFDM S y s t em swi t hC h a n n e lI n d e p e n d e n tP r e c o d e r s , IEEE Tr a n s a c t i o n so n S i g n a lP r o c e s s i n gVo l . 5 1 ,No . 9 ,p p . 2 3 6 9 - 2 3 8 0 ,S e p t em b e r2 0 0 3 .

[ 8 ]X .Wang a n dK . J . Ray L i uAd a p t i v eC h a n n e lEs t ima t i o nU s i n gCyc l i cP r e f i xi n Mu l t i c a r r i e rMod u l a t i o nSy s t em , IEEE C ommu n i c a t i o nL e t t e r s Vo l .3 ,No .1 0 ,O c t o b e r1 9 9 9 .p p . 2 9 1 - 2 9 3 .

[ 9 ]Wah y u lAmien S y a f e i ,Kun i t o s h iN i s h i j o ,Yuh e i N a g a o ,Ma s a y u k iKu ro s a k i ,a n dH i r os hi O c hi ." P e r f o r m a n c eVe r i f i c a t i o nb e t w e e nOFDMa n dSC-FDE F o rNarrowband M o bi l e C o m mu n i c a t i onSy s t em , IEICE

Soc i e t yC o n f e r e n c eP r o c e d i n g s ,To t t o r i ,J apan ,S e p t em b e r1 2 ,2 0 0 7 .[ 1 0 ]A . Cz y lw i k ,Comp a r i s o nb e t w e e nA d a p t i v eOFDM a n dS ing l eC a r r i e rMod u l a t i o nw i t hF r e q u e n c yDomain Equa l i z a t i on , i n P r o c .Veh i c u l a rTe c h n o l o g yC o n f . ,v o l .2 , Pho en i x ,A Z ,p p . 8 6 5 - 8 6 9 ,1 9 9 7 .

[ 1 1 ]Z .Wang a n dG . B .Gi ann a k i s ,Wi r e l e s sMu l t i c a r r i e rCommun i c a t i o n sw h e r eF o u r i e rMe e t sShannon , IEEE S i gn a l P r o ce s si n gMa g a z i n e .v o l . 4 7 ,n o . 3 ,p p . 2 9 - 4 8 ,May 2 0 0 0 .

[ 1 2 ]S .H a yk i n, A d a p t i veF i l t e rTheo r y, P r e n t i c eHa l l ,E n g l e w o o dCl i f f s ,New J e r s e y,4 t hEd i t i o n ,2 0 0 0 .

F e b .1 7 - 2 0 ,2 0 0 8ICACT 2 0 0 8

1 0 - d e a l+ R e t r a i n i n g

Ad a p t i v e

1 0

1 0

1. 10 -3n0

I S BN9 7 8 - 8 9 - 5 5 1 9 - 1 3 6 - 3-

1 0 3 5 -

Adaptive Channel Estimation Using Cyclic Prefix

Documents