Research ArticleSpace-Time Block Coding with Beamforming forTriple-Polarized Uniform Linear Array Systems

Xin Su1 and Kyung Hi Chang2

1The Program in IT amp Media Convergence Studies Inha University Incheon 402-751 Republic of Korea2The Electronic Engineering Department Inha University Incheon 402-751 Republic of Korea

Correspondence should be addressed to Kyung Hi Chang khchanginhaackr

Received 25 May 2015 Revised 8 August 2015 Accepted 20 August 2015

Academic Editor Mourad Nedil

Copyright copy 2015 X Su and K H Chang This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

Generally space-time block coding (STBC) and beamforming (BF) gains cannot be obtained simultaneously because the formerperforms well under a low correlated MIMO channel and the latter works efficiently in an environment with high correlationHowever array systems with antenna polarization have the potential to achieve gains with both techniques simultaneously becausethe cross-branch links in the system are usually uncorrelated The cross-array links on the other hand can be highly correlatedby setting the array element space equal to or less than a half-wavelength This paper proposes a scheme to explore STBC and BFsimultaneously via a triple-polarized uniform linear array (TPULA) system The proposed scheme was verified based on the LongTermEvolution-Advanced (LTE-A) specification under a polarizedMIMO (PMIMO) channelmodel and therewith the simulationresults confirmed the validity of our proposed scheme

1 Introduction

Recently there has been a gradual demand for the use of apolarized antenna system especially for 5G networks This ismainly because antenna polarization is a crucial resource tobe exploited for the design of space-limited wireless devicesTechniques such as space-time diversity multiplexing andarray processing can be applied to polarized antenna systemsto boost system throughput Althoughmany research articlesand ongoing projects have been working on antenna polar-ization [1ndash3] exploration into using space-time block coding(STBC) and beamforming (BF) techniques for polarizedantenna systems has not received much attention in theliterature Typically there is no efficient approach to simul-taneously obtaining STBC and BF gains in polarized antennasystems This paper proposes a robust scheme to cope withthis constraint by applying STBC in three transmit antennaeand BF techniques via a triple-polarized uniform linear array(TPULA) systemThat is the polarized antenna branches aregrouped into three sets that steer three orthogonal beams attheX-Y Z-X andY-Z planes STBC is additionally applied atthree orthogonally colocated branches of each array element(AE) to accommodate the space-time diversity gain

Via the proposed scheme by exploring the characteristicsof antenna polarization theMIMO techniques that is STBCand BF can be realized at mobile stations (MSs) simultane-ously to guarantee the system performance For exampletwo AEs could be employed at an MS with a space of 8 cmwhen the carrier frequency is 2GHz At each AE threeorthogonally colocated antenna branches perform STBC thatcan further enhance the system performance The imple-mentation complexity is increased by the proposed schemethat costs higher power consumption for the MS becauseit needs to calculate the BF weights as well as performingthe STBC decoding However it is very pivotal for cell-edge MSs to keep the performance in 5G communicationsystems especially for the applications without tolerance ofretransmissions In the simulation results the block error rate(BLER) performance improves which shows the validity ofour proposed scheme

The remainder of this paper is organized as followsSection 2 provides the TPULA configuration and Section 3describes our proposed scheme The simulation results arediscussed in Section 4 and finally our conclusions are drawnin Section 5

Hindawi Publishing CorporationInternational Journal of Antennas and PropagationVolume 2015 Article ID 819701 9 pageshttpdxdoiorg1011552015819701

2 International Journal of Antennas and Propagation

ZZ

YX

YX

TxB1

TxB3

TxB2

RxB1

RxB2

RxB3

(a) TPMIMOB

X

Z

Y

X

Z

Y

TxA1B1

TxA1B3

TxA1B2

RxA1B1 RxAQB1

RxA1B3

RxA1B2

RxAQB3

RxAQB2

middot middot middotmiddot middot middot

TxAPB1

TxAPB3

TxAPB2

(b) TPMIMOAampB (TPULA)

Figure 1 Orthogonal TPMIMO system

2 TPULA Configuration

Figure 1 provides examples of orthogonal triple-polarizedMIMO (TPMIMO) system in the array and branch (AampB)multiple antennae configuration According to Figure 1(a)there are three orthogonally colocated antenna branchesat the transmitter (Tx) and receiver (Rx) ports Thereforethis PMIMO transceiver is defined at the branch level asTPMIMOB By uniformly aligning the polarized antennaports (AEs) at the Tx and the Rx as illustrated in Figure 1(b)a PMIMO system is defined as TPMIMOAampB which indicatesthat antenna polarization is realized at both branch and arraylevels

The beamwidth is relevant to the array configurationwhere it is inversely proportional to the number of AEsand the AE space Because the space among three colocatedbranches at each AE is fixed at zero as shown in Figure 1which makes beamwidth scale up to 360∘ the beams shouldbe generated via corresponding cross-array branches ratherthan the colocated branches at each AE [3] Therefore aTPULA system can generate three orthogonal beams asfollows

(i) beam steered and varied in the X-Y plane is formedby the set of branches of TRxA

119901(119902)B1 where 119901(119902) =

1 2 119875(119876)(ii) beam steered and varied in the Z-X plane is formed

by the set of branches of TRxA119901(119902)

B2

(iii) beam steered and varied in the Y-Z plane is formedby the set of branches of TRxA

119901(119902)B3

Figure 2 depicts the 3D beams generated by a TPULAsystem where three orthogonal beams can be steered and

varied separately on the X-Y Z-X and Y-Z planes Byapplying the TPULA into an orthogonal frequency divisionmultiplexing (OFDM) system considering STBC and BF arobust MIMO transceiver is provided as seen in Figure 3which is explained in detail in the following section

3 TPULA System withSTBC and Beamforming

As studied by [1] the cross-array links (eg from TxA1B1

to RxA1B1and RxA

2B1) are highly correlated by setting

the AE space equal to or less than a half-wavelengthThe cross-branch links (eg from TxA

1B1and TxA

1B2to

RxA1B1) however are usually uncorrelated due to the space

polarization This inspired us to incorporate STBC and BFtechniques simultaneously in the TPULA system to boostperformance Moreover in order to achieve full-rate codingwith an odd number of transmit antennae quasiorthogonalSTBC (QO-STBC) has emerged in the literature [4 5] whichfully explores diversity gain but increases the complexity ofdecoding due to nonorthogonal interference In this paperwe propose a scheme to combine the QO-STBC for threetransmit antennae with BF techniques via a TPULA systemas follows

In Figure 3 the BF weights (119908Tx119901119887

and 119908Rx119902119887

) are multipliedbefore the inverse fast Fourier transform (IFFT) block of theTx and after the FFT block of the Rx where 119901 (1 2 119875)119902 (1 2 119876) and 119887 (1 2 3) denote the indexes of the TxAERxAE and antenna branch respectively Note that we haveconsidered the digital BF where the TxBF weights obtainedby zero-forcing BF are determined by the Rx location and theRxBF weights obtained by MMSE BF are determined based

International Journal of Antennas and Propagation 3

minus1 minus05 0 05 1

minus1 minus05 0 051

minus1minus050051minus1

minus08

minus06

minus04

minus02

0

02

04

06

08

1

Z

XY

Y minus1minus050051

minus1

minus08

minus06

minus04

minus02

0

02

04

06

08

1

Z

X

Figure 2 3D beams generated by a TPULA system

Y

X

QO-STBC

Y

X

QO-STBCdecodingQO-STBCdecoding

QO-STBCdecoding

QO-STBCdecodingQO-STBCdecodingQO-STBCdecoding

QO-STBCdecodingQO-STBCdecoding

QO-STBCdecoding

Rx

Rx

combining

Tx

IFFT and GI extension GI removal and FFT

Xp

Xp

Xp

wTx12

wTx13

wTx11

A1B2A1B2

A1B3A1B3

A1B1A1B1

wTx22

wTx23

wTx21

A2B3 A2B3

A2B2

A2B1

A2B2

A2B1

wRx12

wRx13

wRx11

wRx22

wRx23

wRx21

wRxQ2

wRxQ3

wRxQ1

AQB3AQB2

AQB1

APB3

APB2

APB1

wTxP2

wTxP3

wTxP1

Figure 3 OFDM-based TPULA system with STBC and BF

on reference signal The BF weights then differentiate eachother based on the polarized antenna branches of each arrayelement instead of each subcarrierX

119901in expression (1) is the

transmitted QO-STBC symbol matrix [4] where the Romannumeral is the symbol index 119894 is the time index modulo of 4(119868 = 4) and 119895 is the antenna index modulo of 3 (119869 = 3)

X119901=(

119909II 119909III 119909IV

119909lowast

I minus119909lowast

IV 119909lowast

III

119909IV 119909I 119909II

119909lowast

III minus119909lowast

II 119909lowast

I

)

119868minus119887119910minus119869

(1)

For the cross-branch links from the pth TxAE with threebranches to a single Rx branch 119902

119887 that is h

119901sim119902119887 the received

signal before the RxBF weights multiplication is given as

R119901sim119902119887= (h119901sim119902119887(WTx119901

⊙ X119901)119879

)119879

+ n119901sim119902119887

= (WTx119901

⊙ X119901) h119879119901sim119902119887

+ n119901sim119902119887

= (119903119901sim119902119887 I 119903119901sim119902119887 II 119903119901sim119902119887 III 119903119901sim119902119887 IV)

119879

(2)

4 International Journal of Antennas and Propagation

where ldquo⊙rdquo denotes the Hadamard product and WTx119901

is theTxBF weighting matrix given by

WTx119901

=(

119908Tx1199011

119908Tx1199012

119908Tx1199013

119908Tx1199011

119908Tx1199012

119908Tx1199013

119908Tx1199011

119908Tx1199012

119908Tx1199013

119908Tx1199011

119908Tx1199012

119908Tx1199013

) (3)

h119901sim119902119887

is the vector of the polarized cross-branch links givenas

h119901sim119902119887= (ℎ1199011sim119902119887ℎ1199012sim119902119887ℎ1199013sim119902119887) (4)

and n119901sim119902119887

is the noise vector given by

n119901sim119902119887= (120590119901sim119902119887 I 120590119901sim119902119887 II 120590119901sim119902119887 III 120590119901sim119902119887 IV)

119879

(5)

The minimum mean square error (MMSE) RxBF is usedto minimize the error between the RxBF output and theavailable reference signal where the optimal weights119908Rx

119902119887

canbe obtained by theWiener-Hoff equation [6]Then the signalafter RxBF is given by

R119901sim119902119887= 119908

Rx119902119887

R119901sim119902119887 (6)

The STBC decoding applied after the RxBF gives

X119901sim119902119887= D119901sim119902119887

T119901sim119902119887

= (119909119901sim119902119887 I 119909119901sim119902119887 II 119909119901sim119902119887 III 119909119901sim119902119887 IV)

119879

(7)

Here D119901sim119902119887

is the STBC decoding matrix which is modifiedbased on equation (12) in [4] by considering TxBF weightsgiven by

D119901sim119902119887

=(

(

0 119908Tx1199011

ℎ1199011sim119902119887(119908

Tx1199012

)lowast

ℎlowast

1199012sim119902119887

119908Tx1199013

ℎ1199013sim119902119887

(119908Tx1199011

)lowast

ℎlowast

1199011sim119902119887

0 (119908Tx1199013

)lowast

ℎlowast

1199013sim119902119887

minus119908Tx1199012

ℎ1199012sim119902119887

(119908Tx1199012

)lowast

ℎlowast

1199012sim119902119887

119908Tx1199013

ℎ1199013sim119902119887

0 119908Tx1199011

ℎ1199011sim119902119887

(119908Tx1199013

)lowast

ℎlowast

1199013sim119902119887

minus119908Tx1199012

ℎ1199012sim119902119887(119908

Tx1199011

)lowast

ℎlowast

1199011sim119902119887

0

)

)

(8)

And Τ119901sim119902119887

is defined as

Τ119901sim119902119887= (119903119901sim119902119887 I 119903

lowast

119901sim119902119887 II 119903119901sim119902119887 III 119903lowast

119901sim119902119887 IV)119879 (9)

by taking the conjugate of the second and the fourth elementsof R119901sim119902119887

The final output of the proposed scheme is obtainedby the last functional block in Figure 3 (ie the Rx combin-ing) given as

X =119876

sum

119902=1

3

sum

119887=1

(119909119901sim119902119887 I 119909119901sim119902119887 II 119909119901sim119902119887 III 119909119901sim119902119887 IV)

119879

(10)

Li et al [4] used the receiver reported by Liu et al [7]to cope with the interference caused by nonorthogonality ofthe QO-STBS In this paper the complexity of interferencecancellation can be simplified by the adaptive setting of TxBFweights in (3) By calculation the decoded signal in (10) isgiven by

119909119901sim119902119887 I = (119908

Rx119902119887

)2

(10038161003816100381610038161003816119908

Tx1199011

ℎ1199011sim119902119887

10038161003816100381610038161003816

2

+10038161003816100381610038161003816119908

Tx1199012

ℎ1199012sim119902119887

10038161003816100381610038161003816

2

+10038161003816100381610038161003816119908

Tx1199013

ℎ1199013sim119902119887

10038161003816100381610038161003816

2

) 119909I⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟

Desired singal

+ (119908Rx119902119887

)2

[119908Tx1199011

ℎ1199011sim119902119887(119908

Tx1199013

)lowast

ℎlowast

1199013sim119902119887

+ 119908Tx1199013

ℎ1199013sim119902119887(119908

Tx1199011

)lowast

ℎlowast

1199011sim119902119887

] 119909III⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟

Interference caused by adjacent signal

+ (119908Rx119902119887

)2

[119908Tx1199011

ℎ1199011sim119902119887120590lowast

119901sim119902119887 II + (119908Tx1199012

)lowast

ℎlowast

1199012sim119902119887

120590119901sim119902119887 III + 119908

Tx1199013

ℎ1199013sim119902119887120590lowast

119901sim119902119887 IV]⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟

Noise

119909119901sim119902119887 II = (119908

Rx119902119887

)2

(10038161003816100381610038161003816119908

Tx1199011

ℎ1199011sim119902119887

10038161003816100381610038161003816

2

+10038161003816100381610038161003816119908

Tx1199012

ℎ1199012sim119902119887

10038161003816100381610038161003816

2

+10038161003816100381610038161003816119908

Tx1199013

ℎ1199013sim119902119887

10038161003816100381610038161003816

2

) 119909II

+ (119908Rx119902119887

)2

[119908Tx1199011

ℎ1199011sim119902119887(119908

Tx1199013

)lowast

ℎlowast

1199013sim119902119887

+ 119908Tx1199013

ℎ1199013sim119902119887(119908

Tx1199011

)lowast

ℎlowast

1199011sim119902119887

] 119909IV

+ (119908Rx119902119887

)2

[119908Tx1199011

ℎ1199011sim119902119887120590lowast

119901sim119902119887 II + (119908Tx1199012

)lowast

ℎlowast

1199012sim119902119887

120590119901sim119902119887 III + 119908

Tx1199013

ℎ1199013sim119902119887120590lowast

119901sim119902119887 IV]

119909119901sim119902119887 III = (119908

Rx119902119887

)2

(10038161003816100381610038161003816119908

Tx1199011

ℎ1199011sim119902119887

10038161003816100381610038161003816

2

+10038161003816100381610038161003816119908

Tx1199012

ℎ1199012sim119902119887

10038161003816100381610038161003816

2

+10038161003816100381610038161003816119908

Tx1199013

ℎ1199013sim119902119887

10038161003816100381610038161003816

2

) 119909III

+ (119908Rx119902119887

)2

[119908Tx1199011

ℎ1199011sim119902119887(119908

Tx1199013

)lowast

ℎlowast

1199013sim119902119887

+ 119908Tx1199013

ℎ1199013sim119902119887(119908

Tx1199011

)lowast

ℎlowast

1199011sim119902119887

] 119909I

+ (119908Rx119902119887

)2

[119908Tx1199011

ℎ1199011sim119902119887120590lowast

119901sim119902119887 II + (119908Tx1199012

)lowast

ℎlowast

1199012sim119902119887

120590119901sim119902119887 III + 119908

Tx1199013

ℎ1199013sim119902119887120590lowast

119901sim119902119887 IV]

International Journal of Antennas and Propagation 5

119909119901sim119902119887 IV = (119908

Rx119902119887

)2

(10038161003816100381610038161003816119908

Tx1199011

ℎ1199011sim119902119887

10038161003816100381610038161003816

2

+10038161003816100381610038161003816119908

Tx1199012

ℎ1199012sim119902119887

10038161003816100381610038161003816

2

+10038161003816100381610038161003816119908

Tx1199013

ℎ1199013sim119902119887

10038161003816100381610038161003816

2

) 119909IV

+ (119908Rx119902119887

)2

[119908Tx1199011

ℎ1199011sim119902119887(119908

Tx1199013

)lowast

ℎlowast

1199013sim119902119887

+ 119908Tx1199013

ℎ1199013sim119902119887(119908

Tx1199011

)lowast

ℎlowast

1199011sim119902119887

] 119909II

+ (119908Rx119902119887

)2

[119908Tx1199011

ℎ1199011sim119902119887120590lowast

119901sim119902119887 II + (119908Tx1199012

)lowast

ℎlowast

1199012sim119902119887

120590119901sim119902119887 III + 119908

Tx1199013

ℎ1199013sim119902119887120590lowast

119901sim119902119887 IV]

(11)

Formulae (11) indicate that 119909I and 119909III interfere with eachother as well as 119909II and 119909IV For interference cancellationit is better to set the scaling factor that is either 119908Tx

1199011

or119908

Tx1199013

to zero The setting of zero needs to be determined bythe off boresight angle (OBSA) effect discussed by Liu [8]where boresight refers to the axis vertical to the orientationof the array alignment For example in Figure 4 the solidbeam is generated by the set of branches of TxApB1 whilethe dotted beam is generated by the set of branches ofTxApB3 According to Liu [8] the beamwidth is known andis directly proportional to the OBSA value Therefore 119908Tx

1199013

=

0 (when 120572 lt 120573) or 119908Tx1199011

= 0 (when 120572 gt 120573) can be suggestedto avoid increasing the beamwidth Note that in this paperthe location of the Rx is assumed to be known by the TxAdditionally by using the zero-forcing TxBF we can have twooptions for TxBF weight settings that is (119908

1199011= 0 119908

1199012=

ℎminus1

1199012

1199081199013= ℎminus1

1199013

) or (1199081199013= 0 119908

1199012= ℎminus1

1199012

1199081199011= ℎminus1

1199011

) which isdeterminedmainly based on the location of the Rx Note thatwe did not focus on the TDD system therefore the channelreciprocity is not performed Since we assume the location ofRx is known by the Tx the BF weight used in the paper isinverse of the array propagation factor at Tx side [3]

According to Figure 3 the proposed TPULA system canbe easily extended as the polarized massive MIMO (PM-MIMO) system by employing a large number of AEs whichrequires a large number (tens or hundreds) of antennaeserving at both Tx and Rx Consequently the system designthen should consider the plenty of power consumption andhigh cost of RF chains Antenna selection scheme is necessaryfor PM-MIMOsystem in both transmitting end and receivingendAn energy efficient antenna selection algorithmbased onconvex optimization might be proposed where the numberof transmit antenna and the number of receive antenna needto be jointly optimized to maximize energy efficiency Goodperformance gain of energy efficiency should be obtainedcomparing with no antenna selection case

4 Performance Verification

We verified the proposed scheme via Monte Carlo simula-tions based on the LTE-A specification where Table 1 liststhe details of the parameter settings The downlink datamapping for LTE-A resource blocks used in the simulationsis demonstrated in Figure 5 which is based on the QO-STBC symbol matrix of expression (1) 119875A119901B1 119875A119901B2 and119875A119901B3 denote the pilots used for practical channel estimation

and a total of 24 pilots were used separately for 24 transmitantennae An extension of work by Su et al [9] is appliedto simulations of the polarized MIMO channel for whichchannel characteristics are also listed in Table 1 Note that thelocation of the Rx is updated periodically at each simulationloop

Sincewe have proposed the scheme to combine STBCandBF firstly and could not find any existing combined STBC-BF scheme for the performance comparison according toliteratures we only compared four transmission cases in thispaper to verify our proposal by discussing the achievementof STBC and BF gains Figure 6 shows the simulated BLERcurves including Case 1 of TPULA only (without STBC andBF) Case 2 of TPULA with STBC Case 3 of TPULA withSTBC and TxBF and Case 4 of TPULA with STBC andTRxBF where the target BLER for performance comparisonis fixed at 10minus2 According to Figure 6 Case 1 has the worstperformance because there is no space-time coding and BFgain achieved By applying the receiver studied by Liu et al [7]for interference cancellation almost 1 dB of signal-to-noiseratio (SNR) gain at the target BLER can be achievedwithCase2 under the indoor environment compared with Case 1 Inaddition compared with Case 2 about 23 dB of SNR gainunder the indoor environment at the target BLER is obtainedwith Case 3 when the TxBF is applied Compared with Case3 05 dB SNR gain at the target BLER is achieved by usingthe MMSE RxBF via Case 4 because the amount of error fedinto the STBC decoding process is further decreased Theerror floors occur under the outdoor environment due tohigh mobility at 100 kmh This indicates that the numberof pilots in the TPULA system is not enough to compensatethe channel correctly under an environment with fast time-varying phase response Again the efficiency of our proposedscheme can be quantified in terms of throughput Figure 7illustrates the simulated throughput curves By focusing onthe curves obtained under the indoor environment we canobserve that Case 4 of TPULA with STBC and TRxBF canachieve the highest throughput of 21Mbps at 119864

1198871198730of 0 dB

and 20Mbps 19Mbps and 17Mbps for Case 3 Case 2 andCase 1 at 119864

1198871198730of 0 dB respectively

In addition we evaluate the average BLER performancewhen two mobile users are employed In this case the eightAEs at Tx are divided into two groups where each group iswith four AEs and performs TxBF for a mobile user Figure 8illustrates the simulation results of average BLER for twousers with the transmission Case 4 According to Figure 8we can observe that the performance decreases a lot dueto the loss of diversity order at Tx because multiusers need

6 International Journal of Antennas and Propagation

30

210

60

240

90

270

120

300

150

330

180 0

Z

X

120573

120572

Figure 4 Setting of zero for TxBF weights

Subframe20 1 3 4 5 6 7 8 9 10 11 12 13

0

1

11

Reso

urce

blo

ck

10

9

8

7

6

5

4

3

2

x98x4xlowast1

x102x8xlowast5

x106x12xlowast9

x110x16xlowast13

x114x20xlowast17

x118x24xlowast21

x122x28xlowast25

x126x32xlowast29

x130x36xlowast33

x134x40xlowast37

x138x44xlowast41

x142

x50

x54

x58

x62

x66

x70

x74

x78

x82

x86

x90

x94x48xlowast45

xlowast3

xlowast7

xlowast11

xlowast15

xlowast19

xlowast23

xlowast27

xlowast31

xlowast35

xlowast39

xlowast43

xlowast47

xlowast49

xlowast53

xlowast57

xlowast61

xlowast65

xlowast69

xlowast73

xlowast77

xlowast81

xlowast85

xlowast89

xlowast93

x52

x56

x60

x64

x68

x72

x76

x80

x84

x88

x92

x96

xlowast51

xlowast55

xlowast59

xlowast63

xlowast67

xlowast71

xlowast75

xlowast79

xlowast83

xlowast87

xlowast91

xlowast95

xlowast97

xlowast101

xlowast105

xlowast109

xlowast113

xlowast117

xlowast121

xlowast125

xlowast129

xlowast133

xlowast137

xlowast141

x100

x104

x108

x112

x116

x120

x124

x128

x132

x136

x140

x144

xlowast99

xlowast103

xlowast107

xlowast111

xlowast115

xlowast119

xlowast123

xlowast127

xlowast131

xlowast135

xlowast139

xlowast143

PA1B1

PA2B1

PA3B1

PA4B1

PA5B1

PA6B1

PA7B1

PA8B1

x2

x6

x10

x14

x18

x22

x26

x30

x34

x38

x42

x46

(a) Data mapping at TxApB1

20 1 3 4 5 6 7 8 9 10 11 12 13

0

1

11

Reso

urce

blo

ck

10

9

8

7

6

5

4

3

2

Subframe

x1

x5

x9

x13

x17

x21

x25

x29

x33

x37

x41

x45

x3

x7

x11

x15

x19

x23

x27

x31

x35

x39

x43

x47

x49

x53

x57

x61

x65

x69

x73

x77

x81

x85

x89

x93

x51

x55

x59

x63

x67

x71

x75

x79

x83

x87

x91

x95

x97

x101

x105

x109

x113

x117

x121

x125

x129

x133

x137

x141

x99

x103

x107

x111

x115

x119

x123

x127

x131

x135

x139

x143

minusxlowast100

minusxlowast104

minusxlowast108

minusxlowast112

minusxlowast116

minusxlowast120

minusxlowast124

minusxlowast128

minusxlowast132

minusxlowast136

minusxlowast140

minusxlowast144

minusxlowast98

minusxlowast102

minusxlowast106

minusxlowast110

minusxlowast114

minusxlowast118

minusxlowast122

minusxlowast126

minusxlowast130

minusxlowast134

minusxlowast138

minusxlowast142

minusxlowast50

minusxlowast54

minusxlowast58

minusxlowast62

minusxlowast66

minusxlowast70

minusxlowast74

minusxlowast78

minusxlowast82

minusxlowast86

minusxlowast90

minusxlowast94

minusxlowast52

minusxlowast56

minusxlowast52

minusxlowast56

minusxlowast52

minusxlowast56

minusxlowast52

minusxlowast56

minusxlowast56

minusxlowast52

minusxlowast65

minusxlowast96

minusxlowast2

minusxlowast6

minusxlowast10

minusxlowast14

minusxlowast18

minusxlowast22

minusxlowast26

minusxlowast30

minusxlowast34

minusxlowast38

minusxlowast42

minusxlowast46

minusxlowast4

minusxlowast8

minusxlowast12

minusxlowast16

minusxlowast20

minusxlowast24

minusxlowast28

minusxlowast32

minusxlowast36

minusxlowast40

minusxlowast44

minusxlowast48

PA1B2

PA2B2

PA3B2

PA4B2

PA5B2

PA6B2

PA7B2

PA8B2

(b) Data mapping at TxApB2

0

1

11

Reso

urce

blo

ck

10

9

8

7

6

5

4

3

2

20 1 3 4 5 6 7 8 9 10 11 12 13Subframe

x98

x102

x106

x110

x114

x118

x122

x126

x130

x134

x138

x142

x50

x54

x58

x62

x66

x70

x74

x78

x82

x86

x90

x94

xlowast1

xlowast5

xlowast9

xlowast13

xlowast17

xlowast21

xlowast25

xlowast29

xlowast33

xlowast37

xlowast41

xlowast45

xlowast3

xlowast7

xlowast11

xlowast15

xlowast19

xlowast23

xlowast27

xlowast31

xlowast35

xlowast39

xlowast43

xlowast47

xlowast49

xlowast53

xlowast57

xlowast61

xlowast65

xlowast69

xlowast73

xlowast77

xlowast81

xlowast85

xlowast89

xlowast93

x52

x56

x60

x64

x68

x72

x76

x80

x84

x88

x92

x96

xlowast51

xlowast55

xlowast59

xlowast63

xlowast67

xlowast71

xlowast75

xlowast79

xlowast83

xlowast87

xlowast91

xlowast95

xlowast97

xlowast101

xlowast105

xlowast109

xlowast113

xlowast117

xlowast121

xlowast125

xlowast129

xlowast133

xlowast137

xlowast141

x100

x104

x108

x112

x116

x120

x124

x128

x132

x136

x140

x144

xlowast99

xlowast103

xlowast107

xlowast111

xlowast115

xlowast119

xlowast123

xlowast127

xlowast131

xlowast135

xlowast139

xlowast143

x2

x6

x10

x14

x18

x22

x26

x30

x34

x38

x42

x46

x4

x8

x12

x16

x20

x24

x28

x32

x36

x40

x44

x48

PA1B3

PA2B3

PA3B3

PA4B3

PA5B3

PA6B3

PA7B3

PA8B3

(c) Data mapping at TxApB3

Figure 5 Downlink data mapping of the LTE-A resource blocks for the TPULA system

International Journal of Antennas and Propagation 7

Table 1 Simulation parameters

System parameter ValueCarrier frequency 18GHzSystem bandwidth 20MHzFFT size 2048Number of data carriers 1200Number of samples in CP 144Subcarrier spacing 15 KHzModulation QPSKChannel coding NoMIMO configuration 8A lowast 3B(Tx) by 2A lowast 3B(Rx)Array spacing Half-wavelengthChannel parameter Value

Number of scatterersIndoor 64

Outdoor 4 clusters with 16scatterers per cluster (64 in total)

Scatterer radius 10m

Rx velocity 3 kmh for indoor100 kmh for outdoor

Fading FlatXPD 58 dBPractical channel estimation Zero-forcing

to separate the total number of AEs and only four AEs arededicated to a mobile user For example there is about 5 dBSNR loss at target BLER in case of employing two userscompared with the single user case Meanwhile Figure 9demonstrates the simulated throughput curves for two userswith the transmission Case 4 where system throughput isincreased a lot when two users are employed From Figure 9we can observe that the single user cases have a higherthroughput than the multiusers cases with a low SNR forexample minus10 dB mainly due to a lower BLER By increasingthe SNR the performance of multiusers gets better than thesingle user casesThis is because the proposed scheme is goodat keeping the system BLER performance When multiusersare employed the system throughput then can be guaranteedto be linearly enhanced with the number of users beingincreased

With the consideration of tradeoff regarding the spec-trum efficiency the blind channel estimation (BCE) [10]which requires no or a minimal number of pilots might bebetter employed in the TPULA system in our future researchBy using BCE not only the system spectrum efficiency can beincreased but also the error floor caused by the high mobilitymight be solved

5 Conclusions and Future Works

This paper proposes a TPULA system associated with STBCand BF techniques to exploit diversity gain under a trans-mission scenario with both low and high correlated linksThe polarized antenna branches are grouped into three setswhich steer three orthogonal beams at theX-Y Z-X and Y-Z

minus10 minus5 0 5 10 15 20 25

BLER

TPULA only outdoorTPULA w STBC outdoorTPULA w STBC and TxBF outdoorTPULA w STBC and TRxBF outdoorTPULA only indoorTPULA w STBC indoorTPULA w STBC and TxBF indoorTPULA w STBC and TRxBF indoor

TargetBLER

100

10minus1

10minus2

EbN0 (dB)

Figure 6 Simulation results of BLER for single user

0minus10 minus5 0 5 10 15

05

1

15

2

25

3

Thro

ughp

ut (b

ps)

times107

EbN0 (dB)

TPULA w STBC and TRxBF indoorTPULA w STBC and TxBF indoorTPULA w STBC indoorTPULA only indoorTPULA w STBC and TRxBF outdoorTPULA w STBC and TxBF outdoorTPULA w STBC outdoorTPULA only outdoor

Figure 7 Simulation results of throughput for single user

8 International Journal of Antennas and Propagation

minus10 minus5 0 5 10 15 20 25

BLER

TPULA w STBC and TRxBF outdoor two usersTPULA w STBC and TRxBF outdoor single userTPULA w STBC and TRxBF indoor two usersTPULA w STBC and TRxBF indoor single user

100

10minus1

10minus2

EbN0 (dB)

Figure 8 Simulation results of average BLER for two users

minus10 minus5 0 5 10 150

1

2

3

4

5

6

Thro

ughp

ut (b

ps)

TPULA w STBC and TRxBF indoor two usersTPULA w STBC and TRxBF outdoor two usersTPULA w STBC and TRxBF indoor single userTPULA w STBC and TRxBF outdoor single user

times107

EbN0 (dB)

Figure 9 Simulation results of throughput for two users

planes STBC is additionally applied at three orthogonallycolocated branches of each AE to accommodate the space-time diversity gain From the simulation results the proposedscheme in Case 4 (TPULA with SBTC and TRxBF) proveswe can obtain about 4 dB SNR gain at the target BLERcompared with Case 1 (TPULA without STBC and TRxBF)

under the indoor environment Although the performancein an outdoor environment is also increased by the proposedscheme the error-floors occur due to the high Rxmobility Inthis paper we have verified the proposed scheme under theflat fading environment only As a future work the selectivefading environment will be examined and the AE selectionscheme for OFDM-based TPULA system might be proposedto overcome the ISI caused by multipath

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgment

This research was supported by the MSIP (Ministry ofScience ICT and Future Planning) Korea under the ITRC(InformationTechnologyResearchCenter) Support Program(IITP-2015-H8501-15-1019) supervised by the IITP (Institutefor Information amp Communications Technology Promo-tion)

References

[1] M-T Dao V-A Nguyen Y-T Im S-O Park and G Yoonldquo3D polarized channel modeling and performance comparisonof MIMO antenna configurations with different polarizationsrdquoIEEE Transactions on Antennas and Propagation vol 59 no 7pp 2672ndash2682 2011

[2] A S Y Poon and D N C Tse ldquoDegree-of-freedom gain fromusing polarimetric antenna elementsrdquo Institute of Electrical andElectronics Engineers Transactions on Information Theory vol57 no 9 pp 5695ndash5709 2011

[3] X Su and K H Chang ldquoPolarized uniform linear arraysystem beam radiation pattern beamforming diversity orderand channel capacityrdquo International Journal of Antennas andPropagation vol 2015 Article ID 371236 9 pages 2015

[4] J Li U Park and S Y Kim ldquoAn efficient rate one STBCscheme with 3 transmit antennasrdquo in Proceedings of the Interna-tional Conference onWireless Communications Networking andMobile Computing (WiCOM rsquo08) pp 1ndash4 IEEE Dalian ChinaOctober 2008

[5] U Park S Kim K Lim and J Li ldquoA novel QO-STBC schemewith linear decoding for three and four transmit antennasrdquoIEEE Communications Letters vol 12 no 12 pp 868ndash870 2008

[6] V Madisetti Digital Signal Processing Handbook CRC Press2nd edition 1999

[7] L J Liu S Y Kim andM-S Lim ldquoAn efficient selective receiverfor STBC schemerdquo in Proceedings of the IEEE InternationalConference on Communications (ICC rsquo07) pp 4196ndash4200 IEEEGlasgow Scotland June 2007

[8] W Liu ldquoAdaptive wideband beamforming with sensor delay-linesrdquo Signal Processing vol 89 no 5 pp 876ndash882 2009

[9] X Su B Hui and K H Chang ldquo3-D MIMO channel mod-eling with beamforming analysis for dual-polarized antenna

International Journal of Antennas and Propagation 9

systemsrdquo in Proceedings of the IEEE 78th Vehicular TechnologyConference (VTC Fall rsquo13) pp 1ndash5 September 2013

[10] E Beres and R Adve ldquoBlind channel estimation for orthogonalSTBC in MISO systemsrdquo IEEE Transactions on Vehicular Tech-nology vol 56 no 4 pp 2042ndash2050 2007

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of

2 International Journal of Antennas and Propagation

ZZ

YX

YX

TxB1

TxB3

TxB2

RxB1

RxB2

RxB3

(a) TPMIMOB

X

Z

Y

X

Z

Y

TxA1B1

TxA1B3

TxA1B2

RxA1B1 RxAQB1

RxA1B3

RxA1B2

RxAQB3

RxAQB2

middot middot middotmiddot middot middot

TxAPB1

TxAPB3

TxAPB2

(b) TPMIMOAampB (TPULA)

Figure 1 Orthogonal TPMIMO system

2 TPULA Configuration

Figure 1 provides examples of orthogonal triple-polarizedMIMO (TPMIMO) system in the array and branch (AampB)multiple antennae configuration According to Figure 1(a)there are three orthogonally colocated antenna branchesat the transmitter (Tx) and receiver (Rx) ports Thereforethis PMIMO transceiver is defined at the branch level asTPMIMOB By uniformly aligning the polarized antennaports (AEs) at the Tx and the Rx as illustrated in Figure 1(b)a PMIMO system is defined as TPMIMOAampB which indicatesthat antenna polarization is realized at both branch and arraylevels

The beamwidth is relevant to the array configurationwhere it is inversely proportional to the number of AEsand the AE space Because the space among three colocatedbranches at each AE is fixed at zero as shown in Figure 1which makes beamwidth scale up to 360∘ the beams shouldbe generated via corresponding cross-array branches ratherthan the colocated branches at each AE [3] Therefore aTPULA system can generate three orthogonal beams asfollows

(i) beam steered and varied in the X-Y plane is formedby the set of branches of TRxA

119901(119902)B1 where 119901(119902) =

1 2 119875(119876)(ii) beam steered and varied in the Z-X plane is formed

by the set of branches of TRxA119901(119902)

B2

(iii) beam steered and varied in the Y-Z plane is formedby the set of branches of TRxA

119901(119902)B3

Figure 2 depicts the 3D beams generated by a TPULAsystem where three orthogonal beams can be steered and

varied separately on the X-Y Z-X and Y-Z planes Byapplying the TPULA into an orthogonal frequency divisionmultiplexing (OFDM) system considering STBC and BF arobust MIMO transceiver is provided as seen in Figure 3which is explained in detail in the following section

3 TPULA System withSTBC and Beamforming

As studied by [1] the cross-array links (eg from TxA1B1

to RxA1B1and RxA

2B1) are highly correlated by setting

the AE space equal to or less than a half-wavelengthThe cross-branch links (eg from TxA

1B1and TxA

1B2to

RxA1B1) however are usually uncorrelated due to the space

polarization This inspired us to incorporate STBC and BFtechniques simultaneously in the TPULA system to boostperformance Moreover in order to achieve full-rate codingwith an odd number of transmit antennae quasiorthogonalSTBC (QO-STBC) has emerged in the literature [4 5] whichfully explores diversity gain but increases the complexity ofdecoding due to nonorthogonal interference In this paperwe propose a scheme to combine the QO-STBC for threetransmit antennae with BF techniques via a TPULA systemas follows

In Figure 3 the BF weights (119908Tx119901119887

and 119908Rx119902119887

) are multipliedbefore the inverse fast Fourier transform (IFFT) block of theTx and after the FFT block of the Rx where 119901 (1 2 119875)119902 (1 2 119876) and 119887 (1 2 3) denote the indexes of the TxAERxAE and antenna branch respectively Note that we haveconsidered the digital BF where the TxBF weights obtainedby zero-forcing BF are determined by the Rx location and theRxBF weights obtained by MMSE BF are determined based

International Journal of Antennas and Propagation 3

minus1 minus05 0 05 1

minus1 minus05 0 051

minus1minus050051minus1

minus08

minus06

minus04

minus02

0

02

04

06

08

1

Z

XY

Y minus1minus050051

minus1

minus08

minus06

minus04

minus02

0

02

04

06

08

1

Z

X

Figure 2 3D beams generated by a TPULA system

Y

X

QO-STBC

Y

X

QO-STBCdecodingQO-STBCdecoding

QO-STBCdecoding

QO-STBCdecodingQO-STBCdecodingQO-STBCdecoding

QO-STBCdecodingQO-STBCdecoding

QO-STBCdecoding

Rx

Rx

combining

Tx

IFFT and GI extension GI removal and FFT

Xp

Xp

Xp

wTx12

wTx13

wTx11

A1B2A1B2

A1B3A1B3

A1B1A1B1

wTx22

wTx23

wTx21

A2B3 A2B3

A2B2

A2B1

A2B2

A2B1

wRx12

wRx13

wRx11

wRx22

wRx23

wRx21

wRxQ2

wRxQ3

wRxQ1

AQB3AQB2

AQB1

APB3

APB2

APB1

wTxP2

wTxP3

wTxP1

Figure 3 OFDM-based TPULA system with STBC and BF

on reference signal The BF weights then differentiate eachother based on the polarized antenna branches of each arrayelement instead of each subcarrierX

119901in expression (1) is the

transmitted QO-STBC symbol matrix [4] where the Romannumeral is the symbol index 119894 is the time index modulo of 4(119868 = 4) and 119895 is the antenna index modulo of 3 (119869 = 3)

X119901=(

119909II 119909III 119909IV

119909lowast

I minus119909lowast

IV 119909lowast

III

119909IV 119909I 119909II

119909lowast

III minus119909lowast

II 119909lowast

I

)

119868minus119887119910minus119869

(1)

For the cross-branch links from the pth TxAE with threebranches to a single Rx branch 119902

119887 that is h

119901sim119902119887 the received

signal before the RxBF weights multiplication is given as

R119901sim119902119887= (h119901sim119902119887(WTx119901

⊙ X119901)119879

)119879

+ n119901sim119902119887

= (WTx119901

⊙ X119901) h119879119901sim119902119887

+ n119901sim119902119887

= (119903119901sim119902119887 I 119903119901sim119902119887 II 119903119901sim119902119887 III 119903119901sim119902119887 IV)

119879

(2)

4 International Journal of Antennas and Propagation

where ldquo⊙rdquo denotes the Hadamard product and WTx119901

is theTxBF weighting matrix given by

WTx119901

=(

119908Tx1199011

119908Tx1199012

119908Tx1199013

119908Tx1199011

119908Tx1199012

119908Tx1199013

119908Tx1199011

119908Tx1199012

119908Tx1199013

119908Tx1199011

119908Tx1199012

119908Tx1199013

) (3)

h119901sim119902119887

is the vector of the polarized cross-branch links givenas

h119901sim119902119887= (ℎ1199011sim119902119887ℎ1199012sim119902119887ℎ1199013sim119902119887) (4)

and n119901sim119902119887

is the noise vector given by

n119901sim119902119887= (120590119901sim119902119887 I 120590119901sim119902119887 II 120590119901sim119902119887 III 120590119901sim119902119887 IV)

119879

(5)

The minimum mean square error (MMSE) RxBF is usedto minimize the error between the RxBF output and theavailable reference signal where the optimal weights119908Rx

119902119887

canbe obtained by theWiener-Hoff equation [6]Then the signalafter RxBF is given by

R119901sim119902119887= 119908

Rx119902119887

R119901sim119902119887 (6)

The STBC decoding applied after the RxBF gives

X119901sim119902119887= D119901sim119902119887

T119901sim119902119887

= (119909119901sim119902119887 I 119909119901sim119902119887 II 119909119901sim119902119887 III 119909119901sim119902119887 IV)

119879

(7)

Here D119901sim119902119887

is the STBC decoding matrix which is modifiedbased on equation (12) in [4] by considering TxBF weightsgiven by

D119901sim119902119887

=(

(

0 119908Tx1199011

ℎ1199011sim119902119887(119908

Tx1199012

)lowast

ℎlowast

1199012sim119902119887

119908Tx1199013

ℎ1199013sim119902119887

(119908Tx1199011

)lowast

ℎlowast

1199011sim119902119887

0 (119908Tx1199013

)lowast

ℎlowast

1199013sim119902119887

minus119908Tx1199012

ℎ1199012sim119902119887

(119908Tx1199012

)lowast

ℎlowast

1199012sim119902119887

119908Tx1199013

ℎ1199013sim119902119887

0 119908Tx1199011

ℎ1199011sim119902119887

(119908Tx1199013

)lowast

ℎlowast

1199013sim119902119887

minus119908Tx1199012

ℎ1199012sim119902119887(119908

Tx1199011

)lowast

ℎlowast

1199011sim119902119887

0

)

)

(8)

And Τ119901sim119902119887

is defined as

Τ119901sim119902119887= (119903119901sim119902119887 I 119903

lowast

119901sim119902119887 II 119903119901sim119902119887 III 119903lowast

119901sim119902119887 IV)119879 (9)

by taking the conjugate of the second and the fourth elementsof R119901sim119902119887

The final output of the proposed scheme is obtainedby the last functional block in Figure 3 (ie the Rx combin-ing) given as

X =119876

sum

119902=1

3

sum

119887=1

(119909119901sim119902119887 I 119909119901sim119902119887 II 119909119901sim119902119887 III 119909119901sim119902119887 IV)

119879

(10)

Li et al [4] used the receiver reported by Liu et al [7]to cope with the interference caused by nonorthogonality ofthe QO-STBS In this paper the complexity of interferencecancellation can be simplified by the adaptive setting of TxBFweights in (3) By calculation the decoded signal in (10) isgiven by

119909119901sim119902119887 I = (119908

Rx119902119887

)2

(10038161003816100381610038161003816119908

Tx1199011

ℎ1199011sim119902119887

10038161003816100381610038161003816

2

+10038161003816100381610038161003816119908

Tx1199012

ℎ1199012sim119902119887

10038161003816100381610038161003816

2

+10038161003816100381610038161003816119908

Tx1199013

ℎ1199013sim119902119887

10038161003816100381610038161003816

2

) 119909I⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟

Desired singal

+ (119908Rx119902119887

)2

[119908Tx1199011

ℎ1199011sim119902119887(119908

Tx1199013

)lowast

ℎlowast

1199013sim119902119887

+ 119908Tx1199013

ℎ1199013sim119902119887(119908

Tx1199011

)lowast

ℎlowast

1199011sim119902119887

] 119909III⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟

Interference caused by adjacent signal

+ (119908Rx119902119887

)2

[119908Tx1199011

ℎ1199011sim119902119887120590lowast

119901sim119902119887 II + (119908Tx1199012

)lowast

ℎlowast

1199012sim119902119887

120590119901sim119902119887 III + 119908

Tx1199013

ℎ1199013sim119902119887120590lowast

119901sim119902119887 IV]⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟

Noise

119909119901sim119902119887 II = (119908

Rx119902119887

)2

(10038161003816100381610038161003816119908

Tx1199011

ℎ1199011sim119902119887

10038161003816100381610038161003816

2

+10038161003816100381610038161003816119908

Tx1199012

ℎ1199012sim119902119887

10038161003816100381610038161003816

2

+10038161003816100381610038161003816119908

Tx1199013

ℎ1199013sim119902119887

10038161003816100381610038161003816

2

) 119909II

+ (119908Rx119902119887

)2

[119908Tx1199011

ℎ1199011sim119902119887(119908

Tx1199013

)lowast

ℎlowast

1199013sim119902119887

+ 119908Tx1199013

ℎ1199013sim119902119887(119908

Tx1199011

)lowast

ℎlowast

1199011sim119902119887

] 119909IV

+ (119908Rx119902119887

)2

[119908Tx1199011

ℎ1199011sim119902119887120590lowast

119901sim119902119887 II + (119908Tx1199012

)lowast

ℎlowast

1199012sim119902119887

120590119901sim119902119887 III + 119908

Tx1199013

ℎ1199013sim119902119887120590lowast

119901sim119902119887 IV]

119909119901sim119902119887 III = (119908

Rx119902119887

)2

(10038161003816100381610038161003816119908

Tx1199011

ℎ1199011sim119902119887

10038161003816100381610038161003816

2

+10038161003816100381610038161003816119908

Tx1199012

ℎ1199012sim119902119887

10038161003816100381610038161003816

2

+10038161003816100381610038161003816119908

Tx1199013

ℎ1199013sim119902119887

10038161003816100381610038161003816

2

) 119909III

+ (119908Rx119902119887

)2

[119908Tx1199011

ℎ1199011sim119902119887(119908

Tx1199013

)lowast

ℎlowast

1199013sim119902119887

+ 119908Tx1199013

ℎ1199013sim119902119887(119908

Tx1199011

)lowast

ℎlowast

1199011sim119902119887

] 119909I

+ (119908Rx119902119887

)2

[119908Tx1199011

ℎ1199011sim119902119887120590lowast

119901sim119902119887 II + (119908Tx1199012

)lowast

ℎlowast

1199012sim119902119887

120590119901sim119902119887 III + 119908

Tx1199013

ℎ1199013sim119902119887120590lowast

119901sim119902119887 IV]

International Journal of Antennas and Propagation 5

119909119901sim119902119887 IV = (119908

Rx119902119887

)2

(10038161003816100381610038161003816119908

Tx1199011

ℎ1199011sim119902119887

10038161003816100381610038161003816

2

+10038161003816100381610038161003816119908

Tx1199012

ℎ1199012sim119902119887

10038161003816100381610038161003816

2

+10038161003816100381610038161003816119908

Tx1199013

ℎ1199013sim119902119887

10038161003816100381610038161003816

2

) 119909IV

+ (119908Rx119902119887

)2

[119908Tx1199011

ℎ1199011sim119902119887(119908

Tx1199013

)lowast

ℎlowast

1199013sim119902119887

+ 119908Tx1199013

ℎ1199013sim119902119887(119908

Tx1199011

)lowast

ℎlowast

1199011sim119902119887

] 119909II

+ (119908Rx119902119887

)2

[119908Tx1199011

ℎ1199011sim119902119887120590lowast

119901sim119902119887 II + (119908Tx1199012

)lowast

ℎlowast

1199012sim119902119887

120590119901sim119902119887 III + 119908

Tx1199013

ℎ1199013sim119902119887120590lowast

119901sim119902119887 IV]

(11)

Formulae (11) indicate that 119909I and 119909III interfere with eachother as well as 119909II and 119909IV For interference cancellationit is better to set the scaling factor that is either 119908Tx

1199011

or119908

Tx1199013

to zero The setting of zero needs to be determined bythe off boresight angle (OBSA) effect discussed by Liu [8]where boresight refers to the axis vertical to the orientationof the array alignment For example in Figure 4 the solidbeam is generated by the set of branches of TxApB1 whilethe dotted beam is generated by the set of branches ofTxApB3 According to Liu [8] the beamwidth is known andis directly proportional to the OBSA value Therefore 119908Tx

1199013

=

0 (when 120572 lt 120573) or 119908Tx1199011

= 0 (when 120572 gt 120573) can be suggestedto avoid increasing the beamwidth Note that in this paperthe location of the Rx is assumed to be known by the TxAdditionally by using the zero-forcing TxBF we can have twooptions for TxBF weight settings that is (119908

1199011= 0 119908

1199012=

ℎminus1

1199012

1199081199013= ℎminus1

1199013

) or (1199081199013= 0 119908

1199012= ℎminus1

1199012

1199081199011= ℎminus1

1199011

) which isdeterminedmainly based on the location of the Rx Note thatwe did not focus on the TDD system therefore the channelreciprocity is not performed Since we assume the location ofRx is known by the Tx the BF weight used in the paper isinverse of the array propagation factor at Tx side [3]

According to Figure 3 the proposed TPULA system canbe easily extended as the polarized massive MIMO (PM-MIMO) system by employing a large number of AEs whichrequires a large number (tens or hundreds) of antennaeserving at both Tx and Rx Consequently the system designthen should consider the plenty of power consumption andhigh cost of RF chains Antenna selection scheme is necessaryfor PM-MIMOsystem in both transmitting end and receivingendAn energy efficient antenna selection algorithmbased onconvex optimization might be proposed where the numberof transmit antenna and the number of receive antenna needto be jointly optimized to maximize energy efficiency Goodperformance gain of energy efficiency should be obtainedcomparing with no antenna selection case

4 Performance Verification

We verified the proposed scheme via Monte Carlo simula-tions based on the LTE-A specification where Table 1 liststhe details of the parameter settings The downlink datamapping for LTE-A resource blocks used in the simulationsis demonstrated in Figure 5 which is based on the QO-STBC symbol matrix of expression (1) 119875A119901B1 119875A119901B2 and119875A119901B3 denote the pilots used for practical channel estimation

and a total of 24 pilots were used separately for 24 transmitantennae An extension of work by Su et al [9] is appliedto simulations of the polarized MIMO channel for whichchannel characteristics are also listed in Table 1 Note that thelocation of the Rx is updated periodically at each simulationloop

Sincewe have proposed the scheme to combine STBCandBF firstly and could not find any existing combined STBC-BF scheme for the performance comparison according toliteratures we only compared four transmission cases in thispaper to verify our proposal by discussing the achievementof STBC and BF gains Figure 6 shows the simulated BLERcurves including Case 1 of TPULA only (without STBC andBF) Case 2 of TPULA with STBC Case 3 of TPULA withSTBC and TxBF and Case 4 of TPULA with STBC andTRxBF where the target BLER for performance comparisonis fixed at 10minus2 According to Figure 6 Case 1 has the worstperformance because there is no space-time coding and BFgain achieved By applying the receiver studied by Liu et al [7]for interference cancellation almost 1 dB of signal-to-noiseratio (SNR) gain at the target BLER can be achievedwithCase2 under the indoor environment compared with Case 1 Inaddition compared with Case 2 about 23 dB of SNR gainunder the indoor environment at the target BLER is obtainedwith Case 3 when the TxBF is applied Compared with Case3 05 dB SNR gain at the target BLER is achieved by usingthe MMSE RxBF via Case 4 because the amount of error fedinto the STBC decoding process is further decreased Theerror floors occur under the outdoor environment due tohigh mobility at 100 kmh This indicates that the numberof pilots in the TPULA system is not enough to compensatethe channel correctly under an environment with fast time-varying phase response Again the efficiency of our proposedscheme can be quantified in terms of throughput Figure 7illustrates the simulated throughput curves By focusing onthe curves obtained under the indoor environment we canobserve that Case 4 of TPULA with STBC and TRxBF canachieve the highest throughput of 21Mbps at 119864

1198871198730of 0 dB

and 20Mbps 19Mbps and 17Mbps for Case 3 Case 2 andCase 1 at 119864

1198871198730of 0 dB respectively

In addition we evaluate the average BLER performancewhen two mobile users are employed In this case the eightAEs at Tx are divided into two groups where each group iswith four AEs and performs TxBF for a mobile user Figure 8illustrates the simulation results of average BLER for twousers with the transmission Case 4 According to Figure 8we can observe that the performance decreases a lot dueto the loss of diversity order at Tx because multiusers need

6 International Journal of Antennas and Propagation

30

210

60

240

90

270

120

300

150

330

180 0

Z

X

120573

120572

Figure 4 Setting of zero for TxBF weights

Subframe20 1 3 4 5 6 7 8 9 10 11 12 13

0

1

11

Reso

urce

blo

ck

10

9

8

7

6

5

4

3

2

x98x4xlowast1

x102x8xlowast5

x106x12xlowast9

x110x16xlowast13

x114x20xlowast17

x118x24xlowast21

x122x28xlowast25

x126x32xlowast29

x130x36xlowast33

x134x40xlowast37

x138x44xlowast41

x142

x50

x54

x58

x62

x66

x70

x74

x78

x82

x86

x90

x94x48xlowast45

xlowast3

xlowast7

xlowast11

xlowast15

xlowast19

xlowast23

xlowast27

xlowast31

xlowast35

xlowast39

xlowast43

xlowast47

xlowast49

xlowast53

xlowast57

xlowast61

xlowast65

xlowast69

xlowast73

xlowast77

xlowast81

xlowast85

xlowast89

xlowast93

x52

x56

x60

x64

x68

x72

x76

x80

x84

x88

x92

x96

xlowast51

xlowast55

xlowast59

xlowast63

xlowast67

xlowast71

xlowast75

xlowast79

xlowast83

xlowast87

xlowast91

xlowast95

xlowast97

xlowast101

xlowast105

xlowast109

xlowast113

xlowast117

xlowast121

xlowast125

xlowast129

xlowast133

xlowast137

xlowast141

x100

x104

x108

x112

x116

x120

x124

x128

x132

x136

x140

x144

xlowast99

xlowast103

xlowast107

xlowast111

xlowast115

xlowast119

xlowast123

xlowast127

xlowast131

xlowast135

xlowast139

xlowast143

PA1B1

PA2B1

PA3B1

PA4B1

PA5B1

PA6B1

PA7B1

PA8B1

x2

x6

x10

x14

x18

x22

x26

x30

x34

x38

x42

x46

(a) Data mapping at TxApB1

20 1 3 4 5 6 7 8 9 10 11 12 13

0

1

11

Reso

urce

blo

ck

10

9

8

7

6

5

4

3

2

Subframe

x1

x5

x9

x13

x17

x21

x25

x29

x33

x37

x41

x45

x3

x7

x11

x15

x19

x23

x27

x31

x35

x39

x43

x47

x49

x53

x57

x61

x65

x69

x73

x77

x81

x85

x89

x93

x51

x55

x59

x63

x67

x71

x75

x79

x83

x87

x91

x95

x97

x101

x105

x109

x113

x117

x121

x125

x129

x133

x137

x141

x99

x103

x107

x111

x115

x119

x123

x127

x131

x135

x139

x143

minusxlowast100

minusxlowast104

minusxlowast108

minusxlowast112

minusxlowast116

minusxlowast120

minusxlowast124

minusxlowast128

minusxlowast132

minusxlowast136

minusxlowast140

minusxlowast144

minusxlowast98

minusxlowast102

minusxlowast106

minusxlowast110

minusxlowast114

minusxlowast118

minusxlowast122

minusxlowast126

minusxlowast130

minusxlowast134

minusxlowast138

minusxlowast142

minusxlowast50

minusxlowast54

minusxlowast58

minusxlowast62

minusxlowast66

minusxlowast70

minusxlowast74

minusxlowast78

minusxlowast82

minusxlowast86

minusxlowast90

minusxlowast94

minusxlowast52

minusxlowast56

minusxlowast52

minusxlowast56

minusxlowast52

minusxlowast56

minusxlowast52

minusxlowast56

minusxlowast56

minusxlowast52

minusxlowast65

minusxlowast96

minusxlowast2

minusxlowast6

minusxlowast10

minusxlowast14

minusxlowast18

minusxlowast22

minusxlowast26

minusxlowast30

minusxlowast34

minusxlowast38

minusxlowast42

minusxlowast46

minusxlowast4

minusxlowast8

minusxlowast12

minusxlowast16

minusxlowast20

minusxlowast24

minusxlowast28

minusxlowast32

minusxlowast36

minusxlowast40

minusxlowast44

minusxlowast48

PA1B2

PA2B2

PA3B2

PA4B2

PA5B2

PA6B2

PA7B2

PA8B2

(b) Data mapping at TxApB2

0

1

11

Reso

urce

blo

ck

10

9

8

7

6

5

4

3

2

20 1 3 4 5 6 7 8 9 10 11 12 13Subframe

x98

x102

x106

x110

x114

x118

x122

x126

x130

x134

x138

x142

x50

x54

x58

x62

x66

x70

x74

x78

x82

x86

x90

x94

xlowast1

xlowast5

xlowast9

xlowast13

xlowast17

xlowast21

xlowast25

xlowast29

xlowast33

xlowast37

xlowast41

xlowast45

xlowast3

xlowast7

xlowast11

xlowast15

xlowast19

xlowast23

xlowast27

xlowast31

xlowast35

xlowast39

xlowast43

xlowast47

xlowast49

xlowast53

xlowast57

xlowast61

xlowast65

xlowast69

xlowast73

xlowast77

xlowast81

xlowast85

xlowast89

xlowast93

x52

x56

x60

x64

x68

x72

x76

x80

x84

x88

x92

x96

xlowast51

xlowast55

xlowast59

xlowast63

xlowast67

xlowast71

xlowast75

xlowast79

xlowast83

xlowast87

xlowast91

xlowast95

xlowast97

xlowast101

xlowast105

xlowast109

xlowast113

xlowast117

xlowast121

xlowast125

xlowast129

xlowast133

xlowast137

xlowast141

x100

x104

x108

x112

x116

x120

x124

x128

x132

x136

x140

x144

xlowast99

xlowast103

xlowast107

xlowast111

xlowast115

xlowast119

xlowast123

xlowast127

xlowast131

xlowast135

xlowast139

xlowast143

x2

x6

x10

x14

x18

x22

x26

x30

x34

x38

x42

x46

x4

x8

x12

x16

x20

x24

x28

x32

x36

x40

x44

x48

PA1B3

PA2B3

PA3B3

PA4B3

PA5B3

PA6B3

PA7B3

PA8B3

(c) Data mapping at TxApB3

Figure 5 Downlink data mapping of the LTE-A resource blocks for the TPULA system

International Journal of Antennas and Propagation 7

Table 1 Simulation parameters

System parameter ValueCarrier frequency 18GHzSystem bandwidth 20MHzFFT size 2048Number of data carriers 1200Number of samples in CP 144Subcarrier spacing 15 KHzModulation QPSKChannel coding NoMIMO configuration 8A lowast 3B(Tx) by 2A lowast 3B(Rx)Array spacing Half-wavelengthChannel parameter Value

Number of scatterersIndoor 64

Outdoor 4 clusters with 16scatterers per cluster (64 in total)

Scatterer radius 10m

Rx velocity 3 kmh for indoor100 kmh for outdoor

Fading FlatXPD 58 dBPractical channel estimation Zero-forcing

to separate the total number of AEs and only four AEs arededicated to a mobile user For example there is about 5 dBSNR loss at target BLER in case of employing two userscompared with the single user case Meanwhile Figure 9demonstrates the simulated throughput curves for two userswith the transmission Case 4 where system throughput isincreased a lot when two users are employed From Figure 9we can observe that the single user cases have a higherthroughput than the multiusers cases with a low SNR forexample minus10 dB mainly due to a lower BLER By increasingthe SNR the performance of multiusers gets better than thesingle user casesThis is because the proposed scheme is goodat keeping the system BLER performance When multiusersare employed the system throughput then can be guaranteedto be linearly enhanced with the number of users beingincreased

With the consideration of tradeoff regarding the spec-trum efficiency the blind channel estimation (BCE) [10]which requires no or a minimal number of pilots might bebetter employed in the TPULA system in our future researchBy using BCE not only the system spectrum efficiency can beincreased but also the error floor caused by the high mobilitymight be solved

5 Conclusions and Future Works

This paper proposes a TPULA system associated with STBCand BF techniques to exploit diversity gain under a trans-mission scenario with both low and high correlated linksThe polarized antenna branches are grouped into three setswhich steer three orthogonal beams at theX-Y Z-X and Y-Z

minus10 minus5 0 5 10 15 20 25

BLER

TPULA only outdoorTPULA w STBC outdoorTPULA w STBC and TxBF outdoorTPULA w STBC and TRxBF outdoorTPULA only indoorTPULA w STBC indoorTPULA w STBC and TxBF indoorTPULA w STBC and TRxBF indoor

TargetBLER

100

10minus1

10minus2

EbN0 (dB)

Figure 6 Simulation results of BLER for single user

0minus10 minus5 0 5 10 15

05

1

15

2

25

3

Thro

ughp

ut (b

ps)

times107

EbN0 (dB)

TPULA w STBC and TRxBF indoorTPULA w STBC and TxBF indoorTPULA w STBC indoorTPULA only indoorTPULA w STBC and TRxBF outdoorTPULA w STBC and TxBF outdoorTPULA w STBC outdoorTPULA only outdoor

Figure 7 Simulation results of throughput for single user

8 International Journal of Antennas and Propagation

minus10 minus5 0 5 10 15 20 25

BLER

TPULA w STBC and TRxBF outdoor two usersTPULA w STBC and TRxBF outdoor single userTPULA w STBC and TRxBF indoor two usersTPULA w STBC and TRxBF indoor single user

100

10minus1

10minus2

EbN0 (dB)

Figure 8 Simulation results of average BLER for two users

minus10 minus5 0 5 10 150

1

2

3

4

5

6

Thro

ughp

ut (b

ps)

TPULA w STBC and TRxBF indoor two usersTPULA w STBC and TRxBF outdoor two usersTPULA w STBC and TRxBF indoor single userTPULA w STBC and TRxBF outdoor single user

times107

EbN0 (dB)

Figure 9 Simulation results of throughput for two users

planes STBC is additionally applied at three orthogonallycolocated branches of each AE to accommodate the space-time diversity gain From the simulation results the proposedscheme in Case 4 (TPULA with SBTC and TRxBF) proveswe can obtain about 4 dB SNR gain at the target BLERcompared with Case 1 (TPULA without STBC and TRxBF)

under the indoor environment Although the performancein an outdoor environment is also increased by the proposedscheme the error-floors occur due to the high Rxmobility Inthis paper we have verified the proposed scheme under theflat fading environment only As a future work the selectivefading environment will be examined and the AE selectionscheme for OFDM-based TPULA system might be proposedto overcome the ISI caused by multipath

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgment

This research was supported by the MSIP (Ministry ofScience ICT and Future Planning) Korea under the ITRC(InformationTechnologyResearchCenter) Support Program(IITP-2015-H8501-15-1019) supervised by the IITP (Institutefor Information amp Communications Technology Promo-tion)

References

[1] M-T Dao V-A Nguyen Y-T Im S-O Park and G Yoonldquo3D polarized channel modeling and performance comparisonof MIMO antenna configurations with different polarizationsrdquoIEEE Transactions on Antennas and Propagation vol 59 no 7pp 2672ndash2682 2011

[2] A S Y Poon and D N C Tse ldquoDegree-of-freedom gain fromusing polarimetric antenna elementsrdquo Institute of Electrical andElectronics Engineers Transactions on Information Theory vol57 no 9 pp 5695ndash5709 2011

[3] X Su and K H Chang ldquoPolarized uniform linear arraysystem beam radiation pattern beamforming diversity orderand channel capacityrdquo International Journal of Antennas andPropagation vol 2015 Article ID 371236 9 pages 2015

[4] J Li U Park and S Y Kim ldquoAn efficient rate one STBCscheme with 3 transmit antennasrdquo in Proceedings of the Interna-tional Conference onWireless Communications Networking andMobile Computing (WiCOM rsquo08) pp 1ndash4 IEEE Dalian ChinaOctober 2008

[5] U Park S Kim K Lim and J Li ldquoA novel QO-STBC schemewith linear decoding for three and four transmit antennasrdquoIEEE Communications Letters vol 12 no 12 pp 868ndash870 2008

[6] V Madisetti Digital Signal Processing Handbook CRC Press2nd edition 1999

[7] L J Liu S Y Kim andM-S Lim ldquoAn efficient selective receiverfor STBC schemerdquo in Proceedings of the IEEE InternationalConference on Communications (ICC rsquo07) pp 4196ndash4200 IEEEGlasgow Scotland June 2007

[8] W Liu ldquoAdaptive wideband beamforming with sensor delay-linesrdquo Signal Processing vol 89 no 5 pp 876ndash882 2009

[9] X Su B Hui and K H Chang ldquo3-D MIMO channel mod-eling with beamforming analysis for dual-polarized antenna

International Journal of Antennas and Propagation 9

systemsrdquo in Proceedings of the IEEE 78th Vehicular TechnologyConference (VTC Fall rsquo13) pp 1ndash5 September 2013

[10] E Beres and R Adve ldquoBlind channel estimation for orthogonalSTBC in MISO systemsrdquo IEEE Transactions on Vehicular Tech-nology vol 56 no 4 pp 2042ndash2050 2007

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of

International Journal of Antennas and Propagation 3

minus1 minus05 0 05 1

minus1 minus05 0 051

minus1minus050051minus1

minus08

minus06

minus04

minus02

0

02

04

06

08

1

Z

XY

Y minus1minus050051

minus1

minus08

minus06

minus04

minus02

0

02

04

06

08

1

Z

X

Figure 2 3D beams generated by a TPULA system

Y

X

QO-STBC

Y

X

QO-STBCdecodingQO-STBCdecoding

QO-STBCdecoding

QO-STBCdecodingQO-STBCdecodingQO-STBCdecoding

QO-STBCdecodingQO-STBCdecoding

QO-STBCdecoding

Rx

Rx

combining

Tx

IFFT and GI extension GI removal and FFT

Xp

Xp

Xp

wTx12

wTx13

wTx11

A1B2A1B2

A1B3A1B3

A1B1A1B1

wTx22

wTx23

wTx21

A2B3 A2B3

A2B2

A2B1

A2B2

A2B1

wRx12

wRx13

wRx11

wRx22

wRx23

wRx21

wRxQ2

wRxQ3

wRxQ1

AQB3AQB2

AQB1

APB3

APB2

APB1

wTxP2

wTxP3

wTxP1

Figure 3 OFDM-based TPULA system with STBC and BF

on reference signal The BF weights then differentiate eachother based on the polarized antenna branches of each arrayelement instead of each subcarrierX

119901in expression (1) is the

transmitted QO-STBC symbol matrix [4] where the Romannumeral is the symbol index 119894 is the time index modulo of 4(119868 = 4) and 119895 is the antenna index modulo of 3 (119869 = 3)

X119901=(

119909II 119909III 119909IV

119909lowast

I minus119909lowast

IV 119909lowast

III

119909IV 119909I 119909II

119909lowast

III minus119909lowast

II 119909lowast

I

)

119868minus119887119910minus119869

(1)

For the cross-branch links from the pth TxAE with threebranches to a single Rx branch 119902

119887 that is h

119901sim119902119887 the received

signal before the RxBF weights multiplication is given as

R119901sim119902119887= (h119901sim119902119887(WTx119901

⊙ X119901)119879

)119879

+ n119901sim119902119887

= (WTx119901

⊙ X119901) h119879119901sim119902119887

+ n119901sim119902119887

= (119903119901sim119902119887 I 119903119901sim119902119887 II 119903119901sim119902119887 III 119903119901sim119902119887 IV)

119879

(2)

4 International Journal of Antennas and Propagation

where ldquo⊙rdquo denotes the Hadamard product and WTx119901

is theTxBF weighting matrix given by

WTx119901

=(

119908Tx1199011

119908Tx1199012

119908Tx1199013

119908Tx1199011

119908Tx1199012

119908Tx1199013

119908Tx1199011

119908Tx1199012

119908Tx1199013

119908Tx1199011

119908Tx1199012

119908Tx1199013

) (3)

h119901sim119902119887

is the vector of the polarized cross-branch links givenas

h119901sim119902119887= (ℎ1199011sim119902119887ℎ1199012sim119902119887ℎ1199013sim119902119887) (4)

and n119901sim119902119887

is the noise vector given by

n119901sim119902119887= (120590119901sim119902119887 I 120590119901sim119902119887 II 120590119901sim119902119887 III 120590119901sim119902119887 IV)

119879

(5)

The minimum mean square error (MMSE) RxBF is usedto minimize the error between the RxBF output and theavailable reference signal where the optimal weights119908Rx

119902119887

canbe obtained by theWiener-Hoff equation [6]Then the signalafter RxBF is given by

R119901sim119902119887= 119908

Rx119902119887

R119901sim119902119887 (6)

The STBC decoding applied after the RxBF gives

X119901sim119902119887= D119901sim119902119887

T119901sim119902119887

= (119909119901sim119902119887 I 119909119901sim119902119887 II 119909119901sim119902119887 III 119909119901sim119902119887 IV)

119879

(7)

Here D119901sim119902119887

is the STBC decoding matrix which is modifiedbased on equation (12) in [4] by considering TxBF weightsgiven by

D119901sim119902119887

=(

(

0 119908Tx1199011

ℎ1199011sim119902119887(119908

Tx1199012

)lowast

ℎlowast

1199012sim119902119887

119908Tx1199013

ℎ1199013sim119902119887

(119908Tx1199011

)lowast

ℎlowast

1199011sim119902119887

0 (119908Tx1199013

)lowast

ℎlowast

1199013sim119902119887

minus119908Tx1199012

ℎ1199012sim119902119887

(119908Tx1199012

)lowast

ℎlowast

1199012sim119902119887

119908Tx1199013

ℎ1199013sim119902119887

0 119908Tx1199011

ℎ1199011sim119902119887

(119908Tx1199013

)lowast

ℎlowast

1199013sim119902119887

minus119908Tx1199012

ℎ1199012sim119902119887(119908

Tx1199011

)lowast

ℎlowast

1199011sim119902119887

0

)

)

(8)

And Τ119901sim119902119887

is defined as

Τ119901sim119902119887= (119903119901sim119902119887 I 119903

lowast

119901sim119902119887 II 119903119901sim119902119887 III 119903lowast

119901sim119902119887 IV)119879 (9)

by taking the conjugate of the second and the fourth elementsof R119901sim119902119887

The final output of the proposed scheme is obtainedby the last functional block in Figure 3 (ie the Rx combin-ing) given as

X =119876

sum

119902=1

3

sum

119887=1

(119909119901sim119902119887 I 119909119901sim119902119887 II 119909119901sim119902119887 III 119909119901sim119902119887 IV)

119879

(10)

Li et al [4] used the receiver reported by Liu et al [7]to cope with the interference caused by nonorthogonality ofthe QO-STBS In this paper the complexity of interferencecancellation can be simplified by the adaptive setting of TxBFweights in (3) By calculation the decoded signal in (10) isgiven by

119909119901sim119902119887 I = (119908

Rx119902119887

)2

(10038161003816100381610038161003816119908

Tx1199011

ℎ1199011sim119902119887

10038161003816100381610038161003816

2

+10038161003816100381610038161003816119908

Tx1199012

ℎ1199012sim119902119887

10038161003816100381610038161003816

2

+10038161003816100381610038161003816119908

Tx1199013

ℎ1199013sim119902119887

10038161003816100381610038161003816

2

) 119909I⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟

Desired singal

+ (119908Rx119902119887

)2

[119908Tx1199011

ℎ1199011sim119902119887(119908

Tx1199013

)lowast

ℎlowast

1199013sim119902119887

+ 119908Tx1199013

ℎ1199013sim119902119887(119908

Tx1199011

)lowast

ℎlowast

1199011sim119902119887

] 119909III⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟

Interference caused by adjacent signal

+ (119908Rx119902119887

)2

[119908Tx1199011

ℎ1199011sim119902119887120590lowast

119901sim119902119887 II + (119908Tx1199012

)lowast

ℎlowast

1199012sim119902119887

120590119901sim119902119887 III + 119908

Tx1199013

ℎ1199013sim119902119887120590lowast

119901sim119902119887 IV]⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟

Noise

119909119901sim119902119887 II = (119908

Rx119902119887

)2

(10038161003816100381610038161003816119908

Tx1199011

ℎ1199011sim119902119887

10038161003816100381610038161003816

2

+10038161003816100381610038161003816119908

Tx1199012

ℎ1199012sim119902119887

10038161003816100381610038161003816

2

+10038161003816100381610038161003816119908

Tx1199013

ℎ1199013sim119902119887

10038161003816100381610038161003816

2

) 119909II

+ (119908Rx119902119887

)2

[119908Tx1199011

ℎ1199011sim119902119887(119908

Tx1199013

)lowast

ℎlowast

1199013sim119902119887

+ 119908Tx1199013

ℎ1199013sim119902119887(119908

Tx1199011

)lowast

ℎlowast

1199011sim119902119887

] 119909IV

+ (119908Rx119902119887

)2

[119908Tx1199011

ℎ1199011sim119902119887120590lowast

119901sim119902119887 II + (119908Tx1199012

)lowast

ℎlowast

1199012sim119902119887

120590119901sim119902119887 III + 119908

Tx1199013

ℎ1199013sim119902119887120590lowast

119901sim119902119887 IV]

119909119901sim119902119887 III = (119908

Rx119902119887

)2

(10038161003816100381610038161003816119908

Tx1199011

ℎ1199011sim119902119887

10038161003816100381610038161003816

2

+10038161003816100381610038161003816119908

Tx1199012

ℎ1199012sim119902119887

10038161003816100381610038161003816

2

+10038161003816100381610038161003816119908

Tx1199013

ℎ1199013sim119902119887

10038161003816100381610038161003816

2

) 119909III

+ (119908Rx119902119887

)2

[119908Tx1199011

ℎ1199011sim119902119887(119908

Tx1199013

)lowast

ℎlowast

1199013sim119902119887

+ 119908Tx1199013

ℎ1199013sim119902119887(119908

Tx1199011

)lowast

ℎlowast

1199011sim119902119887

] 119909I

+ (119908Rx119902119887

)2

[119908Tx1199011

ℎ1199011sim119902119887120590lowast

119901sim119902119887 II + (119908Tx1199012

)lowast

ℎlowast

1199012sim119902119887

120590119901sim119902119887 III + 119908

Tx1199013

ℎ1199013sim119902119887120590lowast

119901sim119902119887 IV]

International Journal of Antennas and Propagation 5

119909119901sim119902119887 IV = (119908

Rx119902119887

)2

(10038161003816100381610038161003816119908

Tx1199011

ℎ1199011sim119902119887

10038161003816100381610038161003816

2

+10038161003816100381610038161003816119908

Tx1199012

ℎ1199012sim119902119887

10038161003816100381610038161003816

2

+10038161003816100381610038161003816119908

Tx1199013

ℎ1199013sim119902119887

10038161003816100381610038161003816

2

) 119909IV

+ (119908Rx119902119887

)2

[119908Tx1199011

ℎ1199011sim119902119887(119908

Tx1199013

)lowast

ℎlowast

1199013sim119902119887

+ 119908Tx1199013

ℎ1199013sim119902119887(119908

Tx1199011

)lowast

ℎlowast

1199011sim119902119887

] 119909II

+ (119908Rx119902119887

)2

[119908Tx1199011

ℎ1199011sim119902119887120590lowast

119901sim119902119887 II + (119908Tx1199012

)lowast

ℎlowast

1199012sim119902119887

120590119901sim119902119887 III + 119908

Tx1199013

ℎ1199013sim119902119887120590lowast

119901sim119902119887 IV]

(11)

Formulae (11) indicate that 119909I and 119909III interfere with eachother as well as 119909II and 119909IV For interference cancellationit is better to set the scaling factor that is either 119908Tx

1199011

or119908

Tx1199013

to zero The setting of zero needs to be determined bythe off boresight angle (OBSA) effect discussed by Liu [8]where boresight refers to the axis vertical to the orientationof the array alignment For example in Figure 4 the solidbeam is generated by the set of branches of TxApB1 whilethe dotted beam is generated by the set of branches ofTxApB3 According to Liu [8] the beamwidth is known andis directly proportional to the OBSA value Therefore 119908Tx

1199013

=

0 (when 120572 lt 120573) or 119908Tx1199011

= 0 (when 120572 gt 120573) can be suggestedto avoid increasing the beamwidth Note that in this paperthe location of the Rx is assumed to be known by the TxAdditionally by using the zero-forcing TxBF we can have twooptions for TxBF weight settings that is (119908

1199011= 0 119908

1199012=

ℎminus1

1199012

1199081199013= ℎminus1

1199013

) or (1199081199013= 0 119908

1199012= ℎminus1

1199012

1199081199011= ℎminus1

1199011

) which isdeterminedmainly based on the location of the Rx Note thatwe did not focus on the TDD system therefore the channelreciprocity is not performed Since we assume the location ofRx is known by the Tx the BF weight used in the paper isinverse of the array propagation factor at Tx side [3]

According to Figure 3 the proposed TPULA system canbe easily extended as the polarized massive MIMO (PM-MIMO) system by employing a large number of AEs whichrequires a large number (tens or hundreds) of antennaeserving at both Tx and Rx Consequently the system designthen should consider the plenty of power consumption andhigh cost of RF chains Antenna selection scheme is necessaryfor PM-MIMOsystem in both transmitting end and receivingendAn energy efficient antenna selection algorithmbased onconvex optimization might be proposed where the numberof transmit antenna and the number of receive antenna needto be jointly optimized to maximize energy efficiency Goodperformance gain of energy efficiency should be obtainedcomparing with no antenna selection case

4 Performance Verification

We verified the proposed scheme via Monte Carlo simula-tions based on the LTE-A specification where Table 1 liststhe details of the parameter settings The downlink datamapping for LTE-A resource blocks used in the simulationsis demonstrated in Figure 5 which is based on the QO-STBC symbol matrix of expression (1) 119875A119901B1 119875A119901B2 and119875A119901B3 denote the pilots used for practical channel estimation

and a total of 24 pilots were used separately for 24 transmitantennae An extension of work by Su et al [9] is appliedto simulations of the polarized MIMO channel for whichchannel characteristics are also listed in Table 1 Note that thelocation of the Rx is updated periodically at each simulationloop

Sincewe have proposed the scheme to combine STBCandBF firstly and could not find any existing combined STBC-BF scheme for the performance comparison according toliteratures we only compared four transmission cases in thispaper to verify our proposal by discussing the achievementof STBC and BF gains Figure 6 shows the simulated BLERcurves including Case 1 of TPULA only (without STBC andBF) Case 2 of TPULA with STBC Case 3 of TPULA withSTBC and TxBF and Case 4 of TPULA with STBC andTRxBF where the target BLER for performance comparisonis fixed at 10minus2 According to Figure 6 Case 1 has the worstperformance because there is no space-time coding and BFgain achieved By applying the receiver studied by Liu et al [7]for interference cancellation almost 1 dB of signal-to-noiseratio (SNR) gain at the target BLER can be achievedwithCase2 under the indoor environment compared with Case 1 Inaddition compared with Case 2 about 23 dB of SNR gainunder the indoor environment at the target BLER is obtainedwith Case 3 when the TxBF is applied Compared with Case3 05 dB SNR gain at the target BLER is achieved by usingthe MMSE RxBF via Case 4 because the amount of error fedinto the STBC decoding process is further decreased Theerror floors occur under the outdoor environment due tohigh mobility at 100 kmh This indicates that the numberof pilots in the TPULA system is not enough to compensatethe channel correctly under an environment with fast time-varying phase response Again the efficiency of our proposedscheme can be quantified in terms of throughput Figure 7illustrates the simulated throughput curves By focusing onthe curves obtained under the indoor environment we canobserve that Case 4 of TPULA with STBC and TRxBF canachieve the highest throughput of 21Mbps at 119864

1198871198730of 0 dB

and 20Mbps 19Mbps and 17Mbps for Case 3 Case 2 andCase 1 at 119864

1198871198730of 0 dB respectively

In addition we evaluate the average BLER performancewhen two mobile users are employed In this case the eightAEs at Tx are divided into two groups where each group iswith four AEs and performs TxBF for a mobile user Figure 8illustrates the simulation results of average BLER for twousers with the transmission Case 4 According to Figure 8we can observe that the performance decreases a lot dueto the loss of diversity order at Tx because multiusers need

6 International Journal of Antennas and Propagation

30

210

60

240

90

270

120

300

150

330

180 0

Z

X

120573

120572

Figure 4 Setting of zero for TxBF weights

Subframe20 1 3 4 5 6 7 8 9 10 11 12 13

0

1

11

Reso

urce

blo

ck

10

9

8

7

6

5

4

3

2

x98x4xlowast1

x102x8xlowast5

x106x12xlowast9

x110x16xlowast13

x114x20xlowast17

x118x24xlowast21

x122x28xlowast25

x126x32xlowast29

x130x36xlowast33

x134x40xlowast37

x138x44xlowast41

x142

x50

x54

x58

x62

x66

x70

x74

x78

x82

x86

x90

x94x48xlowast45

xlowast3

xlowast7

xlowast11

xlowast15

xlowast19

xlowast23

xlowast27

xlowast31

xlowast35

xlowast39

xlowast43

xlowast47

xlowast49

xlowast53

xlowast57

xlowast61

xlowast65

xlowast69

xlowast73

xlowast77

xlowast81

xlowast85

xlowast89

xlowast93

x52

x56

x60

x64

x68

x72

x76

x80

x84

x88

x92

x96

xlowast51

xlowast55

xlowast59

xlowast63

xlowast67

xlowast71

xlowast75

xlowast79

xlowast83

xlowast87

xlowast91

xlowast95

xlowast97

xlowast101

xlowast105

xlowast109

xlowast113

xlowast117

xlowast121

xlowast125

xlowast129

xlowast133

xlowast137

xlowast141

x100

x104

x108

x112

x116

x120

x124

x128

x132

x136

x140

x144

xlowast99

xlowast103

xlowast107

xlowast111

xlowast115

xlowast119

xlowast123

xlowast127

xlowast131

xlowast135

xlowast139

xlowast143

PA1B1

PA2B1

PA3B1

PA4B1

PA5B1

PA6B1

PA7B1

PA8B1

x2

x6

x10

x14

x18

x22

x26

x30

x34

x38

x42

x46

(a) Data mapping at TxApB1

20 1 3 4 5 6 7 8 9 10 11 12 13

0

1

11

Reso

urce

blo

ck

10

9

8

7

6

5

4

3

2

Subframe

x1

x5

x9

x13

x17

x21

x25

x29

x33

x37

x41

x45

x3

x7

x11

x15

x19

x23

x27

x31

x35

x39

x43

x47

x49

x53

x57

x61

x65

x69

x73

x77

x81

x85

x89

x93

x51

x55

x59

x63

x67

x71

x75

x79

x83

x87

x91

x95

x97

x101

x105

x109

x113

x117

x121

x125

x129

x133

x137

x141

x99

x103

x107

x111

x115

x119

x123

x127

x131

x135

x139

x143

minusxlowast100

minusxlowast104

minusxlowast108

minusxlowast112

minusxlowast116

minusxlowast120

minusxlowast124

minusxlowast128

minusxlowast132

minusxlowast136

minusxlowast140

minusxlowast144

minusxlowast98

minusxlowast102

minusxlowast106

minusxlowast110

minusxlowast114

minusxlowast118

minusxlowast122

minusxlowast126

minusxlowast130

minusxlowast134

minusxlowast138

minusxlowast142

minusxlowast50

minusxlowast54

minusxlowast58

minusxlowast62

minusxlowast66

minusxlowast70

minusxlowast74

minusxlowast78

minusxlowast82

minusxlowast86

minusxlowast90

minusxlowast94

minusxlowast52

minusxlowast56

minusxlowast52

minusxlowast56

minusxlowast52

minusxlowast56

minusxlowast52

minusxlowast56

minusxlowast56

minusxlowast52

minusxlowast65

minusxlowast96

minusxlowast2

minusxlowast6

minusxlowast10

minusxlowast14

minusxlowast18

minusxlowast22

minusxlowast26

minusxlowast30

minusxlowast34

minusxlowast38

minusxlowast42

minusxlowast46

minusxlowast4

minusxlowast8

minusxlowast12

minusxlowast16

minusxlowast20

minusxlowast24

minusxlowast28

minusxlowast32

minusxlowast36

minusxlowast40

minusxlowast44

minusxlowast48

PA1B2

PA2B2

PA3B2

PA4B2

PA5B2

PA6B2

PA7B2

PA8B2

(b) Data mapping at TxApB2

0

1

11

Reso

urce

blo

ck

10

9

8

7

6

5

4

3

2

20 1 3 4 5 6 7 8 9 10 11 12 13Subframe

x98

x102

x106

x110

x114

x118

x122

x126

x130

x134

x138

x142

x50

x54

x58

x62

x66

x70

x74

x78

x82

x86

x90

x94

xlowast1

xlowast5

xlowast9

xlowast13

xlowast17

xlowast21

xlowast25

xlowast29

xlowast33

xlowast37

xlowast41

xlowast45

xlowast3

xlowast7

xlowast11

xlowast15

xlowast19

xlowast23

xlowast27

xlowast31

xlowast35

xlowast39

xlowast43

xlowast47

xlowast49

xlowast53

xlowast57

xlowast61

xlowast65

xlowast69

xlowast73

xlowast77

xlowast81

xlowast85

xlowast89

xlowast93

x52

x56

x60

x64

x68

x72

x76

x80

x84

x88

x92

x96

xlowast51

xlowast55

xlowast59

xlowast63

xlowast67

xlowast71

xlowast75

xlowast79

xlowast83

xlowast87

xlowast91

xlowast95

xlowast97

xlowast101

xlowast105

xlowast109

xlowast113

xlowast117

xlowast121

xlowast125

xlowast129

xlowast133

xlowast137

xlowast141

x100

x104

x108

x112

x116

x120

x124

x128

x132

x136

x140

x144

xlowast99

xlowast103

xlowast107

xlowast111

xlowast115

xlowast119

xlowast123

xlowast127

xlowast131

xlowast135

xlowast139

xlowast143

x2

x6

x10

x14

x18

x22

x26

x30

x34

x38

x42

x46

x4

x8

x12

x16

x20

x24

x28

x32

x36

x40

x44

x48

PA1B3

PA2B3

PA3B3

PA4B3

PA5B3

PA6B3

PA7B3

PA8B3

(c) Data mapping at TxApB3

Figure 5 Downlink data mapping of the LTE-A resource blocks for the TPULA system

International Journal of Antennas and Propagation 7

Table 1 Simulation parameters

System parameter ValueCarrier frequency 18GHzSystem bandwidth 20MHzFFT size 2048Number of data carriers 1200Number of samples in CP 144Subcarrier spacing 15 KHzModulation QPSKChannel coding NoMIMO configuration 8A lowast 3B(Tx) by 2A lowast 3B(Rx)Array spacing Half-wavelengthChannel parameter Value

Number of scatterersIndoor 64

Outdoor 4 clusters with 16scatterers per cluster (64 in total)

Scatterer radius 10m

Rx velocity 3 kmh for indoor100 kmh for outdoor

Fading FlatXPD 58 dBPractical channel estimation Zero-forcing

to separate the total number of AEs and only four AEs arededicated to a mobile user For example there is about 5 dBSNR loss at target BLER in case of employing two userscompared with the single user case Meanwhile Figure 9demonstrates the simulated throughput curves for two userswith the transmission Case 4 where system throughput isincreased a lot when two users are employed From Figure 9we can observe that the single user cases have a higherthroughput than the multiusers cases with a low SNR forexample minus10 dB mainly due to a lower BLER By increasingthe SNR the performance of multiusers gets better than thesingle user casesThis is because the proposed scheme is goodat keeping the system BLER performance When multiusersare employed the system throughput then can be guaranteedto be linearly enhanced with the number of users beingincreased

With the consideration of tradeoff regarding the spec-trum efficiency the blind channel estimation (BCE) [10]which requires no or a minimal number of pilots might bebetter employed in the TPULA system in our future researchBy using BCE not only the system spectrum efficiency can beincreased but also the error floor caused by the high mobilitymight be solved

5 Conclusions and Future Works

This paper proposes a TPULA system associated with STBCand BF techniques to exploit diversity gain under a trans-mission scenario with both low and high correlated linksThe polarized antenna branches are grouped into three setswhich steer three orthogonal beams at theX-Y Z-X and Y-Z

minus10 minus5 0 5 10 15 20 25

BLER

TPULA only outdoorTPULA w STBC outdoorTPULA w STBC and TxBF outdoorTPULA w STBC and TRxBF outdoorTPULA only indoorTPULA w STBC indoorTPULA w STBC and TxBF indoorTPULA w STBC and TRxBF indoor

TargetBLER

100

10minus1

10minus2

EbN0 (dB)

Figure 6 Simulation results of BLER for single user

0minus10 minus5 0 5 10 15

05

1

15

2

25

3

Thro

ughp

ut (b

ps)

times107

EbN0 (dB)

TPULA w STBC and TRxBF indoorTPULA w STBC and TxBF indoorTPULA w STBC indoorTPULA only indoorTPULA w STBC and TRxBF outdoorTPULA w STBC and TxBF outdoorTPULA w STBC outdoorTPULA only outdoor

Figure 7 Simulation results of throughput for single user

8 International Journal of Antennas and Propagation

minus10 minus5 0 5 10 15 20 25

BLER

TPULA w STBC and TRxBF outdoor two usersTPULA w STBC and TRxBF outdoor single userTPULA w STBC and TRxBF indoor two usersTPULA w STBC and TRxBF indoor single user

100

10minus1

10minus2

EbN0 (dB)

Figure 8 Simulation results of average BLER for two users

minus10 minus5 0 5 10 150

1

2

3

4

5

6

Thro

ughp

ut (b

ps)

TPULA w STBC and TRxBF indoor two usersTPULA w STBC and TRxBF outdoor two usersTPULA w STBC and TRxBF indoor single userTPULA w STBC and TRxBF outdoor single user

times107

EbN0 (dB)

Figure 9 Simulation results of throughput for two users

planes STBC is additionally applied at three orthogonallycolocated branches of each AE to accommodate the space-time diversity gain From the simulation results the proposedscheme in Case 4 (TPULA with SBTC and TRxBF) proveswe can obtain about 4 dB SNR gain at the target BLERcompared with Case 1 (TPULA without STBC and TRxBF)

under the indoor environment Although the performancein an outdoor environment is also increased by the proposedscheme the error-floors occur due to the high Rxmobility Inthis paper we have verified the proposed scheme under theflat fading environment only As a future work the selectivefading environment will be examined and the AE selectionscheme for OFDM-based TPULA system might be proposedto overcome the ISI caused by multipath

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgment

This research was supported by the MSIP (Ministry ofScience ICT and Future Planning) Korea under the ITRC(InformationTechnologyResearchCenter) Support Program(IITP-2015-H8501-15-1019) supervised by the IITP (Institutefor Information amp Communications Technology Promo-tion)

References

[1] M-T Dao V-A Nguyen Y-T Im S-O Park and G Yoonldquo3D polarized channel modeling and performance comparisonof MIMO antenna configurations with different polarizationsrdquoIEEE Transactions on Antennas and Propagation vol 59 no 7pp 2672ndash2682 2011

[2] A S Y Poon and D N C Tse ldquoDegree-of-freedom gain fromusing polarimetric antenna elementsrdquo Institute of Electrical andElectronics Engineers Transactions on Information Theory vol57 no 9 pp 5695ndash5709 2011

[3] X Su and K H Chang ldquoPolarized uniform linear arraysystem beam radiation pattern beamforming diversity orderand channel capacityrdquo International Journal of Antennas andPropagation vol 2015 Article ID 371236 9 pages 2015

[4] J Li U Park and S Y Kim ldquoAn efficient rate one STBCscheme with 3 transmit antennasrdquo in Proceedings of the Interna-tional Conference onWireless Communications Networking andMobile Computing (WiCOM rsquo08) pp 1ndash4 IEEE Dalian ChinaOctober 2008

[5] U Park S Kim K Lim and J Li ldquoA novel QO-STBC schemewith linear decoding for three and four transmit antennasrdquoIEEE Communications Letters vol 12 no 12 pp 868ndash870 2008

[6] V Madisetti Digital Signal Processing Handbook CRC Press2nd edition 1999

[7] L J Liu S Y Kim andM-S Lim ldquoAn efficient selective receiverfor STBC schemerdquo in Proceedings of the IEEE InternationalConference on Communications (ICC rsquo07) pp 4196ndash4200 IEEEGlasgow Scotland June 2007

[8] W Liu ldquoAdaptive wideband beamforming with sensor delay-linesrdquo Signal Processing vol 89 no 5 pp 876ndash882 2009

[9] X Su B Hui and K H Chang ldquo3-D MIMO channel mod-eling with beamforming analysis for dual-polarized antenna

International Journal of Antennas and Propagation 9

systemsrdquo in Proceedings of the IEEE 78th Vehicular TechnologyConference (VTC Fall rsquo13) pp 1ndash5 September 2013

[10] E Beres and R Adve ldquoBlind channel estimation for orthogonalSTBC in MISO systemsrdquo IEEE Transactions on Vehicular Tech-nology vol 56 no 4 pp 2042ndash2050 2007

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of

4 International Journal of Antennas and Propagation

where ldquo⊙rdquo denotes the Hadamard product and WTx119901

is theTxBF weighting matrix given by

WTx119901

=(

119908Tx1199011

119908Tx1199012

119908Tx1199013

119908Tx1199011

119908Tx1199012

119908Tx1199013

119908Tx1199011

119908Tx1199012

119908Tx1199013

119908Tx1199011

119908Tx1199012

119908Tx1199013

) (3)

h119901sim119902119887

is the vector of the polarized cross-branch links givenas

h119901sim119902119887= (ℎ1199011sim119902119887ℎ1199012sim119902119887ℎ1199013sim119902119887) (4)

and n119901sim119902119887

is the noise vector given by

n119901sim119902119887= (120590119901sim119902119887 I 120590119901sim119902119887 II 120590119901sim119902119887 III 120590119901sim119902119887 IV)

119879

(5)

The minimum mean square error (MMSE) RxBF is usedto minimize the error between the RxBF output and theavailable reference signal where the optimal weights119908Rx

119902119887

canbe obtained by theWiener-Hoff equation [6]Then the signalafter RxBF is given by

R119901sim119902119887= 119908

Rx119902119887

R119901sim119902119887 (6)

The STBC decoding applied after the RxBF gives

X119901sim119902119887= D119901sim119902119887

T119901sim119902119887

= (119909119901sim119902119887 I 119909119901sim119902119887 II 119909119901sim119902119887 III 119909119901sim119902119887 IV)

119879

(7)

Here D119901sim119902119887

is the STBC decoding matrix which is modifiedbased on equation (12) in [4] by considering TxBF weightsgiven by

D119901sim119902119887

=(

(

0 119908Tx1199011

ℎ1199011sim119902119887(119908

Tx1199012

)lowast

ℎlowast

1199012sim119902119887

119908Tx1199013

ℎ1199013sim119902119887

(119908Tx1199011

)lowast

ℎlowast

1199011sim119902119887

0 (119908Tx1199013

)lowast

ℎlowast

1199013sim119902119887

minus119908Tx1199012

ℎ1199012sim119902119887

(119908Tx1199012

)lowast

ℎlowast

1199012sim119902119887

119908Tx1199013

ℎ1199013sim119902119887

0 119908Tx1199011

ℎ1199011sim119902119887

(119908Tx1199013

)lowast

ℎlowast

1199013sim119902119887

minus119908Tx1199012

ℎ1199012sim119902119887(119908

Tx1199011

)lowast

ℎlowast

1199011sim119902119887

0

)

)

(8)

And Τ119901sim119902119887

is defined as

Τ119901sim119902119887= (119903119901sim119902119887 I 119903

lowast

119901sim119902119887 II 119903119901sim119902119887 III 119903lowast

119901sim119902119887 IV)119879 (9)

by taking the conjugate of the second and the fourth elementsof R119901sim119902119887

The final output of the proposed scheme is obtainedby the last functional block in Figure 3 (ie the Rx combin-ing) given as

X =119876

sum

119902=1

3

sum

119887=1

(119909119901sim119902119887 I 119909119901sim119902119887 II 119909119901sim119902119887 III 119909119901sim119902119887 IV)

119879

(10)

Li et al [4] used the receiver reported by Liu et al [7]to cope with the interference caused by nonorthogonality ofthe QO-STBS In this paper the complexity of interferencecancellation can be simplified by the adaptive setting of TxBFweights in (3) By calculation the decoded signal in (10) isgiven by

119909119901sim119902119887 I = (119908

Rx119902119887

)2

(10038161003816100381610038161003816119908

Tx1199011

ℎ1199011sim119902119887

10038161003816100381610038161003816

2

+10038161003816100381610038161003816119908

Tx1199012

ℎ1199012sim119902119887

10038161003816100381610038161003816

2

+10038161003816100381610038161003816119908

Tx1199013

ℎ1199013sim119902119887

10038161003816100381610038161003816

2

) 119909I⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟

Desired singal

+ (119908Rx119902119887

)2

[119908Tx1199011

ℎ1199011sim119902119887(119908

Tx1199013

)lowast

ℎlowast

1199013sim119902119887

+ 119908Tx1199013

ℎ1199013sim119902119887(119908

Tx1199011

)lowast

ℎlowast

1199011sim119902119887

] 119909III⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟

Interference caused by adjacent signal

+ (119908Rx119902119887

)2

[119908Tx1199011

ℎ1199011sim119902119887120590lowast

119901sim119902119887 II + (119908Tx1199012

)lowast

ℎlowast

1199012sim119902119887

120590119901sim119902119887 III + 119908

Tx1199013

ℎ1199013sim119902119887120590lowast

119901sim119902119887 IV]⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟

Noise

119909119901sim119902119887 II = (119908

Rx119902119887

)2

(10038161003816100381610038161003816119908

Tx1199011

ℎ1199011sim119902119887

10038161003816100381610038161003816

2

+10038161003816100381610038161003816119908

Tx1199012

ℎ1199012sim119902119887

10038161003816100381610038161003816

2

+10038161003816100381610038161003816119908

Tx1199013

ℎ1199013sim119902119887

10038161003816100381610038161003816

2

) 119909II

+ (119908Rx119902119887

)2

[119908Tx1199011

ℎ1199011sim119902119887(119908

Tx1199013

)lowast

ℎlowast

1199013sim119902119887

+ 119908Tx1199013

ℎ1199013sim119902119887(119908

Tx1199011

)lowast

ℎlowast

1199011sim119902119887

] 119909IV

+ (119908Rx119902119887

)2

[119908Tx1199011

ℎ1199011sim119902119887120590lowast

119901sim119902119887 II + (119908Tx1199012

)lowast

ℎlowast

1199012sim119902119887

120590119901sim119902119887 III + 119908

Tx1199013

ℎ1199013sim119902119887120590lowast

119901sim119902119887 IV]

119909119901sim119902119887 III = (119908

Rx119902119887

)2

(10038161003816100381610038161003816119908

Tx1199011

ℎ1199011sim119902119887

10038161003816100381610038161003816

2

+10038161003816100381610038161003816119908

Tx1199012

ℎ1199012sim119902119887

10038161003816100381610038161003816

2

+10038161003816100381610038161003816119908

Tx1199013

ℎ1199013sim119902119887

10038161003816100381610038161003816

2

) 119909III

+ (119908Rx119902119887

)2

[119908Tx1199011

ℎ1199011sim119902119887(119908

Tx1199013

)lowast

ℎlowast

1199013sim119902119887

+ 119908Tx1199013

ℎ1199013sim119902119887(119908

Tx1199011

)lowast

ℎlowast

1199011sim119902119887

] 119909I

+ (119908Rx119902119887

)2

[119908Tx1199011

ℎ1199011sim119902119887120590lowast

119901sim119902119887 II + (119908Tx1199012

)lowast

ℎlowast

1199012sim119902119887

120590119901sim119902119887 III + 119908

Tx1199013

ℎ1199013sim119902119887120590lowast

119901sim119902119887 IV]

International Journal of Antennas and Propagation 5

119909119901sim119902119887 IV = (119908

Rx119902119887

)2

(10038161003816100381610038161003816119908

Tx1199011

ℎ1199011sim119902119887

10038161003816100381610038161003816

2

+10038161003816100381610038161003816119908

Tx1199012

ℎ1199012sim119902119887

10038161003816100381610038161003816

2

+10038161003816100381610038161003816119908

Tx1199013

ℎ1199013sim119902119887

10038161003816100381610038161003816

2

) 119909IV

+ (119908Rx119902119887

)2

[119908Tx1199011

ℎ1199011sim119902119887(119908

Tx1199013

)lowast

ℎlowast

1199013sim119902119887

+ 119908Tx1199013

ℎ1199013sim119902119887(119908

Tx1199011

)lowast

ℎlowast

1199011sim119902119887

] 119909II

+ (119908Rx119902119887

)2

[119908Tx1199011

ℎ1199011sim119902119887120590lowast

119901sim119902119887 II + (119908Tx1199012

)lowast

ℎlowast

1199012sim119902119887

120590119901sim119902119887 III + 119908

Tx1199013

ℎ1199013sim119902119887120590lowast

119901sim119902119887 IV]

(11)

Formulae (11) indicate that 119909I and 119909III interfere with eachother as well as 119909II and 119909IV For interference cancellationit is better to set the scaling factor that is either 119908Tx

1199011

or119908

Tx1199013

to zero The setting of zero needs to be determined bythe off boresight angle (OBSA) effect discussed by Liu [8]where boresight refers to the axis vertical to the orientationof the array alignment For example in Figure 4 the solidbeam is generated by the set of branches of TxApB1 whilethe dotted beam is generated by the set of branches ofTxApB3 According to Liu [8] the beamwidth is known andis directly proportional to the OBSA value Therefore 119908Tx

1199013

=

0 (when 120572 lt 120573) or 119908Tx1199011

= 0 (when 120572 gt 120573) can be suggestedto avoid increasing the beamwidth Note that in this paperthe location of the Rx is assumed to be known by the TxAdditionally by using the zero-forcing TxBF we can have twooptions for TxBF weight settings that is (119908

1199011= 0 119908

1199012=

ℎminus1

1199012

1199081199013= ℎminus1

1199013

) or (1199081199013= 0 119908

1199012= ℎminus1

1199012

1199081199011= ℎminus1

1199011

) which isdeterminedmainly based on the location of the Rx Note thatwe did not focus on the TDD system therefore the channelreciprocity is not performed Since we assume the location ofRx is known by the Tx the BF weight used in the paper isinverse of the array propagation factor at Tx side [3]

According to Figure 3 the proposed TPULA system canbe easily extended as the polarized massive MIMO (PM-MIMO) system by employing a large number of AEs whichrequires a large number (tens or hundreds) of antennaeserving at both Tx and Rx Consequently the system designthen should consider the plenty of power consumption andhigh cost of RF chains Antenna selection scheme is necessaryfor PM-MIMOsystem in both transmitting end and receivingendAn energy efficient antenna selection algorithmbased onconvex optimization might be proposed where the numberof transmit antenna and the number of receive antenna needto be jointly optimized to maximize energy efficiency Goodperformance gain of energy efficiency should be obtainedcomparing with no antenna selection case

4 Performance Verification

We verified the proposed scheme via Monte Carlo simula-tions based on the LTE-A specification where Table 1 liststhe details of the parameter settings The downlink datamapping for LTE-A resource blocks used in the simulationsis demonstrated in Figure 5 which is based on the QO-STBC symbol matrix of expression (1) 119875A119901B1 119875A119901B2 and119875A119901B3 denote the pilots used for practical channel estimation

and a total of 24 pilots were used separately for 24 transmitantennae An extension of work by Su et al [9] is appliedto simulations of the polarized MIMO channel for whichchannel characteristics are also listed in Table 1 Note that thelocation of the Rx is updated periodically at each simulationloop

Sincewe have proposed the scheme to combine STBCandBF firstly and could not find any existing combined STBC-BF scheme for the performance comparison according toliteratures we only compared four transmission cases in thispaper to verify our proposal by discussing the achievementof STBC and BF gains Figure 6 shows the simulated BLERcurves including Case 1 of TPULA only (without STBC andBF) Case 2 of TPULA with STBC Case 3 of TPULA withSTBC and TxBF and Case 4 of TPULA with STBC andTRxBF where the target BLER for performance comparisonis fixed at 10minus2 According to Figure 6 Case 1 has the worstperformance because there is no space-time coding and BFgain achieved By applying the receiver studied by Liu et al [7]for interference cancellation almost 1 dB of signal-to-noiseratio (SNR) gain at the target BLER can be achievedwithCase2 under the indoor environment compared with Case 1 Inaddition compared with Case 2 about 23 dB of SNR gainunder the indoor environment at the target BLER is obtainedwith Case 3 when the TxBF is applied Compared with Case3 05 dB SNR gain at the target BLER is achieved by usingthe MMSE RxBF via Case 4 because the amount of error fedinto the STBC decoding process is further decreased Theerror floors occur under the outdoor environment due tohigh mobility at 100 kmh This indicates that the numberof pilots in the TPULA system is not enough to compensatethe channel correctly under an environment with fast time-varying phase response Again the efficiency of our proposedscheme can be quantified in terms of throughput Figure 7illustrates the simulated throughput curves By focusing onthe curves obtained under the indoor environment we canobserve that Case 4 of TPULA with STBC and TRxBF canachieve the highest throughput of 21Mbps at 119864

1198871198730of 0 dB

and 20Mbps 19Mbps and 17Mbps for Case 3 Case 2 andCase 1 at 119864

1198871198730of 0 dB respectively

In addition we evaluate the average BLER performancewhen two mobile users are employed In this case the eightAEs at Tx are divided into two groups where each group iswith four AEs and performs TxBF for a mobile user Figure 8illustrates the simulation results of average BLER for twousers with the transmission Case 4 According to Figure 8we can observe that the performance decreases a lot dueto the loss of diversity order at Tx because multiusers need

6 International Journal of Antennas and Propagation

30

210

60

240

90

270

120

300

150

330

180 0

Z

X

120573

120572

Figure 4 Setting of zero for TxBF weights

Subframe20 1 3 4 5 6 7 8 9 10 11 12 13

0

1

11

Reso

urce

blo

ck

10

9

8

7

6

5

4

3

2

x98x4xlowast1

x102x8xlowast5

x106x12xlowast9

x110x16xlowast13

x114x20xlowast17

x118x24xlowast21

x122x28xlowast25

x126x32xlowast29

x130x36xlowast33

x134x40xlowast37

x138x44xlowast41

x142

x50

x54

x58

x62

x66

x70

x74

x78

x82

x86

x90

x94x48xlowast45

xlowast3

xlowast7

xlowast11

xlowast15

xlowast19

xlowast23

xlowast27

xlowast31

xlowast35

xlowast39

xlowast43

xlowast47

xlowast49

xlowast53

xlowast57

xlowast61

xlowast65

xlowast69

xlowast73

xlowast77

xlowast81

xlowast85

xlowast89

xlowast93

x52

x56

x60

x64

x68

x72

x76

x80

x84

x88

x92

x96

xlowast51

xlowast55

xlowast59

xlowast63

xlowast67

xlowast71

xlowast75

xlowast79

xlowast83

xlowast87

xlowast91

xlowast95

xlowast97

xlowast101

xlowast105

xlowast109

xlowast113

xlowast117

xlowast121

xlowast125

xlowast129

xlowast133

xlowast137

xlowast141

x100

x104

x108

x112

x116

x120

x124

x128

x132

x136

x140

x144

xlowast99

xlowast103

xlowast107

xlowast111

xlowast115

xlowast119

xlowast123

xlowast127

xlowast131

xlowast135

xlowast139

xlowast143

PA1B1

PA2B1

PA3B1

PA4B1

PA5B1

PA6B1

PA7B1

PA8B1

x2

x6

x10

x14

x18

x22

x26

x30

x34

x38

x42

x46

(a) Data mapping at TxApB1

20 1 3 4 5 6 7 8 9 10 11 12 13

0

1

11

Reso

urce

blo

ck

10

9

8

7

6

5

4

3

2

Subframe

x1

x5

x9

x13

x17

x21

x25

x29

x33

x37

x41

x45

x3

x7

x11

x15

x19

x23

x27

x31

x35

x39

x43

x47

x49

x53

x57

x61

x65

x69

x73

x77

x81

x85

x89

x93

x51

x55

x59

x63

x67

x71

x75

x79

x83

x87

x91

x95

x97

x101

x105

x109

x113

x117

x121

x125

x129

x133

x137

x141

x99

x103

x107

x111

x115

x119

x123

x127

x131

x135

x139

x143

minusxlowast100

minusxlowast104

minusxlowast108

minusxlowast112

minusxlowast116

minusxlowast120

minusxlowast124

minusxlowast128

minusxlowast132

minusxlowast136

minusxlowast140

minusxlowast144

minusxlowast98

minusxlowast102

minusxlowast106

minusxlowast110

minusxlowast114

minusxlowast118

minusxlowast122

minusxlowast126

minusxlowast130

minusxlowast134

minusxlowast138

minusxlowast142

minusxlowast50

minusxlowast54

minusxlowast58

minusxlowast62

minusxlowast66

minusxlowast70

minusxlowast74

minusxlowast78

minusxlowast82

minusxlowast86

minusxlowast90

minusxlowast94

minusxlowast52

minusxlowast56

minusxlowast52

minusxlowast56

minusxlowast52

minusxlowast56

minusxlowast52

minusxlowast56

minusxlowast56

minusxlowast52

minusxlowast65

minusxlowast96

minusxlowast2

minusxlowast6

minusxlowast10

minusxlowast14

minusxlowast18

minusxlowast22

minusxlowast26

minusxlowast30

minusxlowast34

minusxlowast38

minusxlowast42

minusxlowast46

minusxlowast4

minusxlowast8

minusxlowast12

minusxlowast16

minusxlowast20

minusxlowast24

minusxlowast28

minusxlowast32

minusxlowast36

minusxlowast40

minusxlowast44

minusxlowast48

PA1B2

PA2B2

PA3B2

PA4B2

PA5B2

PA6B2

PA7B2

PA8B2

(b) Data mapping at TxApB2

0

1

11

Reso

urce

blo

ck

10

9

8

7

6

5

4

3

2

20 1 3 4 5 6 7 8 9 10 11 12 13Subframe

x98

x102

x106

x110

x114

x118

x122

x126

x130

x134

x138

x142

x50

x54

x58

x62

x66

x70

x74

x78

x82

x86

x90

x94

xlowast1

xlowast5

xlowast9

xlowast13

xlowast17

xlowast21

xlowast25

xlowast29

xlowast33

xlowast37

xlowast41

xlowast45

xlowast3

xlowast7

xlowast11

xlowast15

xlowast19

xlowast23

xlowast27

xlowast31

xlowast35

xlowast39

xlowast43

xlowast47

xlowast49

xlowast53

xlowast57

xlowast61

xlowast65

xlowast69

xlowast73

xlowast77

xlowast81

xlowast85

xlowast89

xlowast93

x52

x56

x60

x64

x68

x72

x76

x80

x84

x88

x92

x96

xlowast51

xlowast55

xlowast59

xlowast63

xlowast67

xlowast71

xlowast75

xlowast79

xlowast83

xlowast87

xlowast91

xlowast95

xlowast97

xlowast101

xlowast105

xlowast109

xlowast113

xlowast117

xlowast121

xlowast125

xlowast129

xlowast133

xlowast137

xlowast141

x100

x104

x108

x112

x116

x120

x124

x128

x132

x136

x140

x144

xlowast99

xlowast103

xlowast107

xlowast111

xlowast115

xlowast119

xlowast123

xlowast127

xlowast131

xlowast135

xlowast139

xlowast143

x2

x6

x10

x14

x18

x22

x26

x30

x34

x38

x42

x46

x4

x8

x12

x16

x20

x24

x28

x32

x36

x40

x44

x48

PA1B3

PA2B3

PA3B3

PA4B3

PA5B3

PA6B3

PA7B3

PA8B3

(c) Data mapping at TxApB3

Figure 5 Downlink data mapping of the LTE-A resource blocks for the TPULA system

International Journal of Antennas and Propagation 7

Table 1 Simulation parameters

System parameter ValueCarrier frequency 18GHzSystem bandwidth 20MHzFFT size 2048Number of data carriers 1200Number of samples in CP 144Subcarrier spacing 15 KHzModulation QPSKChannel coding NoMIMO configuration 8A lowast 3B(Tx) by 2A lowast 3B(Rx)Array spacing Half-wavelengthChannel parameter Value

Number of scatterersIndoor 64

Outdoor 4 clusters with 16scatterers per cluster (64 in total)

Scatterer radius 10m

Rx velocity 3 kmh for indoor100 kmh for outdoor

Fading FlatXPD 58 dBPractical channel estimation Zero-forcing

to separate the total number of AEs and only four AEs arededicated to a mobile user For example there is about 5 dBSNR loss at target BLER in case of employing two userscompared with the single user case Meanwhile Figure 9demonstrates the simulated throughput curves for two userswith the transmission Case 4 where system throughput isincreased a lot when two users are employed From Figure 9we can observe that the single user cases have a higherthroughput than the multiusers cases with a low SNR forexample minus10 dB mainly due to a lower BLER By increasingthe SNR the performance of multiusers gets better than thesingle user casesThis is because the proposed scheme is goodat keeping the system BLER performance When multiusersare employed the system throughput then can be guaranteedto be linearly enhanced with the number of users beingincreased

With the consideration of tradeoff regarding the spec-trum efficiency the blind channel estimation (BCE) [10]which requires no or a minimal number of pilots might bebetter employed in the TPULA system in our future researchBy using BCE not only the system spectrum efficiency can beincreased but also the error floor caused by the high mobilitymight be solved

5 Conclusions and Future Works

This paper proposes a TPULA system associated with STBCand BF techniques to exploit diversity gain under a trans-mission scenario with both low and high correlated linksThe polarized antenna branches are grouped into three setswhich steer three orthogonal beams at theX-Y Z-X and Y-Z

minus10 minus5 0 5 10 15 20 25

BLER

TPULA only outdoorTPULA w STBC outdoorTPULA w STBC and TxBF outdoorTPULA w STBC and TRxBF outdoorTPULA only indoorTPULA w STBC indoorTPULA w STBC and TxBF indoorTPULA w STBC and TRxBF indoor

TargetBLER

100

10minus1

10minus2

EbN0 (dB)

Figure 6 Simulation results of BLER for single user

0minus10 minus5 0 5 10 15

05

1

15

2

25

3

Thro

ughp

ut (b

ps)

times107

EbN0 (dB)

TPULA w STBC and TRxBF indoorTPULA w STBC and TxBF indoorTPULA w STBC indoorTPULA only indoorTPULA w STBC and TRxBF outdoorTPULA w STBC and TxBF outdoorTPULA w STBC outdoorTPULA only outdoor

Figure 7 Simulation results of throughput for single user

8 International Journal of Antennas and Propagation

minus10 minus5 0 5 10 15 20 25

BLER

TPULA w STBC and TRxBF outdoor two usersTPULA w STBC and TRxBF outdoor single userTPULA w STBC and TRxBF indoor two usersTPULA w STBC and TRxBF indoor single user

100

10minus1

10minus2

EbN0 (dB)

Figure 8 Simulation results of average BLER for two users

minus10 minus5 0 5 10 150

1

2

3

4

5

6

Thro

ughp

ut (b

ps)

TPULA w STBC and TRxBF indoor two usersTPULA w STBC and TRxBF outdoor two usersTPULA w STBC and TRxBF indoor single userTPULA w STBC and TRxBF outdoor single user

times107

EbN0 (dB)

Figure 9 Simulation results of throughput for two users

planes STBC is additionally applied at three orthogonallycolocated branches of each AE to accommodate the space-time diversity gain From the simulation results the proposedscheme in Case 4 (TPULA with SBTC and TRxBF) proveswe can obtain about 4 dB SNR gain at the target BLERcompared with Case 1 (TPULA without STBC and TRxBF)

under the indoor environment Although the performancein an outdoor environment is also increased by the proposedscheme the error-floors occur due to the high Rxmobility Inthis paper we have verified the proposed scheme under theflat fading environment only As a future work the selectivefading environment will be examined and the AE selectionscheme for OFDM-based TPULA system might be proposedto overcome the ISI caused by multipath

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgment

This research was supported by the MSIP (Ministry ofScience ICT and Future Planning) Korea under the ITRC(InformationTechnologyResearchCenter) Support Program(IITP-2015-H8501-15-1019) supervised by the IITP (Institutefor Information amp Communications Technology Promo-tion)

References

[1] M-T Dao V-A Nguyen Y-T Im S-O Park and G Yoonldquo3D polarized channel modeling and performance comparisonof MIMO antenna configurations with different polarizationsrdquoIEEE Transactions on Antennas and Propagation vol 59 no 7pp 2672ndash2682 2011

[2] A S Y Poon and D N C Tse ldquoDegree-of-freedom gain fromusing polarimetric antenna elementsrdquo Institute of Electrical andElectronics Engineers Transactions on Information Theory vol57 no 9 pp 5695ndash5709 2011

[3] X Su and K H Chang ldquoPolarized uniform linear arraysystem beam radiation pattern beamforming diversity orderand channel capacityrdquo International Journal of Antennas andPropagation vol 2015 Article ID 371236 9 pages 2015

[4] J Li U Park and S Y Kim ldquoAn efficient rate one STBCscheme with 3 transmit antennasrdquo in Proceedings of the Interna-tional Conference onWireless Communications Networking andMobile Computing (WiCOM rsquo08) pp 1ndash4 IEEE Dalian ChinaOctober 2008

[5] U Park S Kim K Lim and J Li ldquoA novel QO-STBC schemewith linear decoding for three and four transmit antennasrdquoIEEE Communications Letters vol 12 no 12 pp 868ndash870 2008

[6] V Madisetti Digital Signal Processing Handbook CRC Press2nd edition 1999

[7] L J Liu S Y Kim andM-S Lim ldquoAn efficient selective receiverfor STBC schemerdquo in Proceedings of the IEEE InternationalConference on Communications (ICC rsquo07) pp 4196ndash4200 IEEEGlasgow Scotland June 2007

[8] W Liu ldquoAdaptive wideband beamforming with sensor delay-linesrdquo Signal Processing vol 89 no 5 pp 876ndash882 2009

[9] X Su B Hui and K H Chang ldquo3-D MIMO channel mod-eling with beamforming analysis for dual-polarized antenna

International Journal of Antennas and Propagation 9

systemsrdquo in Proceedings of the IEEE 78th Vehicular TechnologyConference (VTC Fall rsquo13) pp 1ndash5 September 2013

[10] E Beres and R Adve ldquoBlind channel estimation for orthogonalSTBC in MISO systemsrdquo IEEE Transactions on Vehicular Tech-nology vol 56 no 4 pp 2042ndash2050 2007

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of

International Journal of Antennas and Propagation 5

119909119901sim119902119887 IV = (119908

Rx119902119887

)2

(10038161003816100381610038161003816119908

Tx1199011

ℎ1199011sim119902119887

10038161003816100381610038161003816

2

+10038161003816100381610038161003816119908

Tx1199012

ℎ1199012sim119902119887

10038161003816100381610038161003816

2

+10038161003816100381610038161003816119908

Tx1199013

ℎ1199013sim119902119887

10038161003816100381610038161003816

2

) 119909IV

+ (119908Rx119902119887

)2

[119908Tx1199011

ℎ1199011sim119902119887(119908

Tx1199013

)lowast

ℎlowast

1199013sim119902119887

+ 119908Tx1199013

ℎ1199013sim119902119887(119908

Tx1199011

)lowast

ℎlowast

1199011sim119902119887

] 119909II

+ (119908Rx119902119887

)2

[119908Tx1199011

ℎ1199011sim119902119887120590lowast

119901sim119902119887 II + (119908Tx1199012

)lowast

ℎlowast

1199012sim119902119887

120590119901sim119902119887 III + 119908

Tx1199013

ℎ1199013sim119902119887120590lowast

119901sim119902119887 IV]

(11)

Formulae (11) indicate that 119909I and 119909III interfere with eachother as well as 119909II and 119909IV For interference cancellationit is better to set the scaling factor that is either 119908Tx

1199011

or119908

Tx1199013

to zero The setting of zero needs to be determined bythe off boresight angle (OBSA) effect discussed by Liu [8]where boresight refers to the axis vertical to the orientationof the array alignment For example in Figure 4 the solidbeam is generated by the set of branches of TxApB1 whilethe dotted beam is generated by the set of branches ofTxApB3 According to Liu [8] the beamwidth is known andis directly proportional to the OBSA value Therefore 119908Tx

1199013

=

0 (when 120572 lt 120573) or 119908Tx1199011

= 0 (when 120572 gt 120573) can be suggestedto avoid increasing the beamwidth Note that in this paperthe location of the Rx is assumed to be known by the TxAdditionally by using the zero-forcing TxBF we can have twooptions for TxBF weight settings that is (119908

1199011= 0 119908

1199012=

ℎminus1

1199012

1199081199013= ℎminus1

1199013

) or (1199081199013= 0 119908

1199012= ℎminus1

1199012

1199081199011= ℎminus1

1199011

) which isdeterminedmainly based on the location of the Rx Note thatwe did not focus on the TDD system therefore the channelreciprocity is not performed Since we assume the location ofRx is known by the Tx the BF weight used in the paper isinverse of the array propagation factor at Tx side [3]

According to Figure 3 the proposed TPULA system canbe easily extended as the polarized massive MIMO (PM-MIMO) system by employing a large number of AEs whichrequires a large number (tens or hundreds) of antennaeserving at both Tx and Rx Consequently the system designthen should consider the plenty of power consumption andhigh cost of RF chains Antenna selection scheme is necessaryfor PM-MIMOsystem in both transmitting end and receivingendAn energy efficient antenna selection algorithmbased onconvex optimization might be proposed where the numberof transmit antenna and the number of receive antenna needto be jointly optimized to maximize energy efficiency Goodperformance gain of energy efficiency should be obtainedcomparing with no antenna selection case

4 Performance Verification

We verified the proposed scheme via Monte Carlo simula-tions based on the LTE-A specification where Table 1 liststhe details of the parameter settings The downlink datamapping for LTE-A resource blocks used in the simulationsis demonstrated in Figure 5 which is based on the QO-STBC symbol matrix of expression (1) 119875A119901B1 119875A119901B2 and119875A119901B3 denote the pilots used for practical channel estimation

and a total of 24 pilots were used separately for 24 transmitantennae An extension of work by Su et al [9] is appliedto simulations of the polarized MIMO channel for whichchannel characteristics are also listed in Table 1 Note that thelocation of the Rx is updated periodically at each simulationloop

Sincewe have proposed the scheme to combine STBCandBF firstly and could not find any existing combined STBC-BF scheme for the performance comparison according toliteratures we only compared four transmission cases in thispaper to verify our proposal by discussing the achievementof STBC and BF gains Figure 6 shows the simulated BLERcurves including Case 1 of TPULA only (without STBC andBF) Case 2 of TPULA with STBC Case 3 of TPULA withSTBC and TxBF and Case 4 of TPULA with STBC andTRxBF where the target BLER for performance comparisonis fixed at 10minus2 According to Figure 6 Case 1 has the worstperformance because there is no space-time coding and BFgain achieved By applying the receiver studied by Liu et al [7]for interference cancellation almost 1 dB of signal-to-noiseratio (SNR) gain at the target BLER can be achievedwithCase2 under the indoor environment compared with Case 1 Inaddition compared with Case 2 about 23 dB of SNR gainunder the indoor environment at the target BLER is obtainedwith Case 3 when the TxBF is applied Compared with Case3 05 dB SNR gain at the target BLER is achieved by usingthe MMSE RxBF via Case 4 because the amount of error fedinto the STBC decoding process is further decreased Theerror floors occur under the outdoor environment due tohigh mobility at 100 kmh This indicates that the numberof pilots in the TPULA system is not enough to compensatethe channel correctly under an environment with fast time-varying phase response Again the efficiency of our proposedscheme can be quantified in terms of throughput Figure 7illustrates the simulated throughput curves By focusing onthe curves obtained under the indoor environment we canobserve that Case 4 of TPULA with STBC and TRxBF canachieve the highest throughput of 21Mbps at 119864

1198871198730of 0 dB

and 20Mbps 19Mbps and 17Mbps for Case 3 Case 2 andCase 1 at 119864

1198871198730of 0 dB respectively

In addition we evaluate the average BLER performancewhen two mobile users are employed In this case the eightAEs at Tx are divided into two groups where each group iswith four AEs and performs TxBF for a mobile user Figure 8illustrates the simulation results of average BLER for twousers with the transmission Case 4 According to Figure 8we can observe that the performance decreases a lot dueto the loss of diversity order at Tx because multiusers need

6 International Journal of Antennas and Propagation

30

210

60

240

90

270

120

300

150

330

180 0

Z

X

120573

120572

Figure 4 Setting of zero for TxBF weights

Subframe20 1 3 4 5 6 7 8 9 10 11 12 13

0

1

11

Reso

urce

blo

ck

10

9

8

7

6

5

4

3

2

x98x4xlowast1

x102x8xlowast5

x106x12xlowast9

x110x16xlowast13

x114x20xlowast17

x118x24xlowast21

x122x28xlowast25

x126x32xlowast29

x130x36xlowast33

x134x40xlowast37

x138x44xlowast41

x142

x50

x54

x58

x62

x66

x70

x74

x78

x82

x86

x90

x94x48xlowast45

xlowast3

xlowast7

xlowast11

xlowast15

xlowast19

xlowast23

xlowast27

xlowast31

xlowast35

xlowast39

xlowast43

xlowast47

xlowast49

xlowast53

xlowast57

xlowast61

xlowast65

xlowast69

xlowast73

xlowast77

xlowast81

xlowast85

xlowast89

xlowast93

x52

x56

x60

x64

x68

x72

x76

x80

x84

x88

x92

x96

xlowast51

xlowast55

xlowast59

xlowast63

xlowast67

xlowast71

xlowast75

xlowast79

xlowast83

xlowast87

xlowast91

xlowast95

xlowast97

xlowast101

xlowast105

xlowast109

xlowast113

xlowast117

xlowast121

xlowast125

xlowast129

xlowast133

xlowast137

xlowast141

x100

x104

x108

x112

x116

x120

x124

x128

x132

x136

x140

x144

xlowast99

xlowast103

xlowast107

xlowast111

xlowast115

xlowast119

xlowast123

xlowast127

xlowast131

xlowast135

xlowast139

xlowast143

PA1B1

PA2B1

PA3B1

PA4B1

PA5B1

PA6B1

PA7B1

PA8B1

x2

x6

x10

x14

x18

x22

x26

x30

x34

x38

x42

x46

(a) Data mapping at TxApB1

20 1 3 4 5 6 7 8 9 10 11 12 13

0

1

11

Reso

urce

blo

ck

10

9

8

7

6

5

4

3

2

Subframe

x1

x5

x9

x13

x17

x21

x25

x29

x33

x37

x41

x45

x3

x7

x11

x15

x19

x23

x27

x31

x35

x39

x43

x47

x49

x53

x57

x61

x65

x69

x73

x77

x81

x85

x89

x93

x51

x55

x59

x63

x67

x71

x75

x79

x83

x87

x91

x95

x97

x101

x105

x109

x113

x117

x121

x125

x129

x133

x137

x141

x99

x103

x107

x111

x115

x119

x123

x127

x131

x135

x139

x143

minusxlowast100

minusxlowast104

minusxlowast108

minusxlowast112

minusxlowast116

minusxlowast120

minusxlowast124

minusxlowast128

minusxlowast132

minusxlowast136

minusxlowast140

minusxlowast144

minusxlowast98

minusxlowast102

minusxlowast106

minusxlowast110

minusxlowast114

minusxlowast118

minusxlowast122

minusxlowast126

minusxlowast130

minusxlowast134

minusxlowast138

minusxlowast142

minusxlowast50

minusxlowast54

minusxlowast58

minusxlowast62

minusxlowast66

minusxlowast70

minusxlowast74

minusxlowast78

minusxlowast82

minusxlowast86

minusxlowast90

minusxlowast94

minusxlowast52

minusxlowast56

minusxlowast52

minusxlowast56

minusxlowast52

minusxlowast56

minusxlowast52

minusxlowast56

minusxlowast56

minusxlowast52

minusxlowast65

minusxlowast96

minusxlowast2

minusxlowast6

minusxlowast10

minusxlowast14

minusxlowast18

minusxlowast22

minusxlowast26

minusxlowast30

minusxlowast34

minusxlowast38

minusxlowast42

minusxlowast46

minusxlowast4

minusxlowast8

minusxlowast12

minusxlowast16

minusxlowast20

minusxlowast24

minusxlowast28

minusxlowast32

minusxlowast36

minusxlowast40

minusxlowast44

minusxlowast48

PA1B2

PA2B2

PA3B2

PA4B2

PA5B2

PA6B2

PA7B2

PA8B2

(b) Data mapping at TxApB2

0

1

11

Reso

urce

blo

ck

10

9

8

7

6

5

4

3

2

20 1 3 4 5 6 7 8 9 10 11 12 13Subframe

x98

x102

x106

x110

x114

x118

x122

x126

x130

x134

x138

x142

x50

x54

x58

x62

x66

x70

x74

x78

x82

x86

x90

x94

xlowast1

xlowast5

xlowast9

xlowast13

xlowast17

xlowast21

xlowast25

xlowast29

xlowast33

xlowast37

xlowast41

xlowast45

xlowast3

xlowast7

xlowast11

xlowast15

xlowast19

xlowast23

xlowast27

xlowast31

xlowast35

xlowast39

xlowast43

xlowast47

xlowast49

xlowast53

xlowast57

xlowast61

xlowast65

xlowast69

xlowast73

xlowast77

xlowast81

xlowast85

xlowast89

xlowast93

x52

x56

x60

x64

x68

x72

x76

x80

x84

x88

x92

x96

xlowast51

xlowast55

xlowast59

xlowast63

xlowast67

xlowast71

xlowast75

xlowast79

xlowast83

xlowast87

xlowast91

xlowast95

xlowast97

xlowast101

xlowast105

xlowast109

xlowast113

xlowast117

xlowast121

xlowast125

xlowast129

xlowast133

xlowast137

xlowast141

x100

x104

x108

x112

x116

x120

x124

x128

x132

x136

x140

x144

xlowast99

xlowast103

xlowast107

xlowast111

xlowast115

xlowast119

xlowast123

xlowast127

xlowast131

xlowast135

xlowast139

xlowast143

x2

x6

x10

x14

x18

x22

x26

x30

x34

x38

x42

x46

x4

x8

x12

x16

x20

x24

x28

x32

x36

x40

x44

x48

PA1B3

PA2B3

PA3B3

PA4B3

PA5B3

PA6B3

PA7B3

PA8B3

(c) Data mapping at TxApB3

Figure 5 Downlink data mapping of the LTE-A resource blocks for the TPULA system

International Journal of Antennas and Propagation 7

Table 1 Simulation parameters

System parameter ValueCarrier frequency 18GHzSystem bandwidth 20MHzFFT size 2048Number of data carriers 1200Number of samples in CP 144Subcarrier spacing 15 KHzModulation QPSKChannel coding NoMIMO configuration 8A lowast 3B(Tx) by 2A lowast 3B(Rx)Array spacing Half-wavelengthChannel parameter Value

Number of scatterersIndoor 64

Outdoor 4 clusters with 16scatterers per cluster (64 in total)

Scatterer radius 10m

Rx velocity 3 kmh for indoor100 kmh for outdoor

Fading FlatXPD 58 dBPractical channel estimation Zero-forcing

to separate the total number of AEs and only four AEs arededicated to a mobile user For example there is about 5 dBSNR loss at target BLER in case of employing two userscompared with the single user case Meanwhile Figure 9demonstrates the simulated throughput curves for two userswith the transmission Case 4 where system throughput isincreased a lot when two users are employed From Figure 9we can observe that the single user cases have a higherthroughput than the multiusers cases with a low SNR forexample minus10 dB mainly due to a lower BLER By increasingthe SNR the performance of multiusers gets better than thesingle user casesThis is because the proposed scheme is goodat keeping the system BLER performance When multiusersare employed the system throughput then can be guaranteedto be linearly enhanced with the number of users beingincreased

With the consideration of tradeoff regarding the spec-trum efficiency the blind channel estimation (BCE) [10]which requires no or a minimal number of pilots might bebetter employed in the TPULA system in our future researchBy using BCE not only the system spectrum efficiency can beincreased but also the error floor caused by the high mobilitymight be solved

5 Conclusions and Future Works

This paper proposes a TPULA system associated with STBCand BF techniques to exploit diversity gain under a trans-mission scenario with both low and high correlated linksThe polarized antenna branches are grouped into three setswhich steer three orthogonal beams at theX-Y Z-X and Y-Z

minus10 minus5 0 5 10 15 20 25

BLER

TPULA only outdoorTPULA w STBC outdoorTPULA w STBC and TxBF outdoorTPULA w STBC and TRxBF outdoorTPULA only indoorTPULA w STBC indoorTPULA w STBC and TxBF indoorTPULA w STBC and TRxBF indoor

TargetBLER

100

10minus1

10minus2

EbN0 (dB)

Figure 6 Simulation results of BLER for single user

0minus10 minus5 0 5 10 15

05

1

15

2

25

3

Thro

ughp

ut (b

ps)

times107

EbN0 (dB)

TPULA w STBC and TRxBF indoorTPULA w STBC and TxBF indoorTPULA w STBC indoorTPULA only indoorTPULA w STBC and TRxBF outdoorTPULA w STBC and TxBF outdoorTPULA w STBC outdoorTPULA only outdoor

Figure 7 Simulation results of throughput for single user

8 International Journal of Antennas and Propagation

minus10 minus5 0 5 10 15 20 25

BLER

TPULA w STBC and TRxBF outdoor two usersTPULA w STBC and TRxBF outdoor single userTPULA w STBC and TRxBF indoor two usersTPULA w STBC and TRxBF indoor single user

100

10minus1

10minus2

EbN0 (dB)

Figure 8 Simulation results of average BLER for two users

minus10 minus5 0 5 10 150

1

2

3

4

5

6

Thro

ughp

ut (b

ps)

TPULA w STBC and TRxBF indoor two usersTPULA w STBC and TRxBF outdoor two usersTPULA w STBC and TRxBF indoor single userTPULA w STBC and TRxBF outdoor single user

times107

EbN0 (dB)

Figure 9 Simulation results of throughput for two users

planes STBC is additionally applied at three orthogonallycolocated branches of each AE to accommodate the space-time diversity gain From the simulation results the proposedscheme in Case 4 (TPULA with SBTC and TRxBF) proveswe can obtain about 4 dB SNR gain at the target BLERcompared with Case 1 (TPULA without STBC and TRxBF)

under the indoor environment Although the performancein an outdoor environment is also increased by the proposedscheme the error-floors occur due to the high Rxmobility Inthis paper we have verified the proposed scheme under theflat fading environment only As a future work the selectivefading environment will be examined and the AE selectionscheme for OFDM-based TPULA system might be proposedto overcome the ISI caused by multipath

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgment

This research was supported by the MSIP (Ministry ofScience ICT and Future Planning) Korea under the ITRC(InformationTechnologyResearchCenter) Support Program(IITP-2015-H8501-15-1019) supervised by the IITP (Institutefor Information amp Communications Technology Promo-tion)

References

[1] M-T Dao V-A Nguyen Y-T Im S-O Park and G Yoonldquo3D polarized channel modeling and performance comparisonof MIMO antenna configurations with different polarizationsrdquoIEEE Transactions on Antennas and Propagation vol 59 no 7pp 2672ndash2682 2011

[2] A S Y Poon and D N C Tse ldquoDegree-of-freedom gain fromusing polarimetric antenna elementsrdquo Institute of Electrical andElectronics Engineers Transactions on Information Theory vol57 no 9 pp 5695ndash5709 2011

[3] X Su and K H Chang ldquoPolarized uniform linear arraysystem beam radiation pattern beamforming diversity orderand channel capacityrdquo International Journal of Antennas andPropagation vol 2015 Article ID 371236 9 pages 2015

[4] J Li U Park and S Y Kim ldquoAn efficient rate one STBCscheme with 3 transmit antennasrdquo in Proceedings of the Interna-tional Conference onWireless Communications Networking andMobile Computing (WiCOM rsquo08) pp 1ndash4 IEEE Dalian ChinaOctober 2008

[5] U Park S Kim K Lim and J Li ldquoA novel QO-STBC schemewith linear decoding for three and four transmit antennasrdquoIEEE Communications Letters vol 12 no 12 pp 868ndash870 2008

[6] V Madisetti Digital Signal Processing Handbook CRC Press2nd edition 1999

[7] L J Liu S Y Kim andM-S Lim ldquoAn efficient selective receiverfor STBC schemerdquo in Proceedings of the IEEE InternationalConference on Communications (ICC rsquo07) pp 4196ndash4200 IEEEGlasgow Scotland June 2007

[8] W Liu ldquoAdaptive wideband beamforming with sensor delay-linesrdquo Signal Processing vol 89 no 5 pp 876ndash882 2009

[9] X Su B Hui and K H Chang ldquo3-D MIMO channel mod-eling with beamforming analysis for dual-polarized antenna

International Journal of Antennas and Propagation 9

systemsrdquo in Proceedings of the IEEE 78th Vehicular TechnologyConference (VTC Fall rsquo13) pp 1ndash5 September 2013

[10] E Beres and R Adve ldquoBlind channel estimation for orthogonalSTBC in MISO systemsrdquo IEEE Transactions on Vehicular Tech-nology vol 56 no 4 pp 2042ndash2050 2007

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of

6 International Journal of Antennas and Propagation

30

210

60

240

90

270

120

300

150

330

180 0

Z

X

120573

120572

Figure 4 Setting of zero for TxBF weights

Subframe20 1 3 4 5 6 7 8 9 10 11 12 13

0

1

11

Reso

urce

blo

ck

10

9

8

7

6

5

4

3

2

x98x4xlowast1

x102x8xlowast5

x106x12xlowast9

x110x16xlowast13

x114x20xlowast17

x118x24xlowast21

x122x28xlowast25

x126x32xlowast29

x130x36xlowast33

x134x40xlowast37

x138x44xlowast41

x142

x50

x54

x58

x62

x66

x70

x74

x78

x82

x86

x90

x94x48xlowast45

xlowast3

xlowast7

xlowast11

xlowast15

xlowast19

xlowast23

xlowast27

xlowast31

xlowast35

xlowast39

xlowast43

xlowast47

xlowast49

xlowast53

xlowast57

xlowast61

xlowast65

xlowast69

xlowast73

xlowast77

xlowast81

xlowast85

xlowast89

xlowast93

x52

x56

x60

x64

x68

x72

x76

x80

x84

x88

x92

x96

xlowast51

xlowast55

xlowast59

xlowast63

xlowast67

xlowast71

xlowast75

xlowast79

xlowast83

xlowast87

xlowast91

xlowast95

xlowast97

xlowast101

xlowast105

xlowast109

xlowast113

xlowast117

xlowast121

xlowast125

xlowast129

xlowast133

xlowast137

xlowast141

x100

x104

x108

x112

x116

x120

x124

x128

x132

x136

x140

x144

xlowast99

xlowast103

xlowast107

xlowast111

xlowast115

xlowast119

xlowast123

xlowast127

xlowast131

xlowast135

xlowast139

xlowast143

PA1B1

PA2B1

PA3B1

PA4B1

PA5B1

PA6B1

PA7B1

PA8B1

x2

x6

x10

x14

x18

x22

x26

x30

x34

x38

x42

x46

(a) Data mapping at TxApB1

20 1 3 4 5 6 7 8 9 10 11 12 13

0

1

11

Reso

urce

blo

ck

10

9

8

7

6

5

4

3

2

Subframe

x1

x5

x9

x13

x17

x21

x25

x29

x33

x37

x41

x45

x3

x7

x11

x15

x19

x23

x27

x31

x35

x39

x43

x47

x49

x53

x57

x61

x65

x69

x73

x77

x81

x85

x89

x93

x51

x55

x59

x63

x67

x71

x75

x79

x83

x87

x91

x95

x97

x101

x105

x109

x113

x117

x121

x125

x129

x133

x137

x141

x99

x103

x107

x111

x115

x119

x123

x127

x131

x135

x139

x143

minusxlowast100

minusxlowast104

minusxlowast108

minusxlowast112

minusxlowast116

minusxlowast120

minusxlowast124

minusxlowast128

minusxlowast132

minusxlowast136

minusxlowast140

minusxlowast144

minusxlowast98

minusxlowast102

minusxlowast106

minusxlowast110

minusxlowast114

minusxlowast118

minusxlowast122

minusxlowast126

minusxlowast130

minusxlowast134

minusxlowast138

minusxlowast142

minusxlowast50

minusxlowast54

minusxlowast58

minusxlowast62

minusxlowast66

minusxlowast70

minusxlowast74

minusxlowast78

minusxlowast82

minusxlowast86

minusxlowast90

minusxlowast94

minusxlowast52

minusxlowast56

minusxlowast52

minusxlowast56

minusxlowast52

minusxlowast56

minusxlowast52

minusxlowast56

minusxlowast56

minusxlowast52

minusxlowast65

minusxlowast96

minusxlowast2

minusxlowast6

minusxlowast10

minusxlowast14

minusxlowast18

minusxlowast22

minusxlowast26

minusxlowast30

minusxlowast34

minusxlowast38

minusxlowast42

minusxlowast46

minusxlowast4

minusxlowast8

minusxlowast12

minusxlowast16

minusxlowast20

minusxlowast24

minusxlowast28

minusxlowast32

minusxlowast36

minusxlowast40

minusxlowast44

minusxlowast48

PA1B2

PA2B2

PA3B2

PA4B2

PA5B2

PA6B2

PA7B2

PA8B2

(b) Data mapping at TxApB2

0

1

11

Reso

urce

blo

ck

10

9

8

7

6

5

4

3

2

20 1 3 4 5 6 7 8 9 10 11 12 13Subframe

x98

x102

x106

x110

x114

x118

x122

x126

x130

x134

x138

x142

x50

x54

x58

x62

x66

x70

x74

x78

x82

x86

x90

x94

xlowast1

xlowast5

xlowast9

xlowast13

xlowast17

xlowast21

xlowast25

xlowast29

xlowast33

xlowast37

xlowast41

xlowast45

xlowast3

xlowast7

xlowast11

xlowast15

xlowast19

xlowast23

xlowast27

xlowast31

xlowast35

xlowast39

xlowast43

xlowast47

xlowast49

xlowast53

xlowast57

xlowast61

xlowast65

xlowast69

xlowast73

xlowast77

xlowast81

xlowast85

xlowast89

xlowast93

x52

x56

x60

x64

x68

x72

x76

x80

x84

x88

x92

x96

xlowast51

xlowast55

xlowast59

xlowast63

xlowast67

xlowast71

xlowast75

xlowast79

xlowast83

xlowast87

xlowast91

xlowast95

xlowast97

xlowast101

xlowast105

xlowast109

xlowast113

xlowast117

xlowast121

xlowast125

xlowast129

xlowast133

xlowast137

xlowast141

x100

x104

x108

x112

x116

x120

x124

x128

x132

x136

x140

x144

xlowast99

xlowast103

xlowast107

xlowast111

xlowast115

xlowast119

xlowast123

xlowast127

xlowast131

xlowast135

xlowast139

xlowast143

x2

x6

x10

x14

x18

x22

x26

x30

x34

x38

x42

x46

x4

x8

x12

x16

x20

x24

x28

x32

x36

x40

x44

x48

PA1B3

PA2B3

PA3B3

PA4B3

PA5B3

PA6B3

PA7B3

PA8B3

(c) Data mapping at TxApB3

Figure 5 Downlink data mapping of the LTE-A resource blocks for the TPULA system

International Journal of Antennas and Propagation 7

Table 1 Simulation parameters

System parameter ValueCarrier frequency 18GHzSystem bandwidth 20MHzFFT size 2048Number of data carriers 1200Number of samples in CP 144Subcarrier spacing 15 KHzModulation QPSKChannel coding NoMIMO configuration 8A lowast 3B(Tx) by 2A lowast 3B(Rx)Array spacing Half-wavelengthChannel parameter Value

Number of scatterersIndoor 64

Outdoor 4 clusters with 16scatterers per cluster (64 in total)

Scatterer radius 10m

Rx velocity 3 kmh for indoor100 kmh for outdoor

Fading FlatXPD 58 dBPractical channel estimation Zero-forcing

to separate the total number of AEs and only four AEs arededicated to a mobile user For example there is about 5 dBSNR loss at target BLER in case of employing two userscompared with the single user case Meanwhile Figure 9demonstrates the simulated throughput curves for two userswith the transmission Case 4 where system throughput isincreased a lot when two users are employed From Figure 9we can observe that the single user cases have a higherthroughput than the multiusers cases with a low SNR forexample minus10 dB mainly due to a lower BLER By increasingthe SNR the performance of multiusers gets better than thesingle user casesThis is because the proposed scheme is goodat keeping the system BLER performance When multiusersare employed the system throughput then can be guaranteedto be linearly enhanced with the number of users beingincreased

With the consideration of tradeoff regarding the spec-trum efficiency the blind channel estimation (BCE) [10]which requires no or a minimal number of pilots might bebetter employed in the TPULA system in our future researchBy using BCE not only the system spectrum efficiency can beincreased but also the error floor caused by the high mobilitymight be solved

5 Conclusions and Future Works

This paper proposes a TPULA system associated with STBCand BF techniques to exploit diversity gain under a trans-mission scenario with both low and high correlated linksThe polarized antenna branches are grouped into three setswhich steer three orthogonal beams at theX-Y Z-X and Y-Z

minus10 minus5 0 5 10 15 20 25

BLER

TPULA only outdoorTPULA w STBC outdoorTPULA w STBC and TxBF outdoorTPULA w STBC and TRxBF outdoorTPULA only indoorTPULA w STBC indoorTPULA w STBC and TxBF indoorTPULA w STBC and TRxBF indoor

TargetBLER

100

10minus1

10minus2

EbN0 (dB)

Figure 6 Simulation results of BLER for single user

0minus10 minus5 0 5 10 15

05

1

15

2

25

3

Thro

ughp

ut (b

ps)

times107

EbN0 (dB)

TPULA w STBC and TRxBF indoorTPULA w STBC and TxBF indoorTPULA w STBC indoorTPULA only indoorTPULA w STBC and TRxBF outdoorTPULA w STBC and TxBF outdoorTPULA w STBC outdoorTPULA only outdoor

Figure 7 Simulation results of throughput for single user

8 International Journal of Antennas and Propagation

minus10 minus5 0 5 10 15 20 25

BLER

TPULA w STBC and TRxBF outdoor two usersTPULA w STBC and TRxBF outdoor single userTPULA w STBC and TRxBF indoor two usersTPULA w STBC and TRxBF indoor single user

100

10minus1

10minus2

EbN0 (dB)

Figure 8 Simulation results of average BLER for two users

minus10 minus5 0 5 10 150

1

2

3

4

5

6

Thro

ughp

ut (b

ps)

TPULA w STBC and TRxBF indoor two usersTPULA w STBC and TRxBF outdoor two usersTPULA w STBC and TRxBF indoor single userTPULA w STBC and TRxBF outdoor single user

times107

EbN0 (dB)

Figure 9 Simulation results of throughput for two users

planes STBC is additionally applied at three orthogonallycolocated branches of each AE to accommodate the space-time diversity gain From the simulation results the proposedscheme in Case 4 (TPULA with SBTC and TRxBF) proveswe can obtain about 4 dB SNR gain at the target BLERcompared with Case 1 (TPULA without STBC and TRxBF)

under the indoor environment Although the performancein an outdoor environment is also increased by the proposedscheme the error-floors occur due to the high Rxmobility Inthis paper we have verified the proposed scheme under theflat fading environment only As a future work the selectivefading environment will be examined and the AE selectionscheme for OFDM-based TPULA system might be proposedto overcome the ISI caused by multipath

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgment

This research was supported by the MSIP (Ministry ofScience ICT and Future Planning) Korea under the ITRC(InformationTechnologyResearchCenter) Support Program(IITP-2015-H8501-15-1019) supervised by the IITP (Institutefor Information amp Communications Technology Promo-tion)

References

[1] M-T Dao V-A Nguyen Y-T Im S-O Park and G Yoonldquo3D polarized channel modeling and performance comparisonof MIMO antenna configurations with different polarizationsrdquoIEEE Transactions on Antennas and Propagation vol 59 no 7pp 2672ndash2682 2011

[2] A S Y Poon and D N C Tse ldquoDegree-of-freedom gain fromusing polarimetric antenna elementsrdquo Institute of Electrical andElectronics Engineers Transactions on Information Theory vol57 no 9 pp 5695ndash5709 2011

[3] X Su and K H Chang ldquoPolarized uniform linear arraysystem beam radiation pattern beamforming diversity orderand channel capacityrdquo International Journal of Antennas andPropagation vol 2015 Article ID 371236 9 pages 2015

[4] J Li U Park and S Y Kim ldquoAn efficient rate one STBCscheme with 3 transmit antennasrdquo in Proceedings of the Interna-tional Conference onWireless Communications Networking andMobile Computing (WiCOM rsquo08) pp 1ndash4 IEEE Dalian ChinaOctober 2008

[5] U Park S Kim K Lim and J Li ldquoA novel QO-STBC schemewith linear decoding for three and four transmit antennasrdquoIEEE Communications Letters vol 12 no 12 pp 868ndash870 2008

[6] V Madisetti Digital Signal Processing Handbook CRC Press2nd edition 1999

[7] L J Liu S Y Kim andM-S Lim ldquoAn efficient selective receiverfor STBC schemerdquo in Proceedings of the IEEE InternationalConference on Communications (ICC rsquo07) pp 4196ndash4200 IEEEGlasgow Scotland June 2007

[8] W Liu ldquoAdaptive wideband beamforming with sensor delay-linesrdquo Signal Processing vol 89 no 5 pp 876ndash882 2009

[9] X Su B Hui and K H Chang ldquo3-D MIMO channel mod-eling with beamforming analysis for dual-polarized antenna

International Journal of Antennas and Propagation 9

systemsrdquo in Proceedings of the IEEE 78th Vehicular TechnologyConference (VTC Fall rsquo13) pp 1ndash5 September 2013

[10] E Beres and R Adve ldquoBlind channel estimation for orthogonalSTBC in MISO systemsrdquo IEEE Transactions on Vehicular Tech-nology vol 56 no 4 pp 2042ndash2050 2007

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of

International Journal of Antennas and Propagation 7

Table 1 Simulation parameters

System parameter ValueCarrier frequency 18GHzSystem bandwidth 20MHzFFT size 2048Number of data carriers 1200Number of samples in CP 144Subcarrier spacing 15 KHzModulation QPSKChannel coding NoMIMO configuration 8A lowast 3B(Tx) by 2A lowast 3B(Rx)Array spacing Half-wavelengthChannel parameter Value

Number of scatterersIndoor 64

Outdoor 4 clusters with 16scatterers per cluster (64 in total)

Scatterer radius 10m

Rx velocity 3 kmh for indoor100 kmh for outdoor

Fading FlatXPD 58 dBPractical channel estimation Zero-forcing

to separate the total number of AEs and only four AEs arededicated to a mobile user For example there is about 5 dBSNR loss at target BLER in case of employing two userscompared with the single user case Meanwhile Figure 9demonstrates the simulated throughput curves for two userswith the transmission Case 4 where system throughput isincreased a lot when two users are employed From Figure 9we can observe that the single user cases have a higherthroughput than the multiusers cases with a low SNR forexample minus10 dB mainly due to a lower BLER By increasingthe SNR the performance of multiusers gets better than thesingle user casesThis is because the proposed scheme is goodat keeping the system BLER performance When multiusersare employed the system throughput then can be guaranteedto be linearly enhanced with the number of users beingincreased

With the consideration of tradeoff regarding the spec-trum efficiency the blind channel estimation (BCE) [10]which requires no or a minimal number of pilots might bebetter employed in the TPULA system in our future researchBy using BCE not only the system spectrum efficiency can beincreased but also the error floor caused by the high mobilitymight be solved

5 Conclusions and Future Works

This paper proposes a TPULA system associated with STBCand BF techniques to exploit diversity gain under a trans-mission scenario with both low and high correlated linksThe polarized antenna branches are grouped into three setswhich steer three orthogonal beams at theX-Y Z-X and Y-Z

minus10 minus5 0 5 10 15 20 25

BLER

TPULA only outdoorTPULA w STBC outdoorTPULA w STBC and TxBF outdoorTPULA w STBC and TRxBF outdoorTPULA only indoorTPULA w STBC indoorTPULA w STBC and TxBF indoorTPULA w STBC and TRxBF indoor

TargetBLER

100

10minus1

10minus2

EbN0 (dB)

Figure 6 Simulation results of BLER for single user

0minus10 minus5 0 5 10 15

05

1

15

2

25

3

Thro

ughp

ut (b

ps)

times107

EbN0 (dB)

TPULA w STBC and TRxBF indoorTPULA w STBC and TxBF indoorTPULA w STBC indoorTPULA only indoorTPULA w STBC and TRxBF outdoorTPULA w STBC and TxBF outdoorTPULA w STBC outdoorTPULA only outdoor

Figure 7 Simulation results of throughput for single user

8 International Journal of Antennas and Propagation

minus10 minus5 0 5 10 15 20 25

BLER

TPULA w STBC and TRxBF outdoor two usersTPULA w STBC and TRxBF outdoor single userTPULA w STBC and TRxBF indoor two usersTPULA w STBC and TRxBF indoor single user

100

10minus1

10minus2

EbN0 (dB)

Figure 8 Simulation results of average BLER for two users

minus10 minus5 0 5 10 150

1

2

3

4

5

6

Thro

ughp

ut (b

ps)

TPULA w STBC and TRxBF indoor two usersTPULA w STBC and TRxBF outdoor two usersTPULA w STBC and TRxBF indoor single userTPULA w STBC and TRxBF outdoor single user

times107

EbN0 (dB)

Figure 9 Simulation results of throughput for two users

planes STBC is additionally applied at three orthogonallycolocated branches of each AE to accommodate the space-time diversity gain From the simulation results the proposedscheme in Case 4 (TPULA with SBTC and TRxBF) proveswe can obtain about 4 dB SNR gain at the target BLERcompared with Case 1 (TPULA without STBC and TRxBF)

under the indoor environment Although the performancein an outdoor environment is also increased by the proposedscheme the error-floors occur due to the high Rxmobility Inthis paper we have verified the proposed scheme under theflat fading environment only As a future work the selectivefading environment will be examined and the AE selectionscheme for OFDM-based TPULA system might be proposedto overcome the ISI caused by multipath

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgment

This research was supported by the MSIP (Ministry ofScience ICT and Future Planning) Korea under the ITRC(InformationTechnologyResearchCenter) Support Program(IITP-2015-H8501-15-1019) supervised by the IITP (Institutefor Information amp Communications Technology Promo-tion)

References

[1] M-T Dao V-A Nguyen Y-T Im S-O Park and G Yoonldquo3D polarized channel modeling and performance comparisonof MIMO antenna configurations with different polarizationsrdquoIEEE Transactions on Antennas and Propagation vol 59 no 7pp 2672ndash2682 2011

[2] A S Y Poon and D N C Tse ldquoDegree-of-freedom gain fromusing polarimetric antenna elementsrdquo Institute of Electrical andElectronics Engineers Transactions on Information Theory vol57 no 9 pp 5695ndash5709 2011

[3] X Su and K H Chang ldquoPolarized uniform linear arraysystem beam radiation pattern beamforming diversity orderand channel capacityrdquo International Journal of Antennas andPropagation vol 2015 Article ID 371236 9 pages 2015

[4] J Li U Park and S Y Kim ldquoAn efficient rate one STBCscheme with 3 transmit antennasrdquo in Proceedings of the Interna-tional Conference onWireless Communications Networking andMobile Computing (WiCOM rsquo08) pp 1ndash4 IEEE Dalian ChinaOctober 2008

[5] U Park S Kim K Lim and J Li ldquoA novel QO-STBC schemewith linear decoding for three and four transmit antennasrdquoIEEE Communications Letters vol 12 no 12 pp 868ndash870 2008

[6] V Madisetti Digital Signal Processing Handbook CRC Press2nd edition 1999

[7] L J Liu S Y Kim andM-S Lim ldquoAn efficient selective receiverfor STBC schemerdquo in Proceedings of the IEEE InternationalConference on Communications (ICC rsquo07) pp 4196ndash4200 IEEEGlasgow Scotland June 2007

[8] W Liu ldquoAdaptive wideband beamforming with sensor delay-linesrdquo Signal Processing vol 89 no 5 pp 876ndash882 2009

[9] X Su B Hui and K H Chang ldquo3-D MIMO channel mod-eling with beamforming analysis for dual-polarized antenna

International Journal of Antennas and Propagation 9

systemsrdquo in Proceedings of the IEEE 78th Vehicular TechnologyConference (VTC Fall rsquo13) pp 1ndash5 September 2013

[10] E Beres and R Adve ldquoBlind channel estimation for orthogonalSTBC in MISO systemsrdquo IEEE Transactions on Vehicular Tech-nology vol 56 no 4 pp 2042ndash2050 2007

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of

8 International Journal of Antennas and Propagation

minus10 minus5 0 5 10 15 20 25

BLER

TPULA w STBC and TRxBF outdoor two usersTPULA w STBC and TRxBF outdoor single userTPULA w STBC and TRxBF indoor two usersTPULA w STBC and TRxBF indoor single user

100

10minus1

10minus2

EbN0 (dB)

Figure 8 Simulation results of average BLER for two users

minus10 minus5 0 5 10 150

1

2

3

4

5

6

Thro

ughp

ut (b

ps)

TPULA w STBC and TRxBF indoor two usersTPULA w STBC and TRxBF outdoor two usersTPULA w STBC and TRxBF indoor single userTPULA w STBC and TRxBF outdoor single user

times107

EbN0 (dB)

Figure 9 Simulation results of throughput for two users

planes STBC is additionally applied at three orthogonallycolocated branches of each AE to accommodate the space-time diversity gain From the simulation results the proposedscheme in Case 4 (TPULA with SBTC and TRxBF) proveswe can obtain about 4 dB SNR gain at the target BLERcompared with Case 1 (TPULA without STBC and TRxBF)

under the indoor environment Although the performancein an outdoor environment is also increased by the proposedscheme the error-floors occur due to the high Rxmobility Inthis paper we have verified the proposed scheme under theflat fading environment only As a future work the selectivefading environment will be examined and the AE selectionscheme for OFDM-based TPULA system might be proposedto overcome the ISI caused by multipath

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgment

This research was supported by the MSIP (Ministry ofScience ICT and Future Planning) Korea under the ITRC(InformationTechnologyResearchCenter) Support Program(IITP-2015-H8501-15-1019) supervised by the IITP (Institutefor Information amp Communications Technology Promo-tion)

References

[1] M-T Dao V-A Nguyen Y-T Im S-O Park and G Yoonldquo3D polarized channel modeling and performance comparisonof MIMO antenna configurations with different polarizationsrdquoIEEE Transactions on Antennas and Propagation vol 59 no 7pp 2672ndash2682 2011

[2] A S Y Poon and D N C Tse ldquoDegree-of-freedom gain fromusing polarimetric antenna elementsrdquo Institute of Electrical andElectronics Engineers Transactions on Information Theory vol57 no 9 pp 5695ndash5709 2011

[3] X Su and K H Chang ldquoPolarized uniform linear arraysystem beam radiation pattern beamforming diversity orderand channel capacityrdquo International Journal of Antennas andPropagation vol 2015 Article ID 371236 9 pages 2015

[4] J Li U Park and S Y Kim ldquoAn efficient rate one STBCscheme with 3 transmit antennasrdquo in Proceedings of the Interna-tional Conference onWireless Communications Networking andMobile Computing (WiCOM rsquo08) pp 1ndash4 IEEE Dalian ChinaOctober 2008

[5] U Park S Kim K Lim and J Li ldquoA novel QO-STBC schemewith linear decoding for three and four transmit antennasrdquoIEEE Communications Letters vol 12 no 12 pp 868ndash870 2008

[6] V Madisetti Digital Signal Processing Handbook CRC Press2nd edition 1999

[7] L J Liu S Y Kim andM-S Lim ldquoAn efficient selective receiverfor STBC schemerdquo in Proceedings of the IEEE InternationalConference on Communications (ICC rsquo07) pp 4196ndash4200 IEEEGlasgow Scotland June 2007

[8] W Liu ldquoAdaptive wideband beamforming with sensor delay-linesrdquo Signal Processing vol 89 no 5 pp 876ndash882 2009

[9] X Su B Hui and K H Chang ldquo3-D MIMO channel mod-eling with beamforming analysis for dual-polarized antenna

International Journal of Antennas and Propagation 9

systemsrdquo in Proceedings of the IEEE 78th Vehicular TechnologyConference (VTC Fall rsquo13) pp 1ndash5 September 2013

[10] E Beres and R Adve ldquoBlind channel estimation for orthogonalSTBC in MISO systemsrdquo IEEE Transactions on Vehicular Tech-nology vol 56 no 4 pp 2042ndash2050 2007

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of

International Journal of Antennas and Propagation 9

systemsrdquo in Proceedings of the IEEE 78th Vehicular TechnologyConference (VTC Fall rsquo13) pp 1ndash5 September 2013

[10] E Beres and R Adve ldquoBlind channel estimation for orthogonalSTBC in MISO systemsrdquo IEEE Transactions on Vehicular Tech-nology vol 56 no 4 pp 2042ndash2050 2007

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of