Research ArticleCapacity Improvement for DVB-NGH with Dual-Polarized MIMOSpatial Multiplexing and Hybrid Beamforming

Huu Trung Nguyen ,1 Trung Tan Le,1 and Trung Hien Nguyen2

1School of Electronics and Telecommunications, Hanoi University of Science and Technology, Hanoi 10000, Vietnam2Department of Electronics and Telecommunications, Posts and Telecommunications Institute of Technology, Hanoi 10000, Vietnam

Correspondence should be addressed to Huu Trung Nguyen; trung.nguyenhuu@hust.edu.vn

Received 2 September 2019; Revised 13 July 2020; Accepted 27 July 2020; Published 17 August 2020

Academic Editor: Quoc-Tuan Vien

Copyright © 2020 Huu Trung Nguyen et al. 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 isproperly cited.

Multiple-input multiple-output (MIMO) antenna scheme is an effective technique for future terrestrial broadcasting systems suchas digital video broadcasting-next-generation handheld (DVB-NGH) to overcome the limits on information theory of traditionalsingle-antenna wireless communications without any additional bandwidth or increased transmit power. In this paper, we proposea hybrid beamforming scheme for dual-polarized MIMO spatial multiplexing DVB-NGH broadcasting systems. The system ofinterest makes use of phase shifters and amplitude attenuators for the digital-analog precoder at beamforming stage of thetransmitter end to maximize the signal-to-noise ratio to increase the channel capacity of the DVB-NGH systems. At the receiverend, the maximum likelihood (ML) detector is used to evaluate the system performance. We consider the signal-to-interference-and-noise ratio (SINR) and the achievable average capacity for the DVB-NGH MIMO with various FFT sizes, number oftransmit antennas, and different modulation schemes. The performance results on bit error rate, channel capacity, andbeampatterns show that the proposed hybrid beamforming and dual-polarized MIMO spatial multiplexing schemes providemore robustness against signal interference by beamforming and/or nulling techniques. The simulation results also show thatthe proposed system achieves higher capacity than the existing MIMO schemes for the DVB-NGH systems.

1. Introduction

Digital video broadcasting (DVB) is a family of standardizedtechnologies developed for TV broadcasting over terrestrial,cable, satellite, and mobile communication systems [1]. Inthe last decade, DVB is an area of intensive developmentand standardization activities such as DVB-H (Handheld),Media FLO (Forward Link Only), and DVB-SH (Satellite toHandheld) [2–4]. They are developed to support large-scaleconsumption of mass multimedia services such as mobiletelevision (TV). However, mobile TV services did not fulfillthe initial expectations due to the lack of a successful businessmodel and the high costs associated to the development ofnew mobile broadcasting networks. Today, a new generationof mobile broadcasting technologies is emerging due to thecontinuously increasing requirements and expectations ofboth users and operators that incorporate the latest advancesin wireless communications which provide significant capac-

ity and coverage improvements. To date, the latest emergingmember of DVB family standards is DVB-NGH (next-gener-ation handheld) [5]. It is the mobile evolution of the second-generation digital terrestrial TV broadcasting technologybased on the physical layer of DVB-T2, the most advanceddigital terrestrial TV (DTT) technology in the world [6].Therefore, it can be incorporated in DVB-T2 transmissions,allowing the reuse of spectrum and infrastructure of DVB-T2, offering more robustness, flexibility, and at least 50%more spectrum efficiency than any other technology [7].

DVB-NGH was created with the objectives of increasingthe coverage performance area and network capacity notonly outperforming the previous existing mobile broadcast-ing standards DVB-H and DVB-SH but also optimizingDVB-T2 in many aspects [5]. DVB-NGH is of significantinterest among the wireless operators, since it allows thetransmission of IP (Internet Protocol) mass multimedia con-tent to a wide range of mobile devices, from wearable devices

HindawiInternational Journal of Digital Multimedia BroadcastingVolume 2020, Article ID 9578521, 11 pageshttps://doi.org/10.1155/2020/9578521

such as earphones, mobile phones, and MP3 players to lap-tops and vehicle-mounted receivers at very high data rates[8]. Compared to DVB-T2, DVB-NGH uses Scalable VideoCoding (SVC) technology with Multiple Physical Layer Pipes(MPLPs) for graceful service degradation, TFS (Time-Fre-quency Slicing) technique that enables multiple frequencychannels to be combined into a single wider channel in orderto improve the efficiency and robustness of digital televisionterrestrial (DTT) transmissions [9, 10]. In streaming servicedata, DVB-NGH uses RoHC (Robust Header Compression)method to reduce the overhead due to IP encapsulation,additional satellite component for increased coverage area,improved signaling robustness [5]. DVB-NGH allows effi-cient transmission of local services within enhanced Single-Frequency Network (eSFN) [11]. DVB-NGH improves SFNplanning flexibility in the 4K mode [12]. Finally, DVB-NGH is the first broadcasting system to incorporatemultiple-input multiple-output (MIMO) antenna schemesas the key technology not only to improve the robustness ofthe transmitted signal by exploiting the spatial diversity ofthe MIMO channel but also to achieve increased data ratesthrough spatial multiplexing. Recently, reciprocity calibra-tion for massive MIMO has been studied for future terrestrialbroadcasting technologies [13]. In this work, Luo et al.proposed a novel and innovative closed-loop reciprocitycalibration method for massive MIMO systems with a betterperformance compared to the existing methods. A labmeasurement setup is built for BS hardware impairments’measurement and the implementation.

1.1. Related Works. Nowadays, MIMO is a key technology toincrease the capacity and system reliability without any addi-tional wireless bandwidth for terrestrial broadcasting system[14–16]. Dai et al. addressed in [14] the key technologies andresearch trends for next-generation digital television terres-trial broadcasting systems, including the discussion aboutstatus, technical challenges, and more importantly the futureresearch trends. In addition, MIMO technology, OFDM-based transmission, modulation, and channel coding werefocused as the common technologies to increase the systemcapacity and improve the transmission reliability. Theauthors in [15–17] described the benefits of MIMO thatmotivated its incorporation in DVB-NGH. In these works,the structure of MIMO channel precoders for digital terres-trial TV systems and enhanced spatial multiplexing phasehopping (eSM-PH) as well as the required elements at thetransmitter and receiver terminals to decode the MIMOeSM-PH 2 × 2 transmission was presented against the unpre-coded 4 × 2MIMO scheme. Performance of practical MIMOsystems and compared to SISO using DVB-NGH physicallayer has been assessed for the 2 × 2MIMO scheme with spa-tial multiplexing to support four transmit antennas; MIMOcould provide significant carrier to noise ratio (CNR) reduc-tions [15]. The system capacity in bpc vs. the CNR requiredto achieve the selected QoS criterion for eSM-PH with differ-ent deliberated transmitted power over correlated Rice fadingchannel was clearly provided. Moreover, MIMO rate-2 codesallowed to increase data rates through spatial multiplexing.DVB-NGH MIMO rate-2 schemes were the best option

suited for outdoor medium/high signal use cases such as tab-let PCs and automotive reception as the generally lowersignal-to-noise ratios (SNR) of portable/indoor receptionsubstantially reduce the available multiplexing gain thatmay be exploited [16]. However, in the DVB-NGH physicallayer architecture perspective, these studies did not considerthe MIMO beamforming design on the performance of theMIMO channel precoder.

Traditional MIMO-beamforming systems require a ded-icated radio frequency (RF) chain for each antenna element,which becomes impractical with massive MIMO systems dueto either cost or power consumption. To reduce the numberof RF chains, hybrid beamforming (HBF), which combinesRF analog and baseband digital beamformers, has been pro-posed as a promising solution to enhance the network capac-ity and coverage of next-generation mobile wirelesscommunication [18–20]. Sohrabi and Yu presented in [18,19] the transceiver design for maximizing the spectralefficiency of a large-scale MIMO system with hybrid beam-forming architecture where the number of RF chains is equalto the number of data streams. The hybrid beamformingstructure can achieve the same performance as the fullydigital beamforming scheme if the number of RF chains ateach end is greater than twice the number of data streams.The design could achieve a rate close to that of optimalexhaustive search and use the extra RF chains to significantlyimprove the system performance in the case of low-resolution phase shifters (PSs).

Moreover, millimeter wave (mmWave) cellular systemswill enable gigabit-per-second data rates thanks to the largebandwidth available at mmWave frequencies. Due to thehigh cost and power consumption of gigasample mixed-signal devices, mmWave precoding will likely be dividedamong the analog and digital domains. The large numberof antennas and the presence of analog beamforming requirethe development of mmWave-specific channel estimationand precoding algorithms [20–24]. A hybrid analog/digitalprecoding algorithm that overcomes the hardware constrainson the analog-only beamforming, approaching the perfor-mance of digital solutions, was proposed in [20]. In [21],the problem of mmWave precoder design as a sparsity-constrained signal recovery was formulated using orthogonalmatching pursuit. This framework could be applied to theproblem of designing practical MMSE combiners formmWave systems. Most recently, Magueta et al. proposed ahybrid multiuser equalizer for the uplink of broadbandmmWave systems with dynamic subarray antennas [22].The hybrid subconnected architectures were designed forthe number of required PSs which was lower than in fullyconnected architectures with a set of only analog precodedusers transmitting to a base station and sharing the sameradio resources. At the receiver end, the hybrid multiuserequalizer was designed by minimizing the sum of the meansquare error (MSE) of all subcarriers, considering the digitalpart is iteratively computed as a function of the analog partand the analog equalizer with dynamic antenna mapping thatwas derived to connect the best set of antennas to each RFchain. This designed hybrid dynamic two-step equalizerachieved the performance close to the fully connected

2 International Journal of Digital Multimedia Broadcasting

counterpart, although it is less complex in terms of hardwareand signal processing requirements. In addition, the sameauthors in [23] proposed the iterative analog-digital multi-user equalizer by minimizing the sum of the MSE of the dataestimates over the subcarriers assuming that the analog partwas fixed for all subcarriers while the digital part computedon a per subcarrier basis. The bit error rate (BER) of thehybrid system was derived and compared with other hybridequalizer schemes, which were recently designed formmWave MIMO systems. This technique showed that theperformance of the developed analog-digital multiuser equal-izer was close to the full-digital counterpart and outperformsthe previous hybrid approach. In [24], Hefnawi recently pro-posed a hybrid beamforming scheme for mmWave heteroge-neous networks formed with one macrocell base station (BS)and multiple small-cell BS equipped with large-scale antennaarrays that employ hybrid analog and digital beamforming.Digital beamforming weights were optimized to maximizethe received signal-to-interference-plus-noise ratio (SINR)of the effective channels. In this study, the beampatternsand the ergodic channel capacity were evaluated with onlyfour RF chains while requiring considerably less computationcomplexity. To our knowledge, the hybrid beamformingschemes using analog-digital beamforming and dual-polarized MIMO spatial multiplexing for the DVB-NGHsystem have yet to be addressed in the literature.

In this paper, we therefore present a comprehensivestudy on the performance of the analog-digital beamformingusing dual-polarized MIMO spatial multiplexing for theDVB-NGH systems. Unlike previous works, the objective ofthis architecture is to improve capacity as well as the robust-ness of the received signal. MIMO technique is employed toimprove coverage area and increase the channel capacity ofthe DVB-NGH systems. Therefore, the hybrid beamformingscheme that combines with dual-polarized MIMO spatialmultiplexing is supposed to meet the expectation for thefuture broadcasting systems such as DVB-NGH.

1.2. Main Contributions. In this paper, we propose a solutionon the category of performance-oriented technologies todevelop DVB-NGH systems which includes OFDM-basedtransmission, modulation schemes, and channel coding andespecially employing the MIMO technology. Therefore, ourdesign options include the following:

(i) We design and analyze the performance on bit errorrate and ergodic channel capacity of the hybridbeamforming dual-polarized MIMO spatial multi-plexing with the eSM-PH structure. The proposedarchitecture interest makes use of phase shifters andamplitude attenuators for the digital-analog precoderat beamforming stage to maximize the signal-to-noise ratio to increase the capacity of the DVB-NGH systems

(ii) The dual-polarized MIMO analog precoder isdesigned at both the transmitter and receiver sides.In addition, the maximum likelihood (ML) detectoris used at the hybrid receiver to evaluate the system

performance purposes. For the case studies, weclearly describe the signal-to-interference-and-noiseratio (SINR) and the corresponding achievable aver-age capacity for the DVB-NGH MIMO with variousFFT sizes, number of transmit antennas, and differ-ent modulation schemes

The remainder of this paper is organized as follows.Section 2 describes the system model description adoptedin the work. Section 3 describes a MIMO rate-2 case study.In Section 4, we show the main performance results and dis-cussions on bit error rate, ergodic capacity, and beampatternof the proposed scheme. Finally, the conclusions and direc-tions are presented in Section 5.

1.3. Notations. Capital boldface letters denote matrices, andlower boldface letters denote column vectors. The remainingnotations in this paper are presented in Table 1.

2. System Model Description

In this section, we describe the hybrid beamforming schemefor improving the performance of dual-polarized MIMOspatial multiplexing DVB-NGH broadcasting systems.

The block diagram of the proposed dual-polarizedMIMO spatial multiplexing combining with the digital-analog beamforming for DVB-NGH system is illustrated inFigure 1. We consider the system with K data streams maybe simultaneously transmitted in the same bandwidth usingNT = K transmitter antennas. The transmitted signal thenbeing separated into K respective data streams by way of aset of NR = K ×M antennas deployed at the receiver. In thesystemmodel, we use partially connected structure for hybridbeamforming, where each RF chain is connected to an arrayof M dual-polarized antennas. Such a structure has a lowerhardware complexity compared to fully connected, butbeamforming gain is reduced [25].

Table 1: Notation adopted in the paper.

Operator Description

tr(•) Trace of a matrix

(•)∗ Conjugate of a matrix

(•)T Transpose of a matrix

(•)H Hermitian of a matrix

diag(a)Diagonal matrix where the diagonal

entries are equal to vector a

ARepresents the elements of the n-th row

and m-th column of a matrix

a Represents a vector

a Represents a scalar

EH •½ � Expectation

||A||F The Frobenius norm of A

tr(A) The trace of the matrix A

INmIdentity matrix N ×N

3International Journal of Digital Multimedia Broadcasting

At the transmitter side (Figure 1(a)), the Bit-InterleavedCoded Modulation (BICM) module is one of the mostimportant modules as it provides the error correction capa-bility for the system. The input to the BICM module consistsof one or more logical data streams which are separated fromservices data stream d by a stream partitioner. Each logicaldata stream is carried by one Physical Layer Pipe (PLP) andassociated with a modulation constellation, a forward-errorcorrecting (FEC) protection mode, and a time interleavingdepth [26]. The structure of DVB-NGH BICM module con-sists of serial concatenation of a FEC encoding subsystemwhich performs outer coding using Bose-Chaudhuri-Hocquenghem (BCH) codes, inner coding using a low-density parity-check code (LDPC), bit interleaving (BI), anda constellation mapper (MAP) [27].

2.1. Design of Dual-Polarized MIMO Digital Precoder.Figure 2 illustrates the dual-polarized MIMO precoder as adigital precoder for each data stream input. The precoder isdesigned based on spatial multiplexing (SM) to increase thechannel spectral efficiency by providing reliable performanceover erasure channel [28]. The digital precoder consists of

three different steps of linear precoding, phase hopping,and power calibration. In the first step, the transmittedsymbol is performed by the linear precoding to correlatethe transmitted signal on different transmitting antennas.Therefore, even if one or more channel links encounter era-sure phenomenon, the transmitted signal can still be recov-ered from the other link signals [29]. For each PLP kðk = 1⋯ KÞ, the precoded data symbol can be expressed as

x kð Þ =Θs kð Þ, ð1Þ

where the precoding matrix is determined by

Θ =cos θ −sin θ

sin θ cos θ

" #: ð2Þ

In equation (2), if the rotation phase θ = 33:3° and θ =45°, we have eSM (enhanced SM) and Hadamard SM(hSM), respectively. The rotation phase θ has very littleimpact on the system’s performance [16]. The phase hop-ping eSM (PH-eSM) consists of a phase rotation to the

Stre

am p

artit

ione

r

FEC

Datastream

d

BICM

MAPDual-

polarizedantennaMIMO

precoding

Digitalprecoding

MAP

T/C/F-1

T/C/F-1

OFDM DAC RFh

RFv

1

DAC

Analogprecoder

OFDM

PLP1

S(l)

A11ej𝜙11

A1Mej𝜙1MM

DM

UX

BI

FEC

BICMMAP

Dual-polarizedantennaMIMO

precodingMAP

T/C/F-1

T/C/F-1

OFDM DAC RFh

RFv

1

DACOFDM

PLPK

S(K)

AK1ej𝜙K1

AKMej𝜙KMM

DM

UX

BI

Transmitter

v h

v h

v h

v h

(a) Transmitter

PLP1

FEC

BICM demoddeMAP

MLdecoder

ML decoder

deMAP

T/C/F-dI

T/C/F-dI

OFDMdemodADCRFh

RFv ADCOFDMdemod

MU

X

BDI

PLPK

1

Analogbeamformer

A11ej𝜙11

v h

v h

A1Mej𝜙1M

M

1

AK1ej𝜙K1

AKMej𝜙KM

M

FEC

BICM demoddeMAP

MLdecoder

deMAP

T/C/F-dI

T/C/F-dI

OFDMdemodADCRFh

RFv ADC OFDMdemod

MU

X

BDI

v h

v h

Receiver

(b) Receiver

Figure 1: Block diagram of a hybrid digital-analog beamforming MIMO DVB-NGH system: (a) transmitter side; (b) receiver side.

4 International Journal of Digital Multimedia Broadcasting

symbols of the second transmitting antenna by the multipli-cation with the term ejφtðnÞ. The precoded data symbol withphase hopping is therefore expressed as

x kð Þ =ΨΘs kð Þ, ð3Þ

where the Ψ is given by

Ψ =1 0

0 ejφt nð Þ

" #: ð4Þ

In equation (4), φtðnÞ = ð2π/NÞn, n = 0,⋯, ðNcell/2Þ − 1,and N = 9 is the hopping period and Ncell is the number ofcells per FEC codeword. The phase hopping periodicallychanges the phase of symbols transmitted by one of twoantennas within one FEC block. This is performed by themultiplication with the term of ejφtðnÞ. The phase rotationis initialized to 0° at the beginning of each FEC block andis incremented by 2π/9 for every cell pair. This hoppingphase enhances circularly the robustness against the trans-mitting antenna orientation to increase the diversity [30].

The ST coded data symbols with unequal power can bewritten as

x kð Þ =ΨΘPs kð Þ, ð5Þ

where P is defined as a matrix to scale the power between twoantennas. It can be expressed as

P =ffiffiffiρ

p 0

0ffiffiffiffiffiffiffiffiffiffi1 − ρ

p" #

: ð6Þ

For instance, when QPSK and 16-QAM are applied to thetwo antennas, ρ is set to 1/3. This means twice power is allo-cated to 16-QAM than the QPSK symbols. It is to increasethe robustness of the higher order constellation under theconstraint of the total signal power.

The combination of eSM and the phase hopping is the so-called eSM-PH. The MIMO rate-2 code of the DVB-NGHsystem is expressed as

x kð Þ1

x kð Þ2

24

35 =

1 0

0 ejφt nð Þ

" #cos θ −sin θ

sin θ cos θ

" # ffiffiffiρ

p 0

0ffiffiffiffiffiffiffiffiffiffi1 − ρ

p" #

s kð Þ1

s kð Þ2

24

35:ð7Þ

The digital precoder for the overall DVB-NGH system isrepresented as D = diag ½�d1, �d2,⋯�dK �, where �d =ΨΘΡ ∈ℂ2×2. K data symbols are precoded by D; each symbol xðkÞis then passed through the k-th RF chain.

The digital domain signal from one RF chain is then fedto M transmitting antennas to perform transmit analog pre-coding. The analog precoder vector is expressed as

�ak = ak1, ak2,⋯akM½ �T ∈ℂM×1, ð8Þ

where aij = Aijejϕi j , Aij is the amplitude factor and ϕij is the

phase shift. Finally, every data symbol is transmitted by thesubantenna array of M antennas.

2.2. Receivers. Consider the receiver side, where the hybridanalog-digital beamforming architecture is shown inFigure 1(b). The received signal for all K data symbols r =½r1, r2,⋯rK �T is expressed as

r =ffiffiffiffiffiffiffiffiffiSNR

pHADx + n =

ffiffiffiffiffiffiffiffiffiSNR

pHGx + n, ð9Þ

where �hk = ½h11, h12,⋯, hkM� ∈ℂ1×M , hk = ½01×Mðk−1Þ, �hk,01×MðK−kÞ�∈ℂ1×KM , and H = ½h1, h2,⋯, hK � ∈ℂK×KM that isgiven by

00

0 0

00

0

01

2

K

H = =

… ……

…

…

……

…

…

…

h11 h12 h1M

.

hK1

hK2

hKM

ð10Þ

The analog precoder A of the receiver is represented inthe following matrix form:

A =

�a1 00 �a2

⋯ 0⋮ 0

⋮ ⋮

0 0⋱ ⋮

⋮ �aK

266664

377775: ð11Þ

In this expression, the analog precoder vector of receiver,�ai, is defined similarly to those of the transmitter. K datasymbols are represented as x = ½x1, x2,⋯, xK �T . Moreover,in equation (9), the total noise of n = ½n1, n2,⋯, nK �T , wherenk is the complex Gaussian random variable with zero means

MIMOprecoding

DM

UX

Pow

er sc

ale

cos 𝜃

S1(k)

x1(k)

S2(k)

x2(k)

ej𝜙t (n)

S(k)

sin 𝜃 cos 𝜃–sin 𝜃

MAP

MAP

Figure 2: Dual-polarized antenna MIMO precoder.

5International Journal of Digital Multimedia Broadcasting

and a variance of σ2, and G =AD represents the joint hybridprecoding matrix with the size of KM × K .

2.3. Capacity Analysis. Recall the ergodic capacity of MIMOchannel assuming perfect channel knowledge at both trans-mitter and receiver with zero-mean Gaussian distributioninputs. The ergodic capacity can be expressed in general formas [31]

CH = EH maxQ:Tr Qð Þ=P

log2 INm+ SNRHQHH�� ��� �

, ð12Þ

where the covariance of the input data Q is written as

Q =GGH : ð13Þ

Therefore, equation (12) is given by

CH =Nm log2 det INm+ SNRHQXHH� �� �

, ð14Þ

where INmdenotes the identity matrix of dimensions Nm ×

Nm, QX = ð1/NmÞINm, and HH is the complex conjugate

transpose of H, Nm =min ðNT ,NRÞ. When the channelmatrix is square and orthogonal (HHH = I), then with anidentically distributed data input channel capacity can berewritten as [32]

CH = Nm log2 1 +SNRNm

� : ð15Þ

The capacity is linearly scaled with the number oftransmitter antennas for an increasing SNR. In general, itcan be demonstrated that an orthogonal channel as theone used in the previous example maximizes the capacityin MIMO systems. In an identically distributed channelwith flat fading, the channel matrix becomes almostorthogonal when the number of transmitter antennas ishigh [33]. When the number of transmitter and receiver

antennas is different, the capacity increase is limited tothe minimum number of them. The purpose of analogbeamforming is to raise the SNR. In the next section, weconsider a MIMO rate-2 case and estimate the SNR forcapacity evaluation.

3. MIMO Rate-2 Case Study

In this section, we describe the MIMO spatial multiplexingthat is specified in DVB-NGH as MIMO rate-2 codes. Theterm “rate-2” basically stands for the transmission of twoindependent streams. MIMO rate-2 codes in DVB-NGHuse cross-polar antenna arrangement (antennas with orthog-onal polarization) with two transmitter antennas and tworeceiver antennas (the 2 × 2 MIMO system).

The DVB-NGH MIMO rate-2 receiver diagram is illus-trated in Figure 3. In this scheme, we clarify the amplitudefactor of Ai and the phase factor of ϕi. Considering the hori-zontally polarized part of the system, the received signal canbe written as

r1 = h1x + n1,

r2 = A1ejϕ1h2x + A1e

jϕ1n2,

y hð Þ = r1 + r2 = x h1 + A1ejϕ1h2

� �+ n1 + A1e

jϕ1n2,

ð16Þ

where superscript h denotes the polarized antenna hori-zontally, r1 and r2 represent the received signal at the firstreceiver antenna and the received signal at the secondreceiver antenna, respectively. x, n1, and n2 are the trans-mitted signal, the noise component at the first receiverantenna, and the parameter representing noise componentat the second receiver antenna, respectively. θ1 and A1 arethe phase factor between the received signal at the firstand second receiver antennas and the amplitude factor,respectively.

Ant #1 hV r1

r2

ML

deco

der

MU

X

𝜃1A1

y(h) y(h)

VhAnt #2

S

ˆ

ˆ

Figure 3: The DVB-NGH MIMO rate-2 receiver diagram.

Table 2: Study cases.

Case Study cases

1BER for the MIMO rate-2 mode, FFT 8K/16K,

64- or 256-QAM constellation modulation schemes

2Capacity for the MIMO rate-2 mode and Ns = 4,

16 K FFT mode, 64-QAM

Table 3: Simulation parameters.

Parameter Value

PLP 2, 4

Number of substreams in Demux 2

Number of transmitter antennas 1, 2, 4, 8

Number of RF chains 2, 4

Number of receiver antennas 2, 4

Bandwidth 8MHz

Modulation 64-QAM and 256-QAM

Carriers (FFT size) 8K, 16K

Constellation rotation2D (b = 0:2890 16-QAM,b = 0:1495 64-QAM)

Interleavers B-C-T-FI

FEC encoding and code rate 16 K LDPC code 8/15

Guard interval 1/16

PAPR reductionActive constellationextension (ACE)

6 International Journal of Digital Multimedia Broadcasting

When the DVB-NGH MIMO system uses N transmitterand M receiver antennas, the received signal at the receiverside can be represented as

rk = 〠N

i=1Ak−1e

jϕk−1hikx + Ak−1ejϕk−1nk

� �, ð17Þ

and the output of the analog beamformer is expressed as

yk = 〠M

i=1rk, ð18Þ

where A0 = 1 and ϕ0 = 0. For the case study of the DVB-NGHMIMO rate-2 system, the SNR can be determined by maxi-

mizing the following expression for various phase shift ϕand amplitude A factors as

SNR =h1 + A1e

jϕ1h2 2

E n1k k2 + E A1ejϕ1n2 2 =

h1 + A1ejϕ1h2

2σ2 1 + A2

1� � , ð19Þ

where σ2 is the noise power.In addition, multiple interferences in the DVB-NGH

MIMO rate-2 system can be expressed as

r1 = h1x + hI1sI + n1,

r2 = A1ejϕ1 h2x + hI2sI + n2ð Þ,

y hð Þ = r1 + r2 = x h1 + A1ejϕ1h2

� �+ sI hI1 + A1e

jϕ1hI2� �

+ n1 + A1ejϕ1n2,

ð20Þ

–15 –10 –5 0 5 10Eb/No (dB)

10–4

10–2

100

BER

MIMO rate-2, FFT: 8K, 64-QAM, M = 1MIMO rate-2, FFT: 8K, 256-QAM, M = 1MIMO rate-2, FFT: 16K, 64-QAM, M = 1MIMO rate-2, FFT: 16K, 256-QAM, M = 1

(a) BER vs. Eb/No for various modulation schemes, the transmitter

antennas of M = 1

–15 –10 –5 0 5 10Eb/No (dB)

10–4

10–2

100

BER

MIMO rate-2, FFT: 8K, 64-QAM, M = 2MIMO rate-2, FFT: 8K, 256-QAM, M = 2MIMO rate-2, FFT: 16K, 64-QAM, M = 2MIMO rate-2, FFT: 16K, 256-QAM, M = 2

(b) BER vs. Eb/No for various modulation schemes and the FFT size,

the transmitter antennas of M = 2

–15 –10 –5 0 5 10Eb/No (dB)

10–8

10–6

10–4

10–2

100

BER

MIMO rate-2, FFT: 8K, 64-QAM, M = 4MIMO rate-2, FFT: 8K, 256-QAM, M = 4MIMO rate-2, FFT: 16K, 256-QAM, M = 4MIMO rate-2, FFT: 16K, 64-QAM, M = 4

(c) BER vs. Eb/No for various modulation schemes, the transmitter antennas of M = 4

Figure 4: The BER performance against Eb/No for MIMO rate-2 using the modulation schemes of 64-QAM and 256-QAMwith the FFT sizeof 8 K and 16K for various numbers of antenna element M at the analog precoder.

7International Journal of Digital Multimedia Broadcasting

where sI represents the interference signal. For the case ofDVB-NGH MIMO system using N transmitter and Mreceiver antennas, the received signal is represented as

rk = 〠N

i=1Ake

jϕkhikx + AkejϕkhIksI + Ake

jϕknk� �

, ð21Þ

and the output of the analog beamformer is expressed as

yk = 〠M

i=1rk: ð22Þ

The signal-to-interference-and-noise ratio (SINR) whichcorresponds to the signal strength parameter can be thendetermined for the DVB-NGH MIMO rate-2 system. The

SINR can be determined by maximizing the following expres-sion for various phase shift ϕ and amplitude A factors as

SINR =h1 + A1e

jϕ1h2 2

E n1k k2 + E A1ejϕ1n2 2 + hI1 + A1ejϕ1hI2

2=

h1 + A1ejϕ1h2

2σ2 1 + A2

1� �

+ hI1 + A1ejϕ1hI2 2 ,

ð23Þ

where σ2 is the noise power.When the desired signal has i = 1,⋯, P multiple paths

with different delays and the interfering signal has multiplepaths k − 1,⋯, R with different delays, then the maximumSINR solution for the DVB-NGH MIMO rate-2 system inthat case is given by

SINRmax =∑P

i=1 h1 + A1ejϕ1h2

2σ2 1 + A2

1� �

+∑Rk=1 hI1 + A1ejϕ1hI2

2 : ð24Þ

The corresponding achievable average capacity can berewritten as [32]

CH = EH log2 1 + SINRmaxð Þ½ �: ð25Þ

To reduce the computational complexity, numerous sub-optimal MIMO receivers have been used such as the linearzero-forcing (ZF) and the minimum mean square error(MMSE) receivers [34]. The optimum receiver, i.e., the MLdetector at the receiver, evaluates the squared Euclidean dis-tanceDðx1, x2Þ and selects the coupe ðx1, x2Þminimizing thisdistance. The squared Euclidean distance can be expressed as

D x1, x2ð Þ = Y −Hxk k2F = r1 − h11x1 − h12x2k k2+ r2 − h21x1 − h22x2k k2,

ð26Þ

–15 –10 –5 0 5 10Eb/No (dB)

10–6

10–4

10–2

100BE

R

MIMO rate-2, FFT: 16K, 256-QAM, M = 4MIMO rate-2, FFT: 16K, 256-QAM, M = 8MIMO rate-2, FFT: 16K, 256-QAM, M = 1MIMO rate-2, FFT: 16K, 256-QAM, M = 2

Figure 5: The BER performance against Eb/No for MIMO rate-2using the modulation scheme of 256-QAM and the FFT size of16K for various transmitter antennas of M = 1, 2, 4, and 8.

–10 0 10 20 30 40Eb/No (dB)

0

5

10

15

20

25

Capa

city

(bit/

s/H

z)

MIMO rate-2, 64-QAM, LDPC 8/15, M = 1MIMO rate-2, 64-QAM, LDPC 8/15, M = 2MIMO rate-2, 64-QAM, LDPC 8/15, M = 4MIMO rate-2, 64-QAM, LDPC 8/15, M = 8

Figure 6: Capacity achieved by various transmitter antennas for aMIMO rate-2 mode with the modulation scheme of 64-QAM andthe inner coding of LDPC 8/15.

–10 0 10 20 30 400

10

20

30

40

50

MIMO rate-4, 64-QAM, LDPC 8/15, M = 1MIMO rate-4, 64-QAM, LDPC 8/15, M = 2MIMO rate-4, 64-QAM, LDPC 8/15, M = 4MIMO rate-4, 64-QAM, LDPC 8/15, M = 8

Eb/No (dB)

Capa

city

(bit/

s/H

z)

Figure 7: Capacity achieved by various transmitter antennas for aMIMO Ns = 4 mode with the modulation scheme of 64-QAM andthe inner coding of LDPC 8/15.

8 International Journal of Digital Multimedia Broadcasting

where k:k2F is the Frobenius norm and Y is the receivednoisy signal.

4. Simulation Results

In this section, we show the BER performance of the pro-posed system by means of the Monte Carlo simulation. Thesimulation estimates the capacity performance versus thesignal-to-noise ratio and bit error rate versus Eb/No ratio.Table 2 shows two study cases for the performance evalua-tion of the proposed system. In the first case, the MIMOrate-2 for the FFT 8K/16K mode is performed using 64-QAM or 256-QAM modulation constellation schemes. Inthe second case, the MIMO rate-2 for the number of sub-streams of 2 and 4 over the 16K FFT mode is performedusing 64-QAM. Within each study case, we evaluate systemparameters with various analog precoder configurations.The detailed simulation parameters are presented inTable 3. The simulations include inner LDPC codes with aword length size of 16200 bits (16K) and the outer BCH.

Figure 4 shows the bit error rate (BER) performance ver-sus Eb/No for the variousM values of 1 and 2.We evaluate allcombinations of constellation orders 64-QAM and 256-QAM with the FFT size of 8K and 16K. It is clearly notedthat the system’s performance significantly increases whenthe power gain increases. In addition, figures also show thatincreasing the FFT size severely improves the BER perfor-mance, especially at the power gain of 10 dB. However, forthe case ofM = 4, it is noted that the best system performancecan be achieved at the power gain of approximately 7.5 dB forthe case of MIMO rate-2, the FFT size of 16 using 64-QAMmodulation scheme.

In Figure 5, we show the BER performance of the DVB-NGH system versus the Eb/No for MIMO rate-2 using the256-QAM modulation scheme and the FFT size of 16K forvarious transmitter antennas of M = 1, 2, 4, and 8. It is alsoseen that the BER decreases with the increase of Eb/No. Inaddition, the increase of the number of transmitter antennassignificantly increases the DVB-NGH system performance.The power gain of approximately 5 dB is seen when increas-ing the number of transmitter antennas from 1 to 2, 2 to 4,and 4 to 8, at the target BER less than 10-3.

Now, let us evaluate the proposed hybrid beamformingdual-polarized MIMO spatial multiplexing with the eSM-

PH structure. Figures 6 and 7 present the system capacityachieved by the modulation scheme of 64-QAM, the innercoding of LDPC 8/5, and a frame error rate (FER) 1% afterBCH. The analyzed schemes are MIMO rate-2 and MIMOwith four RF chains (Ns = 4) with M = 1, 2, 4, 8 for perfor-mance comparison purposes. The channel model used forthis case study is uncorrelated Rayleigh fading in which theimpairment correlation factor between the antenna elementsis 1%. The results illustrate a better performance with variousM values. The beamforming gains achieved with LDPC coderate 8/15 at BER of 10−4 decoding increase with the numberof elements at analog precoder, i.e., 3 dB (M = 2), 6 dB(M = 4), and 8.9 dB (M = 8).

Finally, Figure 8 shows the beampattern of the proposedhybrid beamformer with two RF chains (Ns = 2) and num-ber of the antennas MT = 8 and MR = 2 (Figure 8(a)) andfour RF chains (Ns = 4) MT = 16 and MR = 4 (Figure 8(b)).It can be seen from the proposed scheme that the optimizedbeamformer has dominant beams. This beampattern meansthat the data streams can be successfully transmittedthrough those beams.

5. Conclusions

In this paper, we developed a new beamforming schemebased on hybrid beamforming and dual-polarized MIMOspatial multiplexing for DVB-NGH systems. In the proposedapproach, the beamforming scheme has maximized thechannel capacity of the MIMO-based DVB-NGH systems.Simulation results showed that the proposed hybrid beam-forming is efficient to achieve higher capacity than theexisting dual-polarized MIMO spatial multiplexing for theDVB-NGH systems. The performance evaluation in termsof the bit error rate, the ergodic channel capacity, and thebeampatterns showed that the proposed hybrid beamform-ing schemes using analog-digital beamforming and dual-polarized MIMO spatial multiplexing for the DVB-NGHsystem have dominant beams. Although some achievementshave been achieved, some challenging problems remain tobe solved, for example, physical size reduction. Therefore,in the future work, the hybrid beamforming configurationscheme could be optimized for the handheld receiver inDVB-NGH systems.

yAz 90EI 0x

Az 0

el

azEI 0

zAz 0EI 90

(a)

y

Az 90EI 0x

Az 0

el

azEI 0

zAz 0EI 90

(b)

Figure 8: Beampattern of the proposed beamforming scheme: (a)MT = 8 andMR = 2 for theMIMO rate-2 mode; (b)MT = 16 andMR = 4 forthe MIMO Ns = 4 mode.

9International Journal of Digital Multimedia Broadcasting

Data Availability

The data used to support the findings of this study areincluded within the article. Simulation experiments were car-ried out using MATLAB®. All data used to support the find-ings of this study are included within the article. The latterare available from the corresponding author upon request.

Conflicts of Interest

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

Acknowledgments

This work is supported by Ministry of Science and Tech-nology of Vietnam under Project No. NDT.32.ITA/17.

References

[1] D. Gómez-Barquero, Ed., Next Generation Mobile Broadcast-ing, CRC Press, 2013.

[2] G. Faria, J. A. Henriksson, E. Stare, and P. Talmola, “DVB-H:digital broadcast services to handheld devices,” Proceedings ofthe IEEE, vol. 94, no. 1, pp. 194–209, 2006.

[3] M. R. Chari, F. Ling, A. Mantravadi et al., “FLO physical layer:an overview,” IEEE Transactions on Broadcasting, vol. 53,no. 1, pp. 145–160, 2007.

[4] I. Andrikopoulos, N. Chuberre, M. Cohen et al., “An overviewof digital video broadcasting via satellite services to handhelds(DVB-SH) technology,” in Handbook of Mobile Broadcasting:DVB-H, DMB, ISDB-T and MediaFLO, CRC Press, 2008.

[5] DVB Document A160, Digital video broadcasting (DVB); nextgeneration broadcasting system to handheld, physical layerspecification (DVB-NGH), 2012.

[6] L. Vangelista, N. Benvenuto, S. Tomasin et al., “Key technolo-gies for next-generation terrestrial digital television standardDVB-T2,” IEEE Communications Magazine, vol. 47, no. 10,pp. 146–153, 2009.

[7] I. Eizmendi, G. Prieto, G. Berjon-Eriz, I. Landa, and M. Velez,“Empirical DVB-T2 thresholds for fixed reception,” IEEETransactions on Broadcasting, vol. 59, no. 2, pp. 306–316,2013.

[8] D. Gómez-Barquero, C. Douillard, P. Moss, and V. Mignone,“DVB-NGH: the next generation of digital broadcast servicesto handheld devices,” IEEE Transactions on Broadcasting,vol. 60, no. 2, pp. 246–257, 2014.

[9] J. J. Giménez, E. Stare, S. Bergsmark, and D. Gomez-Barquero,“Advanced network planning for time frequency slicing (TFS)toward enhanced efficiency of the next-generation terrestrialbroadcast networks,” IEEE Transactions on Broadcasting,vol. 61, no. 2, pp. 309–322, 2015.

[10] H. Schwarz, D. Marpe, and T. Wiegand, “Overview of the scal-able video coding extension of the H.264/AVC standard,”IEEE Transactions on Circuits and Systems for Video Technol-ogy, vol. 17, no. 9, pp. 1103–1120, 2007.

[11] J. Lopez-Sanchez, J. Zollner, S. Atungsiri, E. Stare, andD. Gomez-Barquero, “Technical solutions for local serviceinsertion in DVB-NGH single frequency networks,” IEEETransactions on Broadcasting, vol. 60, no. 2, pp. 293–301,2014.

[12] Y. Zhang, C. Zhang, J. Cosmas et al., “Analysis of DVB-H net-work coverage with the application of transmit diversity,”IEEE Transactions on Broadcasting, vol. 54, no. 3, pp. 568–577, 2008.

[13] H. Luo, Y. Zhang, L.-K. Huang, J. Cosmas, and A. Aggoun, “Aclosed-loop reciprocity calibration method for massive MIMOin terrestrial broadcasting systems,” IEEE Transactions onBroadcasting, vol. 63, no. 1, pp. 11–19, 2017.

[14] L. Dai, Z. Wang, and Z. Yang, “Next-generation digital televi-sion terrestrial broadcasting systems: key technologies andresearch trends,” IEEE Communications Magazine, vol. 50,no. 6, pp. 150–158, 2012.

[15] D. Vargas, Y. J. D. Kim, J. Bajcsy, D. Gomez-Barquero, andN. Cardona, “A MIMO-channel-precoding scheme for nextgeneration terrestrial broadcast TV systems,” IEEE Transac-tions on Broadcasting, vol. 61, no. 3, pp. 445–456, 2015.

[16] D. Vargas, D. Gozalvez, D. Gomez-Barquero, and N. Cardona,“MIMO for DVB-NGH, the next generation mobile TVbroadcasting [accepted from open call],” IEEE Communica-tions Magazine, vol. 51, no. 7, pp. 130–137, 2013.

[17] J. H. Seo, T. J. Jung, H. M. Kim, and D. S. Han, “Improvedpolarized 2x2 MIMO spatial multiplexing method for DVB-NGH system,” IEEE Transactions on Broadcasting, vol. 61,no. 4, pp. 729–733, 2015.

[18] F. Sohrabi and W. Yu, “Hybrid digital and analog beamform-ing design for large-scale MIMO systems,” in 2015 IEEE Inter-national Conference on Acoustics, Speech and Signal Processing(ICASSP), Brisbane, Australia, 2015.

[19] F. Sohrabi and W. Yu, “Hybrid digital and analog beamform-ing design for large-scale antenna arrays,” IEEE Journal ofSelected Topics in Signal Processing, vol. 10, no. 3, pp. 501–513, 2016.

[20] A. Alkhateeb, O. El Ayach, G. Leus, and R. Heath, “Channelestimation and hybrid precoding for millimeter wave cellularsystems,” IEEE Journal of Selected Topics in Signal Processing,vol. 8, no. 5, pp. 831–846, 2014.

[21] O. El Ayach, S. Rajagopal, S. Abu-Surra, Z. Pi, and R. Heath,“Spatially sparse precoding in millimeter wave MIMO sys-tems,” IEEE Transactions on Wireless Communications,vol. 13, no. 3, pp. 1499–1513, 2014.

[22] R. Magueta, D. Castanheira, A. Silva, R. Dinis, andA. Gameiro, “Hybrid multi-user equalizer for massive MIMOmillimeter-wave dynamic subconnected architecture,” IEEEAccess, vol. 7, pp. 79017–79029, 2019.

[23] R. Magueta, D. Castanheira, P. Pedrosa, A. Silva, R. Dinis, andA. Gameiro, “Iterative analog–digital multi-user equalizer forwideband millimeter wave massive MIMO systems,” Sensors,vol. 20, no. 2, p. 575, 2020.

[24] M. Hefnawi, “Hybrid beamforming for millimeter-wave het-erogeneous networks,” Electronics, vol. 8, no. 2, p. 133, 2019.

[25] E. G. Larsson, “Joint beamforming and broadcasting in mas-sive MIMO,” in 2015 IEEE 16th International Workshop onSignal Processing Advances in Wireless Communications(SPAWC), pp. 266–270, Stockholm, Sweden, 2015.

[26] I.-W. Kang, K. H. Suh, H. M. Kim et al., “Performance ofMIMO bit division with polarized MIMO DVB-T2,” inProceedings of International Conference on Systems andElectronic Engineering (ICSEE'2012), pp. 57–60, Phuket,Thailand, 2012.

[27] C. Douillard and C. A. Nour, “The bit interleaved coded mod-ulation module for DVB-NGH: enhanced features for mobile

10 International Journal of Digital Multimedia Broadcasting

reception,” in 2012 19th International Conference on Telecom-munications (ICT), pp. 1–6, Jounieh, Lebanon, 2012.

[28] N. H. Trung, N. T. Anh, N. M. Duc, D. T. Binh, and L. T. Tan,“System theory based multiple beamforming,” Vietnam Jour-nal of Science and Technology, vol. 55, no. 5, p. 637, 2017.

[29] N. H. Trung and D. T. Binh, “Large-scale MIMO MC-CDMAsystem using combined multiple beamforming and spatialmultiplexing,” Vietnam Journal of Science and Technology,vol. 56, no. 1, pp. 102–112, 2018.

[30] F. Zheng, Y. Chen, B. Pang et al., “An efficient CSI feedbackscheme for dual-polarized massive MIMO,” IEEE Access,vol. 6, pp. 23420–23430, 2018.

[31] O. Alluhaibi, Q. Z. Ahmed, C. Pan, and H. Zhu, “Capacitymaximisation for hybrid digital-to-analog beamforming mm-wave systems,” in 2016 IEEE Global Communications Confer-ence (GLOBECOM), Washington, DC, USA, 2016.

[32] E. Telatar, “Capacity of multi-antenna Gaussian channels,”European Transactions on Telecommunications, vol. 10,no. 6, pp. 585–595, 1999.

[33] M. Abdelaziz, L. Anttila, A. Brihuega, F. Tufvesson, andM. Valkama, “Digital predistortion for hybrid MIMO trans-mitters,” IEEE Journal of Selected Topics in Signal Processing,vol. 12, no. 3, pp. 445–454, 2018.

[34] X. Liu, Q. Zhang, W. Chen et al., “Beam-oriented digitalpredistortion for 5G massive MIMO hybrid beamformingtransmitters,” IEEE Transactions on Microwave Theory andTechniques, vol. 66, no. 7, pp. 3419–3432, 2018.

11International Journal of Digital Multimedia Broadcasting