Received May 4, 2017, accepted July 10, 2017, date of publication July 18, 2017, date of current version

August 8, 2017.
Digital Object Identifier 10.1109/ACCESS.2017.2727550

5G Cellular User Equipment: From Theory

to Practical Hardware Design
YIMING HUO1 , (Student Member, IEEE), XIAODAI DONG1 , (Senior Member, IEEE),
AND WEI XU2 , (Senior Member, IEEE)
1 Department of Electrical and Computer Engineering, University of Victoria, Victoria, BC V8P 5C2, Canada
2 National Mobile Communications Research Laboratory, Southeast University, Nanjing 210096, China
Corresponding author: Yiming Huo ([email protected])
This work was supported in part by the NSERC of Canada under Grant 261524. The work of W. Xu was supported in part by the NSFC
under Grant 61471114 and in part by the Six Talent Peaks Project in Jiangsu Province under Grant GDZB-005.

ABSTRACT Research and development on the next generation wireless systems, namely 5G, has
experienced explosive growth in recent years. In the physical layer, the massive multiple-input-multiple-
output (MIMO) technique and the use of high GHz frequency bands are two promising trends for adoption.
Millimeter-wave (mmWave) bands, such as 28, 38, 64, and 71 GHz, which were previously considered not
suitable for commercial cellular networks, will play an important role in 5G. Currently, most 5G research
deals with the algorithms and implementations of modulation and coding schemes, new spatial signal
processing technologies, new spectrum opportunities, channel modeling, 5G proof of concept systems, and
other system-level enabling technologies. In this paper, we first investigate the contemporary wireless user
equipment (UE) hardware design, and unveil the critical 5G UE hardware design constraints on circuits and
systems. On top of the said investigation and design tradeoff analysis, a new, highly reconfigurable system
architecture for 5G cellular user equipment, namely distributed phased arrays based MIMO (DPA-MIMO)
is proposed. Finally, the link budget calculation and data throughput numerical results are presented for the
evaluation of the proposed architecture.

INDEX TERMS 5G, massive multiple-input-multiple-output (MIMO), millimeter-wave (mmWave),

beamforming, distributed phased array, user equipment (UE), hardware, system-on-chip (SoC), spectral
efficiency.

NOMENCLATURE EMI Electromagnetic Interference

ET Envelop Tracking
3GPP Third Generation Partnership
FCC Federal Communications Commission
Project (3GPP)
FDD Frequency Division Duplex
5G Fifth Generation
ADC Analog-to-Digital Converter FOM Figure of Merit
AGC Automatic Gain Control FinFET Fin Field Effect Transistor
BB Baseband GNSS Global Navigation Satellite System
BF Beamforming IRR Infrared Reflective
BS Base Station LNA Low Noise Amplifier
CA Carrier Aggregation LOS Line-of-Sight
CC Carrier Component Li-ion Lithium-ion
CMOS Complementary LO Local Oscillator
Metal-Oxide-Semiconductor MIMO Multiple-Input-Multiple-Output
DAC Digital-to-Analog Converter MLB Main Logic Board
DCPS Digitally Controlled Phase Shifter mmWave Millimeter-Wave
DPA-MIMO Distributed Phased Array MIMO NiCd Nickel-Cadmium
EIRP Effective Isotropic Radiated Power NiMH Nickel-Metal Hydride
EMC Electromagnetic Compatibility NLOS Non-Line-of-Sight

2169-3536
2017 IEEE. Translations and content mining are permitted for academic research only.
13992 Personal use is also permitted, but republication/redistribution requires IEEE permission. VOLUME 5, 2017
See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Y. Huo et al.: 5G Cellular User Equipment: From Theory to Practical Hardware Design

PA Power Amplifier maximum 100 MHz bandwidth (BW) that a mobile network
PAPR Peak-to-Average Power Ratio operator currently can support through enabling 5 CCs. If the
PCB Printed-Circuit Board RF bandwidth is fixed, such high SE requirement leads to
PDLT Peak Downlink Throughput using either higher order modulation scheme, more layers
PMU Power Management Unit of SM, or both. The PDLT is given by
PULT Peak Uplink Throughput
QAM Quadrature Amplitude Modulation PDLT (BRF NQAM NCA NMIMO ) (1)
QoE Quality of Experience where BRF is the RF bandwidth for one single carrier,
QoS Quality of Service NQAM represents the modulation order, NCA is the number
RF Radio Frequency of aggregated carriers, and NMIMO stands for the number of
RSSI Received Signal Strength Indicator MIMO spatial multiplexing layer. Therefore, the SE can be
RX Receiver expressed as
SAR Specific Absorption Rate
SE Spectral Efficiency SE (NQAM NCA NMIMO ). (2)
SiGe Silicon-Germanium Nevertheless, from the implementation point of view, high
SM Spatial Multiplexing modulation order and wide RF bandwidth unavoidably
SNR Signal-to-Noise Ratio require power-hungry, complicated and high-performance RF
SoC System-on-Chip and baseband circuits. On the other hand, high order of
SPDT Single Port Double Throw MIMO is confronted with the limitation of antennas physical
TDD Time Division Duplex dimension, spacing, and radiation efficiency (RE). Based
TSSI Transmitted Signal Strength Indicator on emerging UE design techniques, a mobile phone hand-
TX Transmitter set can accommodate at most 4 4 MIMO antennas with
UE User Equipment 256-QAM modulation [5], which theoretically boosts PDLT
UMa Urban Macrocell up to approximately 1960 Mbps when using 5 CCs. On the
UMi Urban Microcell other hand, such UE design already reaches the maximum
VGA Variable Gain Amplifier spatial multiplexing gain due to the limited hardware area for
embedding MIMO antennas of low GHz frequency bands.
I. INTRODUCTION As a result, at the UE end, the highest achievable SE is around
In the cellular world, tremendous efforts have been devoted 20 bits/s/Hz. With the said SE, achieving 10 Gbps PDLT
to delivering higher quality of service (QoS) and quality would require at least 500 MHz bandwidth, which is currently
of experience (QoE) since the very first cell phone call not possible in terms of the limited spectrum holdings of
was made in 1973. In the Third Generation Partnership service providers.
Project (3GPP) roadmap, several representative techniques Meanwhile, Wi-Fi technologies have been advancing
have marked the milestones to further accelerate such trend, rapidly. The cutting-edge off-the-shelf IEEE 802.11ac com-
namely: device-to-device (D2D) communication for boost- patible wireless products can support a 4 4 multi-user
ing geographic spectrum reusability [1]; heterogeneous and MIMO (MU-MIMO) for downlink. By using unlicensed
small-cell network (HetSNet) targeting at deploying small 60 GHz millimeter-wave (mmWave) frequency bands, wire-
cells in addition to macrocells at the same or different carrier less gigabit alliance (WiGig) IEEE 802.11ad products deliver
frequencies [2]; carrier aggregation (CA) for larger radio even higher data rate for short-range communications [5]. The
frequency (RF) bandwidths; new carrier type (NCT) for yet to be released IEEE 802.11ax and 802.11ay standards
increasing spectral efficiency and spectrum flexibility, and that are deemed as the successors of 802.11ac and 802.11ad
reducing interference and power consumption [3]; higher respectively, are expected to provide improved QoS such as
order modulation schemes and more layers of spatial mul- better communication coverage and reduced latency. Besides
tiplexing (SM) for higher spectral efficiency (SE). In 3GPP WiFi, Bluetooth, near-field communication (NFC) and global
release 12, 8 8 multiple-input-multiple-output (MIMO) navigation satellite system (GNSS) are also integrated in
for downlink, 4 4 MIMO for uplink, 5 carrier com- mobile handset terminals. These wireless technologies will
ponents (CCs) and 256 quadrature amplitude modula- compete with cellular for hardware resources and design
tion (QAM) are supported to satisfy the increasing wireless budget on an already highly compact, multi-functional, multi-
capacity needs. standard wireless handset terminal. For example, a huge chal-
As part of the QoS requirement of 5G networks, the lenge is to integrate antenna systems for different wireless
5G peak downlink throughput (PDLT) is expected to achieve technologies that occupy a very wide range of frequency
10 Gbps in the dense urban environments [4]. Delivering (from 700 MHz to almost 6 GHz), and implementing high
such high speed data to end-users is an essential prerequi- order MIMO can make it even more serious considering the
site for a satisfying QoE which is perceived subjectively. limited dimension of a mobile handset device.
Furthermore, this high PDLT can be translated into a very In light of these challenging issues in sub-6 GHz, the high
high SE requirement of at least 100 bits/s/Hz based on the GHz frequency bands used to be considered unsuitable for

VOLUME 5, 2017 13993

 Y. Huo et al.: 5G Cellular User Equipment: From Theory to Practical Hardware Design

commercial cellular networks are now attracting significant

attention. On July 14, 2016, the Federal Communications
Commission (FCC) voted to adopt a new Upper Microwave
Flexible Use service in the licensed bands, namely 28 GHz
(27.5-28.35 GHz), 37 GHz (37-38.6 GHz), 39 GHz (38.6-
40 GHz), plus a new unlicensed band at 64-71 GHz [7]. This
initiative taken by the FCC helps mitigate the 5G UE design
challenges. First, larger continuous RF bandwidth enables
higher data rates. Second, using mmWave frequencies leads
to a significant reduction of antenna dimension, and as a
result, the form factor of the UE can be maintained while
facilitating beamforming (BF) and spatial multiplexing.
In this paper, we present a novel distributed phased array
MIMO (DPA-MIMO) architecture for 5G UE hardware
design. The remainder of this paper is organized as follows, FIGURE 1. Increment of the wireless capacity versus the battery
Section II investigates the contemporary wireless UE design performance improvement.
and explains the design constraints. Section III presents the
details of 5G UE hardware design challenges, and proposes
a novel system architecture to address these challenges. production, higher energy efficiency of the UE wireless sys-
Section IV conducts the performance evaluation of the tem will be critically relevant.
proposed system based on the link budget calculation
and comparison with state-of-the-art 5G works. Finally, B. CIRCUIT AND SYSTEM DESIGN
Section V concludes this paper. The performance of a wireless system, from a hardware
perspective, depends on the evolution of design arts in
II. WIRELESS UE DESIGN OVERVIEW system-on-chip (SoC), printed-circuit board (PCB), mechan-
Contemporary wireless UE design becomes more chal- ical design, and antenna design. The SoCs of high energy
lenging and complicated than ever. In any mainstream efficiency, small area, low cost and high yield, are always
mobile phone, it does not only need to co-exist with sev- strongly desired. For the current SoC design, a widespread
eral prevailing wireless technologies, but also integrate the fact is that Moores law slows down when the process dimen-
camera(s), audio, battery, display, fingerprint scanner, vibra- sion enters the deep-nanometer regime [12]. Consequently,
tor, gyroscope, wireless charging, etc. In the near future, the speed of energy efficiency improvement is moderated.
there will be a stronger need to enable or enhance vari- Before any proven success with novel IC processes based
ous functions and technologies, such as virtual reality, aug- on new materials, the contemporary silicon and III-IV com-
mented reality, internet of vehicles [8], and so on, all of pound based semiconductor processes, such as complemen-
which further increases the difficulty level of UE design. tary metal-oxide-semiconductor (CMOS), CMOS silicon on
In this section, the major design methods and constraints are insulator (SOI), fin field effect transistor (FinFET), sili-
discussed from several aspects, namely battery, circuit and con germanium (SiGe), gallium arsenide (GaAs), gallium
system, antenna and product design, and other system design nitride (GaN) and indium phosphide (InP), still play a domi-
trade-offs. nant and critical role in the future 5G SoC designs.
Likewise, the multi-layer board design of a 5G mobile
A. BATTERY DESIGN CONSTRAINTS handset will become more compact and integrated to accom-
As plotted in Fig. 1, the data rates of both WiFi and cellular modate an increasing number of SoC chipsets for enabling
increase by around 10 folds for every five years. On the various functions, standards, and technologies. On the main
other hand, during the last 20 years, the battery technique logic board (MLB) of a mobile handset as depicted in Fig. 2,
for mobile devices has been through three major technical there are cellular/WiFi RF transceivers, antenna switch mod-
transitions, which starts with nickel-cadmium (NiCd) battery, ules, power amplifier (PA) modules, baseband (BB) modem,
then the nickel-metal hydride (NiMH) battery, and eventu- NFC, bluetooth, GNSS, application processor (AP), PA man-
ally the current mainstream lithium-ion (Li-ion) battery [9]. agement unit, static random-access memory (SRAM), power
From 1995 to 2014, the wireless capacity has increased by management unit (PMU), etc. Nowadays, these highly cus-
around 10,000 times [10], whereas only 4-5 folds increment tomized chipsets are supplied by various vendors who design
of battery specific power have been achieved for the same and fabricate them with different processes.
period. Apparently, this mismatch becomes one of the current Similar to the trend in IC design, the footprint of PCB is
bottlenecks for mobile handset devices and affects the quality continuously downsizing to smaller trace width and trace
of user experience. Despite the recent battery research on spacing. As a result, more chipsets can be embedded on
new anode materials [11], before the advent of significant one single main logic board, which results in less insertion
breakthrough in battery performance and feasibility for mass loss (IL) and easier impedance matching. Therefore, the

13994 VOLUME 5, 2017

Y. Huo et al.: 5G Cellular User Equipment: From Theory to Practical Hardware Design

TABLE 1. Dimension information of smartphones.

aesthetic appearance. The slim form factor improves the user

experience, and can be seen in several recent mainstream
smartphones as illustrated in Table 1.
The antenna dimension is proportional to the effective
wavelength, and this relation can be approximated as
c0
e = (3)
FIGURE 2. An example of main logic board in contemporary smartphones. f e
where c0 is the speed of light in vacuum, f is the frequency,
and e is the effective dielectric constant that makes the effec-
RF front-end loss caused by IL and impedance mismatching tive wavelength shorter. Furthermore, the effective dielectric
are reduced, and the receiver (RX) sensitivity and transmit- constant can be derived using the following equations [16]
ter (TX) power can be improved. On the other hand, signal r + 1 r 1
integrity is an issue in a more complicated MLB design. For e = +
2 2
example, the clock signals and their harmonics, through com-


 2
plicated signal path and modulation, can end up at the receiver
1 W
q
+ 0.04 1 H ,
end in the form of spur. Therefore, a microwave shield cover
1 + 12 WH
is normally used on MLB to improve the electromagnetic
compatibility/electromagnetic interference (EMC/EMI) per- subject to W /H < 1 (4)
formance. Another frequently seen issue is the degradation r + 1 r 1
e = + q
of sensitivity, or desense, which is typically caused by 2 2 1 + 12 H


W
TX output leaked into the RX path due to insufficient isola-
subject to W /H 1 (5)
tion between TX and RX ports. This issue is more serious
in the case of carrier aggregation, for example, when the where r stands for the relative dielectric constant, W is
harmonic of a transmit signal falls in the receive band of a the width of antenna, and H is the thickness of the antenna
paired CA band [13]. As can be predicted, these issues will substrate. Therefore, the antenna dimension is mainly deter-
become more prevailing in a 5G terminal device. mined by the frequency and substrate material. Although
higher dielectric constant reduces the antenna dimension, it
C. ANTENNA AND PRODUCT DESIGN degrades the antenna performance as more radiation energy
Antenna design is another matter of importance in wireless will be confined inside the substrate instead of being radiated.
systems. Unlike any of its priors [14] in the 2G/3G era, current In the 5G era, the handset antenna design faces more
mobile handset antennas are expected to support not only challenges in order to cover the legacy 3GPP standards and
multi-bands and multi-standards in a wide range of frequency new 5G standards which regulate the use of higher GHz
from 700 MHz to 6 GHz (with some uncovered gaps), but also frequency bands. From this point of view, implementing large
enable certain degrees of diversity and SM. At the same time, scale MIMO at low GHz frequencies becomes very difficult
there is requirement that high efficiency and low specific as it normally requires a minimum spacing to guarantee good
absorption rate (SAR) are both fulfilled after assembling isolation. As for the mmWave antenna design, more antenna
antennas into the handset housing made of metallic casing. elements can be accommodated thanks to downsizing, but the
Therefore, the co-design of antennas, metal casing, and metal casing can deteriorate the antenna performance.
handset housing is enormously challenging since the latter
two factors could generate substantial effects on antenna per- D. SYSTEM DESIGN TRADE-OFFS
formance [15]. Narrow frame and metallic casing are still the In addition to the aforementioned three major design consid-
unswerving trends currently and in the near future, because erations, there are also high-level design constraints between
they enable better protection, portability, heat dissipation and the wireless subsystem and other UE components. Besides

VOLUME 5, 2017 13995

 Y. Huo et al.: 5G Cellular User Equipment: From Theory to Practical Hardware Design

the power budget and hardware area allocation, one more design trade-offs must be considered for a high-performance
critical technical challenge originates from the interference 5G UE. By taking the cellular standard as an example, the
among different components. For example, the display screen figure-of-merit (FOM) of a cellular UE can be formulated as
can cause the RF sensitivity degradation. Therefore, a sheet
of metallic microwave (MW) shield is normally put between Pn,max PDLTBand,n
Non-CA
Pm,max PDLTBands
CC,m
n=1 Beff,n Pn + m=2 Beff,m Pm
the display unit and hardware part to enhance the isolation as FOMCellular,UE =
shown in Fig. 3 which briefly depicts a cell phone opened VUE MUE
from middle. Moreover, this MW shield can minimize the (6)
SAR in the common use cases when the screen side is held
close to the head of a smartphone user, as illustrated in Fig. 4. where PDLTBand,n
non-CA is the PDLT of the 3GPP band n when
carrier aggregation is not enabled (non-CA). Beff,n and Pn
stand for the effective bandwidth and power consumption
respectively, when the wireless UE works in the non-CA
mode. Moreover, the effective bandwidth is the bandwidth of
the used band which has excluded the guard band. Accord-
ingly, PDLTBands
CC,m represents the PDLT of the carrier aggrega-
tion of m CCs, and the superscript m, max is the maximum
number of CCs, defined to be up to 5 in 3GPP Release 13.
Thus, the first and second terms of the numerator add up
the energy-spectral efficiency of both non-CA and CA cases
for all cellular bands and CA combinations supported by the
wireless UE, then we divide the result by the volume VUE and
weight MUE of the wireless UE. The denominator reflects the
score of electrical-mechanical co-design and the portability
of the wireless UE. Therefore, the unit of FOMCellular,UE is
bit/Hz/Joule/mm3 /gram. It is obvious that more bands and
CCs, higher SE, smaller volume and weight, can result in a
FIGURE 3. Disassembly of a smartphone to front and back parts.
higher FOM of wireless UE, which means a better compre-
hensive design. It is worth mentioning that the material cost
of the UE is not taken into consideration in the equation for
well-defined comparison.

III. 5G CELLULAR UE BASED ON A NOVEL

SYSTEM ARCHITECTURE
The foremost challenge of using high GHz frequency
bands comes from the propagation loss that is significantly
higher and more complicated than sub-6 GHz frequency
bands. Based on the most recently published 5G channel
model [17], [18], plus atmospheric absorption and rain atten-
uation models in [19], [20], the path loss comparison for
different propagation scenarios are given in Table 2 for three
frequency bands, namely 2.6 GHz, 28 GHz, and 39 GHz.
FIGURE 4. Talk mode using a specific anthropomorphic mannequin (SAM)
head phantom.
A. CHANNEL MODEL ANALYSIS
In other words, antennas radiate minimal signal through As presented in Table 2, the path loss of non-line-of-
the screen, and therefore it can only propagate the signal in sight (NLOS) is much larger than that of line-of-sight (LOS),
the direction away from the human head. Nevertheless, the and LOS of the urban macrocell (UMa) scenario has similar
shield increases the thickness of the handset and degrades the path loss to the LOS urban microcell (UMi) scenario. How-
form factor. The placements of camera, speaker, finger scan- ever, the path loss in UMi Street Canyon NLOS is much more
ner, battery, MLB, also require careful consideration as they severe than UMa NLOS or UMi Street Open. In addition, for
can change the electromagnetic field and lead to undesired all scenarios, the path loss of 28 and 39 GHz are at least 20 dB
effects. To summarize here, contemporary wireless UEs need larger than LTE band 41. In order to combat such large path
to provide high quality of user experience determined and loss, the FCC regulation allows base station (BS) to transmit
contributed by comprehensive factors which not only lie in at 75 dBm per 100 MHz [7]. Moreover, the power loss caused
the wireless system design, but also mechanical design, prod- by oxygen absorption and rain attenuation are comparatively
uct design, operating system design, etc. Consequently, many small.

13996 VOLUME 5, 2017

Y. Huo et al.: 5G Cellular User Equipment: From Theory to Practical Hardware Design

TABLE 2. Calculation and comparison of path loss.

Apart from the path loss and atmospheric or rain attenua- multiplexing are quite different. It is not feasible and practical
tion loss, the building penetration loss depends on different to make an MN-element antenna array which is designed
materials. Particularly for the concrete wall, the penetration and optimized for beamforming to be perfectly used as MN
loss significantly increases with frequency [18], and it can be MIMO antennas and vice versa.
as high as 117 dB for 28 GHz. In light of these challenges, Consequently, at the UE end, a new system architecture
beamforming is mandatory at both the BS and UE end. needs to be proposed to enable the functionality including
both beamforming and spatial multiplexing, for different
B. NOVEL DISTRIBUTED PHASED ARRAY application scenarios.
BASED MIMO ARCHITECTURE Assuming the appearance of 5G mobile phone is similar to
Implementing mmWave beamforming at the UE end is more the emerging 4G ones, for example, the anticipated volume is
difficult than at the BS end since it is largely constrained by
high energy efficiency requirement, limitations in battery life
and hardware dimension which are key FOM contributors.
The conventional concept of BF is a method to increase the
signal-to-noise-ratio (SNR) and reduce channel interference
of a single data stream, but does not provide spatial multi-
plexing gain by delivering multiple streams.
For the antenna array design of a BF module, the spacing
between antenna elements is critically important because
it is directly relevant to the grating lobe when the array
operates the beam steering. A grating lobe, or side lobe is
undesired in the antenna array because it degrades the gain
and radiation efficiency of an array [22]. Furthermore, for a
given maximum amount of beam steering max , the spacing d
between two neighboring antenna elements has to be main-
tained as [22]
d 1
, (7)
0 1 + cosmax
where 0 is the free space wavelength. Furthermore, a maxi-
mum spacing of 0 /2 is normally used to avoid grating lobes.
On the other hand, in a MIMO system, enough separation
between two antenna elements needs to be maintained so
that a good spatial multiplexing gain can be obtained. The
adjacent element spacing depends on the specific antenna
array design and use case. In order to avoid significant capac-
ity degradation, adjacent element spacing should be larger
than 1.50 for the uniform square array (USA) [23].
Therefore, the design principles for antenna arrays used
FIGURE 5. Proposed DPA-MIMO architecture in mobile phone handset
for beamforming and MIMO antennas used for spatial from back side transparent view.

VOLUME 5, 2017 13997

 Y. Huo et al.: 5G Cellular User Equipment: From Theory to Practical Hardware Design

around 160808 mm, and the SIM card slot is removed due 4) Based on the design methodologies and consideration
to the use of embedded subscriber identity module (eSIM) discussed above, placing BF modules at top two cor-
or virtual SIM that can be integrated into a chipset. To be ners, bottom two corners, and the central part of the
more specific, eSIM is designed and implemented by the UE mobile device is mandatory in order to overcome the
manufacturer, and it removes the need of physical SIM card human body (hand) blockage issue. Therefore, NBF is
while easing the process of switching service provider [21]. flexible but has to be more than 5 as long as it satisfies
It makes the UE device neat and user-friendly by getting rid the minimum isolation spacing. It is worth mentioning
of the physical SIM card tray, which is critical to not only the that, efficient adaptive beam tracking algorithms need
5G UE but also the internet of things devices such as wearable to be employed for both BS and UE ends so that the
devices. two beams from BS and UE can be precisely aligned
A proposed 5G prototyping hardware design is illustrated with acceptable latency.
in Fig. 5 where eight identical 8-element phased array based 5) From the wireless hardware design perspective, the
BF modules are distributed and placed in the back hous- distributed phased arrays based architecture can
ing of a mobile handset. This new architecture is referred help heat dissipation which is largely contributed
to as the distributed phased arrays based MIMO (DPA- by the PAs. In the state-of-the-art PA design for
MIMO) architecture, several advantages can be observed as 5G phased arrays, the power added efficiency (PAE) is
follows. below 20% [27]. Therefore the majority of the
1) Each BF module, embedding one RF transceiver DC power will be converted into the thermal energy
chain, realizes an active phased array of NANT (=8 in which increases the inner temperature of a mobile
this example) antenna elements. Now we can use handset and potentially leads to a critical failure of
an example to estimate the effective isotropic radi- the entire system. This issue is more pronounced when
ated power (EIRP). Assume that a PA has an output multiple mmWave PAs are integrated in the BF mod-
power of PPA , and the output power is split into NANT ules and the handset is operated at cell edge or with
equal parts and fed into a NANT -element phased array. heavy traffic load. By arranging the mmWave BF mod-
At the phased array output, we can obtain an EIRP ules in a distributed manner, it can largely mitigate
which is 10log10 (NANT ) (=9) dB higher than PPA . this self-heating issue. Otherwise a cooling device is
Furthermore, if we remove that PA and place one PA required [28], however it is difficult to implement in a
in the front-end path of each antenna element, and the compact mobile handset.
output power of all PAs is maintained at PPA (not scaled
down with the increased number of front-ends [6]), the C. BEAMFORMING MODULE HARDWARE DESIGN
EIRP can be boosted to be 20log10 (NANT ) (=18) dB The details of a BF module design are given in Fig. 7.
higher than PPA . Therefore, in this scenario, First, the antenna array top-down view is shown in Fig. 7(a).
NPA (= NANT ) PAs contribute the extra 10log10 (NANT ) Second, the layer on which the chipsets are mounted is
(=9) dB gain on top of the first EIRP, but at the cost of shown in Fig. 7(b). A type of material with low dielectric
higher power consumption. constant and small loss tangent is desired. As a matter of fact,
2) A total number of NBF (=8 in this example) there are several suitable integration technology candidates
BF modules, with enough spacing and isolation, can such as low temperature co-fired ceramics (LTCC), hybrid
also cooperate as 8 MIMO antennas to process a maxi- LTCC [29], multi-layer organics (MLO) [30], liquid crystal
mum number of 8 streams. Thus, the spatial multiplex- polymer (LCP) [31], etc. Considering the cost, mass produc-
ing gain can be obtained to further increase the link tion and industrial maturity [32], an MLO-like structure is
throughput by multiple times. adopted as it has shown profound value on commercial mass
3) The DPA-MIMO topology provides a solution to production in IEEE 802.11ad products [6].
human body blockage which could lead to severe In the cross section view of the BF module in Fig. 7(c),
attenuation at mmWave frequencies [24]. For example, Rogers RO4003C material is used for both antenna and
the attenuation can be as high as 30 to 40 dB for PCB substrate because it has a low loss tangent and a suitable
the 73 GHz band as given in [25]. According to the dielectric constant at the high GHz frequency. Accordingly,
study of mobile phone user habits [26], the mobile the effective wavelength e for the 28 GHz carrier can be cal-
handset is usually held in several popular positions as culated using (3)-(5). Moreover, the spacing among antenna
depicted in Fig. 6. In position (i) and (ii), thanks to the elements is set to 0 /2, thus W and L of the BF module
DPA-MIMO architecture, BF modules 1 and 2 can still are calculated as 25 and 18 mm, respectively, with some
work normally, either independently or cooperatively dimension margin.
in a 2 2 MIMO SM mode; in case (iii), BF modules Each BF module is connected with the MLB by coaxial
3-6 can work either independently or cooperatively in cables and coaxial connectors on PCBs, and BF modules
a 4 4 MIMO SM mode. Finally, in position (iv), all are arranged in the back housing of the mobile devices.
BF modules can work simultaneously and support a There are several critical factors in BF module arrangements.
8 8 MIMO SM mode. First, the spacing among BF modules, dISOL , should be kept

13998 VOLUME 5, 2017

Y. Huo et al.: 5G Cellular User Equipment: From Theory to Practical Hardware Design

FIGURE 6. Four popular positions of holding mobile phone handsets.

FIGURE 7. (a) Top-down view of the BF module. (b) Chipsets layer of the BF module. (c) Layout and cross-section view of the
BF module stack-up.

sufficiently large ( 1.50 , or 16 mm). Second, as long A 5G BF module accommodates multiple PAs in a very
as a good spacing is guaranteed, more BF modules can be small space. Therefore, a poor thermal design may cause its
embedded on the back housing of the mobile device thus temperature to rise to a threshold value which results in a
higher order MIMO can be obtained. However, this involves total failure of the BF module. Moreover, an overheated BF
trade-offs between wireless performance and limited hard- module can cause the entire UE system malfunction, and in
ware area or resource on mobile devices. These constraints some extreme cases, it can even compromise the users safety.
are much alleviated on tablet computers. The mmWave front Normally, more than one IC process is used to implement
end, control and calibration circuits can be designed and the mmWave RF transceiver chipsets, and therefore, more
fabricated using various conventional IC processes according than one die is used and shown in Fig. 7(c). With the current
to different design specifications and features. Connection mainstream IC processes, based on the emerging commercial
between the PAs, LNAs and antenna elements is built by products and the existing IC design experience, the total area
the stud bumps and the antenna feed lines routed inside the of chipsets can be well managed below 1010 mm according
package. The through package vias (TPVs) route the signals to mmWave IC designs in [6], [30] and [33]. Moreover, the
between the dies and the PCB, and they also dissipate the thickness of the BF module can be briefly calculated using
heat which is mainly generated from the PA dies in the BF the equation as below
module. The thermal design and TPVs design are very critical
because they are directly relevant to the cooling performance.
HBF = HANT + Hbump + Hdie + HPCB + Hconnector (8)

VOLUME 5, 2017 13999

 Y. Huo et al.: 5G Cellular User Equipment: From Theory to Practical Hardware Design

FIGURE 8. Block diagram of 5G user equipment wireless system architecture.

where HANT is the thickness of the mmWave patch antenna than the FR-4 laminate widely used for the MLB design
with an estimated thickness of 0.4 mm, Hbump is the stand in contemporary smartphones. Therefore, separating
bump thickness with a typical value of 50m, the die thick- BF modules and MLB design leads to cost effective
ness Hdie is usually 254m, the thickness of PCB, HPCB , is manufacturing for mass production.
less than 0.4 mm, and the flat coaxial connector (widely used Third, converting the mmWave frequency to the
in commercial products) has a thickness of only 0.4 mm. IF frequency directly and immediately on a BF module
Therefore, the total thickness including the surface mounted minimizes the front-end insertion loss. Moreover, better
coaxial connector can be made below 1.5 mm, and conse- signal integrity can be achieved since the connection
quently, a good form factor of 5G UE can be well maintained. is through coaxial cables instead of routing traces on
Moreover, the patch antenna design can be flexible and the MLB.
tailored to specific requirement. For example, it can be a Fourth, it offers the flexibility to handle various applica-
rectangular/circular microstrip patch, slot loop, Yagi-Uda, tions and scenarios without the need to reconfigure the
planar inverted-F, substrate integrated waveguide, etc. entire wireless system design. For example, the place-
As long as they fit into the BF module to construct a phased ment and number of BF modules in the handset can be
array, they can be used in our DPA-MIMO architecture. adjusted according to different system specifications and
The frequency band can be 28, 37, or 39 GHz for licensed use cases, which makes it cost-effective.
5G cellular networks. Last, the low GHz IF radio is less challenging to imple-
ment than its mmWave counterpart, and therefore, it
D. RF CIRCUIT DESIGN OF 5G CELLULAR UE can be designed and manufactured in the same IC pro-
Another important feature in the proposed RF circuit design cesses and SoCs, for legacy cellular standards such as
is a split-IF architecture whose block diagram is depicted in 3G and 4G. In addition, it can facilitate the test in
Fig. 8. The BF module not only integrates the active antenna mass production for both BF modules and IF radio plus
array to realize beamforming, but also enables frequency baseband modules [6].
conversion for both uplink and downlink. As shown in Fig. 8 Furthermore, a frequency plan is proposed as follows. Take
the BF module down-converts a high GHz 5G band signal to the 28 GHz frequency band as an example. As shown in
a low GHz intermediate frequency (IF) signal in the downlink Fig. 9, the IF frequency is set to 4.4 GHz, the radio control
path, and up-converts an IF signal to a 5G band signal in the signal is at 600 MHz, the reference clock signal is set to
uplink path. The local oscillator (LO) signal fLO1 can be tuned below 100 MHz, and power supply is a DC signal. These
and realize the frequency conversion for a 28 GHz frequency signals are all supplied over the coaxial cable, and separated
band 5G TDD signal. Furthermore, this RF architecture can or combined using quadplexers on both the BF modules
apply to 37 GHz, 39 GHz, and other 5G frequency bands and IF radio ends. The IF frequency is chosen at 4.4 GHz
after changing the LO frequency and corresponding hardware due to several reasons. First, it does not fall in any LTE
characteristics. band of 3GPP Rel. 14, neither any WiFi/GNSS frequency
The motivation and benefits of utilizing this RF architec- so that the desense issue of LTE/WiFi can be mitigated;
ture lie in several aspects: second, when conducting frequency up-conversion, its image
First, the challenging requirement for the slim form frequency can be easily filtered out since the image frequency
factor of a contemporary handset design strictly limits is separated from the desired frequency by 8.8 GHz; third,
the hardware dimension, and therefore it is not feasible the IF frequency is in the low GHz range and therefore using
to embed all BF modules on a larger MLB. cost-effective coaxial cables can satisfy the performance
Second, the high-performance mmWave circuits design requirement.
necessitates mmWave-enabled PCB such as Rogers In addition, the control signal is chosen to be operated
RO4003C which is more costly but electrically less lossy at 600 MHz. As shown in Fig. 9, the harmonics of the

14000 VOLUME 5, 2017

Y. Huo et al.: 5G Cellular User Equipment: From Theory to Practical Hardware Design

FIGURE 9. 5G user equipment wireless system frequency plan.

control signal do not interfere with the IF radio signal, and

hence the desense issue can be mitigated. A control interface
provides supervision and operation of BF modules through
read-write to the registers using radio control signals such
as RF front end (RFFE) control interface signals. The RFFE
signals carry the information of transmitter signal strength
indicator (TSSI), receiver signal strength indicator (RSSI),
and it executes the calibration and temperature control of a
BF module.
Assume the 28 GHz band contains four sub-bands with
each one occupying a bandwidth of 200 MHz and some guard
bands. The BF module can support a wide bandwidth up to FIGURE 10. Block diagram of 5G beamforming module supporting
800 MHz which is equivalent to four intra sub-bands aggre- FDD duplex scheme.
gated. The local oscillator in the BF module should cover a
frequency range at least from 23.2 to 23.9 GHz. Therefore,
a 4.4 GHz IF radio signal can be obtained after doing fre- E. ADVANCEMENT OF DATA CONVERTER TECHNIQUES
quency down-conversion, and a 28 GHz 5G mmWave signal With respect to the ADCs and digital-to-analog convert-
can be generated after conducting frequency up-conversion. ers (DACs), they should support a wide RF bandwidth
Totally, NBF BF modules and NBF IF radios are integrated with high resolution which depends on the actual applica-
to support a maximum of NBF streams communicating tion, for example the order of digital modulation and the
simultaneously. performance of VGA. In this proposed 5G cellular UE,
As to the detailed BF module implementations, PMUs 256-QAM is supported, which needs a resolution of 12 bits
and low drop-out regulators (LDOs) in each 5G BF module or above. Furthermore, a high spur free dynamic range of
transform the DC voltage of the coax cable to different power ADC needs to be maintained considering that the input signal
supplies for different dies. At the RF front end, each antenna at the receiver end can range from around 25 dBm to
element is connected to one single port double throw (SPDT) 110 dBm [34]. The high dynamic range requirement of the
which is controlled by a radio control signal to enable time wideband VGA can be alleviated by using high-performance
domain duplexing. PAs and low noise amplifiers (LNAs) data converters.
are respectively placed in the uplink and downlink paths There has been a concern of the high power consump-
followed by digitally controlled phase shifters (DCPS) which tion and poor cost effectiveness of ADC for 5G appli-
determine the step resolution of beam steering. For the cations [35]. In fact, several state-of-the-art designs have
IF radio module design, the direct conversion RF architecture recently demonstrated satisfying performance such as low
is employed. Moreover, the differential to single- power consumption and small chip area. In [36], a 12-bit,
ended (D-to-S) amplifier and the single-ended to differential 1.6 GS/s time interleaved ADC only consumes 37.7 mW
(S-to-D) converter are situated in the uplink and downlink with 0.9 mm2 chip area, and it achieves 17.8 fJ/conversion.
paths respectively. The variable gain amplifier (VGA) real- In other words, such ADC can enable a theoretical absolute
izes the function of automatic gain control (AGC) so that physical data throughput of 10.8 Gb/s for an ideal 256-QAM
the dynamic range (DR) requirement of the analog-to-digital demodulation. Moreover, [37][39] have presented high-
converter (ADC) can be mitigated. On the other hand, for performance, energy and area efficient ADC and DAC which
a frequency division duplex (FDD) based 5G beamforming can be considered as good prototype candidates for future
module, the SPDT and filters in the TDD mode are replaced 5G UE data converters. Schreier figure of merit (FOMS ) and
with the diplexers as depicted in Fig. 10. Walden figure of merit (FOMW ) [40] are commonly used

VOLUME 5, 2017 14001

 Y. Huo et al.: 5G Cellular User Equipment: From Theory to Practical Hardware Design

to evaluate the data converters performance, as expressed

below
FOMS = SNDR + 10log10 (B/P) (9)
FOMW = P/(2ENOB min(2B, fs )), (10)
where SNDR is the signal-to-noise-and-distortion ratio,
ENOB stands for the effective number of bits, B is the
analog bandwidth, P is the power consumption, and fS is
the sampling rate. Schreier FOM is also known as ther-
mal FOM and more frequently used when the SNDR is
above 60 dB. According to the data collected [41] which has
summarized the performance of state-of-the-art ADCs pub-
lished in the International Solid-State Circuits Conference
(ISSCC) and the Symposia on VLSI Technology and Circuits
(VLSI Symposia) in recent 20 years, there are a couple
of designs suitable for 200 MHz wide or even 800 MHz
wide analog frequency, with a FOMW smaller than 50
fJ/conversion. Moreover, the FOM of data converters
keeps improving at a steady pace which will further facilitate
the 5G UE hardware design.

IV. LINK BUDGET CALCULATION AND WIRELESS

PERFORMANCE EVALUATION
In this section, link budget analysis and throughput esti-
mation are conducted for the downlink and uplink of the
proposed DPA-MIMO UE design. The 5G channel model
in [17] is used, and the numbers used for insertion loss,
noise figure (NF) and antenna gain are explained first. The
performance of a mmWave antenna switch in [42] shows that
IL can be well managed below 1.9 dB with a TX-RX isolation
better than 38 dB. On the other hand, according to the state-
of-the-art band pass filter (BPF) design [43], the IL is below
FIGURE 11. DPA-MIMO system in a mobile phone handset when
1.5 dB. Therefore, in Table 3 and Table 4, the RX front-end NANT = 16.
loss before the LNA is set to 4.0 dB including extra loss due
to the interface between the LNA and the antenna elements.
The noise figure of the mmWave receiver varies with different
IC processes, and for the state-of-the-art CMOS designs in [6] calculation. Typically, the maximum radius of a microcell
and [46], NF is 7.1 dB and 8 dB, respectively. For the design is 200 meters, and a macrocell BS can cover up to more
using more advanced IC process such as SiGe, BiCMOS, than 1 km. Therefore, the UMa NLOS model is used to
NF of a receiver can achieve 6.8 dB [30] and 5.5 dB [47]. represent communication distance more than 200 meters.
Considering the superior cost-effectiveness of the CMOS As can be analyzed from the results in Table 3, in some
process and its widespread use, it is reasonable to assume that cases, SE becomes smaller as the SNR decreases because
5G receiver NF is around 7 dB. Furthermore, the calculated when the SNR goes down below some threshold value, a
antenna gain of one single patch antenna element can achieve lower digital modulation order is enabled. Furthermore, two
5 to 7 dBi in state-of-the-art designs [28], [29]. Therefore sets of data are given in Table 3 based on 8 and 16 antenna ele-
in the following tables, single antenna element gain is set ments per BF module respectively. The 16 antenna elements
to 5 dBi. based UE architecture is drawn in Fig. 11 which shows that
a sufficiently large physical separation (> 20 in Fig. 11)
A. DOWNLINK BUDGET AND DATA is well maintained. By using more antenna elements, it can
THROUGHPUT ANALYSIS increase the receiver gain, boost the SNR and EIRP so that
The downlink budget calculation under several popular the mobile handset can operate in a more challenging envi-
deployment scenarios is given in Table 3, also with the ronment and handle larger path loss and penetration loss of
results from numerical analysis for the data throughput. The buildings, particularly for mmWave frequency bands.
UMi Street Canyon NLOS and UMa NLOS scenarios in In Fig. 12, the PDLT under several deployment scenarios
the new 5G channel model [17] have been chosen for their is given for 28 GHz and 39 GHz respectively. The num-
larger path loss and shadowing coefficients as the worst-case ber of antenna elements in one UE, denoted as Nue (equals

14002 VOLUME 5, 2017

Y. Huo et al.: 5G Cellular User Equipment: From Theory to Practical Hardware Design

TABLE 3. Calculation and comparison of downlink budget at 28 GHz.

FIGURE 12. 8 8 MIMO, BW=200 MHz, Nue = 8 8, peak downlink throughput versus various
deployment scenarios.

NANT NBF ), is set to 8 8 (=64). The bandwidth is set glass presented in [18] are also added into the propagation
to 200 MHz, and the UE is configured in MIMO with a loss models for analysis. Note that, the results under UMi
maximum of NBF (equals 8 in Fig. 12) layers. The penetration scenarios are not drawn because they are all above 10 Gbps
loss models of regular glass and infrared reflective (IRR) or around it. As can be observed, IRR glass that is widely

VOLUME 5, 2017 14003

 Y. Huo et al.: 5G Cellular User Equipment: From Theory to Practical Hardware Design

TABLE 4. Calculation and comparison of uplink budget at 28 GHz.

used in energy-saving buildings can significantly lower the

PDLT and shorten communication distance. Moreover, if the
human body blockage model is taken into consideration, the
SNR will decrease by 30 to 40 dB and thus PDLT will be
significantly lowered. As explained and illustrated in the
previous sections, the DPA-MIMO architecture can mitigate
the human body blockage issue by enabling the BF modules
not blocked in working modes.

B. UPLINK BUDGET AND DATA THROUGHPUT ANALYSIS

For the uplink, the link budget calculation is given in Table 4.
The maximum EIRP is regulated to be 43 dBm for mobile
stations (MSs) by the FCC [7]. Such level of transmission
power is translated to a substantial challenge for long distance
transmission at 5G mmWave frequency bands. Therefore
there is necessity for the BS to enable the use of large antenna
arrays at the receiver end to compensate for the propagation
loss. Since the UE can operate with NBF NBF MIMO to
receive NBF streams simultaneously, the BS needs to enable
NBF arrays of antenna elements as well, where each array of
antenna elements is referred to as a base station unit. The FIGURE 13. Communication between user equipment and base station in
8 8 MIMO mode.
number of antenna elements in each base station unit, Narray ,
is determined by the link budget calculation and constrained the base station, Nbs is equal to NBF Narray . Assuming NBF
by the hardware resource and implementation feasibility on is 8, two sets of data are given in Table 4, based on 64 and 256
the BS end. Thus, the total number of antenna elements on antenna elements in one antenna array of the BS respectively.

14004 VOLUME 5, 2017

Y. Huo et al.: 5G Cellular User Equipment: From Theory to Practical Hardware Design

FIGURE 14. 28 GHz, 8 8 MIMO, BW=200 MHz, peak uplink throughput versus various deployment scenarios.

FIGURE 15. Regular glass, UMa NLOS, 8 8 MIMO, BW=200 MHz, peak data throughput versus distance for
various number of antenna elements on BS and UE ends.

Fig. 13 illustrates the communication between the and different numbers of antenna units on the BS end.
DPA-MIMO architecture based UE and the BS in 88 MIMO It shows that, a large number of receiver antenna elements
to deliver 8 streams simultaneously. When Narray is 256 and in the antenna array need to be enabled at the BS to com-
the carrier frequency is 28 GHz, the dimension of the 256- pensate for the large propagation loss. Again, the IRR glass
antenna elements array is approximately 100100 mm. In the induced attenuation largely degrades the uplink performance,
given example, there are totally 8 such BS units. It is feasible and therefore more antenna elements should be used at the
to embed 8 or even more groups of such mmWave antenna base station to enable stronger beamforming gain. Also, at the
arrays for practical hardware design of microcells and macro- mobile station, a maximum output of 43 dBm EIRP will limit
cells. As a matter of fact, the antenna array dimension can be the maximum output power of one single PA in the phased
further expanded to 200 200 mm so as to embed a 1024- array, and the relation can be expressed in the equation below:
antenna elements array, thereby increasing the SNR by an
EIRPUE,max = PPA,out + 20log10 (NANT ) (11)
additional 6 dB.
Furthermore, Fig. 14 plots the peak uplink throughput where PPA,out is the output power of one single PA, and NANT
(PULT) versus distance for various deployment scenarios stands for the number of antenna elements in the BF module.

VOLUME 5, 2017 14005

 Y. Huo et al.: 5G Cellular User Equipment: From Theory to Practical Hardware Design

FIGURE 16. IRR glass, UMa NLOS, 8 8 MIMO, BW=200 MHz, peak data throughput versus distance for
various number of antenna elements on BS and UE ends.

This equation is valid only when each antenna element is

connected to one PA. Therefore, when NANT equals 16, the
maximum PPA,out is limited to 19 dBm, and the maximum
PPA,out will increase to 25 dBm if NANT is set to 8. The
specification of 19 dBm output power is less challenging and
more implementable according to the current state-of-the-art
mmWave PA designs [48], [49].

C. ANALYSIS WITH ATTENUATION MODELS

As previously mentioned, the IRR glass can cause very seri-
ous degradation to the uplink performance. More numerical
results of data throughput under various deployment scenar-
ios are plotted for both downlink and uplink modes with two
types of penetration loss in Fig. 15 and Fig. 16, respectively.
It is observed that the strong attenuation caused by IRR glass
can be overcome at the cost of embedding more antennas
at both the UE and BS ends. Moreover, the deployment of
ultra-dense small cells is considered a necessary mechanism
in mmWave communications to deal with strong path loss and
attenuation [50].
On the other hand, the inter-bands carrier aggregation
(Inter-CA) may be needed for 5G mmWave frequency bands.
For example, 28 GHz, 37/39 GHz band, 64-71 GHz band,
etc., are aggregated to provide even larger bandwidths. In that
situation, the DPA-MIMO architecture can still be adopted
for a practical multi-band, multi-mode 5G UE design. As a
FIGURE 17. An example of DPA-MIMO system used in tablet computer
matter of fact, some multi-band 5G hardware components when NANT = 16 and NBF = 16.
have been presented in the literature and can be used in
the DPA-MIMO architecture. Two dual-band 5G mmWave
antenna prototypes are demonstrated in [51], [52], and a In order to further improve the efficiency and overcome
linear Doherty PA supporting 28 GH, 37 GHz and 39 GHz the high peak-to-average power ratio (PAPR) issue, the power
bands is designed and verified using SiGe process with high supply modulation techniques such as envelope tracking (ET)
PAE demonstrated in [53]. will continue to play a critical role in 5G UE design.

14006 VOLUME 5, 2017

Y. Huo et al.: 5G Cellular User Equipment: From Theory to Practical Hardware Design

TABLE 5. 5G state-of-the-art works performance summary and comparison.

One of the major challenges lies in the stringent signal in recent years. We can observe that most of the previ-
bandwidth requirement of the ET modulator, because the ous works have used conventional beamforming, MIMO,
supply generally tracks the signal envelope continuously, Massive-MIMO, and MU-MIMO as the air interface tech-
which can be many times the bandwidth of the I/Q signal nology. According to the numerical analysis, the novel
components [54]. With respect to the new 5G mmWave band DPA-MIMO architecture not only provides the highest down-
occupying hundreds of MHz bandwidth, this challenge can link throughput at long distances, but also is able to address
be more pronounced. Nevertheless, some recently published the human body blockage issue which is not mentioned or
work [55] has shown a 70% efficient envelope modulator for dealt with in the previous 5G state-of-the-art research works.
an X-band PA with 100-MHz signal bandwidth, which can be
considered as a candidate ET technique for 5G applications. V. CONCLUSION
Finally, in the system-level design, the choice of NANT and In this paper, a system architecture and method for next
NBF can be very flexible according to the practical speci- generation wireless user equipment, or 5G cellular user
fication, performance target, and design constraints so that equipment design has been provided. By analyzing the chal-
the user equipment can handle different environments and lenges of contemporary wireless UE designs and emerging
situations. For example, in Fig. 5 and Fig. 11, NANT is 8 5G mmWave techniques, a novel DPA-MIMO architecture
and 16 respectively while NBF is fixed as 8, and the resulted and design method has been presented to overcome the lim-
wireless performance is summarized in Table 3. On the other itations of conventional MIMO structures. This work has
hand, for a specific UE design case where longer battery life provided a solution to the technical constraints and challenges
and more space are available, such as in a tablet computer as of the mobile handset design, such as human blockage, high
illustrated in Fig. 17, both NANT and NBF can be increased path loss, self-heating issues, which are more pronounced for
(to 16 in Fig. 17) to enable better wireless performance. future 5G cellular user equipment but cannot be solved using
the existing system architectures and methods. Moreover,
through numerical analysis, the proposed architecture has
D. PERFORMANCE COMPARISON been shown to increase the wireless link budget and enhance
WITH STATE-OF-THE-ART the data throughput under different use cases and various
Table 5 summarizes and compares the performance of BS deployment scenarios with highly flexible reconfigura-
5G state-of-the-art works from both industry and academics bility. Furthermore, the DPA-MIMO based wireless UE can

VOLUME 5, 2017 14007

 Y. Huo et al.: 5G Cellular User Equipment: From Theory to Practical Hardware Design

be implemented using the state-of-the-art technologies of [19] T. S. Rappaport et al., Millimeter wave mobile communications for 5G
circuits, antennas and systems. As a result, this architecture cellular: It will work! IEEE Access, vol. 1, pp. 335349, 2013.
[20] Q. Zhao and J. Li, Rain attenuation in millimeter wave ranges, in Proc.
can facilitate a peak throughput of more than 10 Gb/s while Int. Symp. Antennas, Propag. EM Theory, Oct. 2006, pp. 14.
maintaining a slim form factor of mobile terminal devices. [21] L. Li, A. G. Mathias, and B. Juang, Provisioning an embedded subscriber
identity module, WO Patent 2 013 169 484 A1, Nov. 14, 2013.
[22] T. S. Rappaport, R. W. Heath. Jr., R. C. Daniels, and J. N. Murdock,
ACKNOWLEDGEMENT Millimeter Wave Wireless Communications. Upper Saddle River, NJ, USA:
The authors would like to acknowledge Natural Sciences and Prentice-Hall, 2014.
[23] F. Rusek et al., Scaling up MIMO: Opportunities and challenges with
Engineering Research Council of Canada, and National Nat- very large arrays, IEEE Signal Process. Mag., vol. 30, no. 1, pp. 4060,
ural Science Foundation of China for support of this project, Jan. 2013.
[24] K. Zhao, J. Helander, D. Sjoberg, Sa. He, T. Bolin, and Z. Ying, User
Dr. Song Hu from Georgia Institute of Technology and body effect on phased array in user equipment for the 5G mmWave
Dr. Adrian Tang from NASA JPL for valuable discussions. communication system, IEEE Antennas Wireless Propag. Lett., vol. 16,
pp. 864867, 2017.
[25] G. R. MacCartney, S. Deng, S. Sun, and T. S. Rappaport, Millimeter-wave
REFERENCES human blockage at 73 GHz with a simple double knife-edge diffraction
[1] K. Doppler, M. Rinne, C. Wijting, C. B. Ribeiro, and K. Hugl, Device-to- model and extension for directional antennas, in Proc. IEEE 84th Veh.
device communication as an underlay to LTE-advanced networks, IEEE Technol. Conf. (VTC-Fall), Sep. 2016, pp. 16.
Commun. Mag., vol. 47, no. 12, pp. 4249, Dec. 2009. [26] I. Vasilev, V. Plicanic, and B. K. Lau, Impact of antenna design
[2] D. Lopez-Perez, I. Guvenc, G. de la Roche, M. Kountouris, T. Q. S. Quek, on MIMO performance for compact terminals with adaptive
and J. Zhang, Enhanced intercell interference coordination challenges impedance matching, IEEE Trans. Antennas Propag., vol. 64, no. 4,
in heterogeneous networks, IEEE Wireless Commun., vol. 18, no. 3, pp. 14541465, Apr. 2016.
pp. 2230, Jun. 2011. [27] S. Shakib, H. C. Park, J. Dunworth, V. Aparin, and K. Entesari, A 28 GHz
[3] C. Hoymann, D. Larsson, H. Koorapaty, and J.-F. Cheng, A lean carrier efficient linear power amplifier for 5G phased arrays in 28 nm bulk
for LTE, IEEE Commun. Mag., vol. 51, no. 2, pp. 7480, Feb. 2013. CMOS, in IEEE Int. Solid-State Circuits Conf. (ISSCC) Dig. Tech. Papers,
[4] M. Agiwal, A. Roy, and N. Saxena, Next generation 5G wireless net- Jan./Feb. 2016, pp. 352353.
works: A comprehensive survey, IEEE Commun. Surveys Tuts., vol. 18, [28] S. Zihir and G. M. Rebeiz, A 60 GHz 64-element phased-array beam-
no. 3, pp. 16171655, 3rd Quart. 2016. pointing communication system for 5G 100 meter links up to 2 Gbps, in
[5] Cellular Insights, Samsung Galaxy S7. The First 4 4 MIMO IEEE MTT-S Int. Microw. Symp. Dig., May 2016, pp. 13.
Smartphone, accessed on Dec. 2, 2016. [Online]. Available: [29] H. Jin, W. Che, K.-S. Chin, Gu. Shen, W. Yang and Q. Xue,
http://cellularinsights.com/samsung-galaxy-s7-the-first-4x4-mimo- 60-GHz LTCC differential-fed patch antenna array with high gain by
smartphone/ using soft-surface structures, IEEE Trans. Antennas Propag., vol. 65,
[6] M. Boers et al., A 16TX/16RX 60 GHz 802.11ad chipset with single no. 1, pp. 206216, Jan. 2017.
coaxial interface and polarization diversity, IEEE J. Solid-State Circuits, [30] A. Natarajan et al., A fully-integrated 16-element phased-array receiver
vol. 49, no. 12, pp. 30313045, Dec. 2014. in SiGe BiCMOS for 60-GHz communications, IEEE J. Solid-State
[7] Report and Order and Future Notice of Proposed Rulemaking, Circuits, vol. 46, no. 5, pp. 10591075, May 2011.
accessed on Dec. 2, 2016. [Online]. Available: https://apps.fcc.gov/ [31] S. Li, T. Chi, J. S. Park, W. T. Khan, H. Wang, and J. Papapolymerou,
edocs_public/attachmatch/FCC-16-89A1.pdf A fully packaged D-band MIMO transmitter using high-density flip-chip
[8] M.-K. Jiau, S.-C. Huang, J.-N. Hwang, and A. V. Vasilakos, Multimedia interconnects on LCP substrate, in IEEE MTT-S Int. Microw. Symp. Dig.,
services in cloud-based vehicular networks, IEEE Intell. Transp. Syst. May 2016, pp. 14.
Mag., vol. 7, no. 3, pp. 6279, Sep. 2015. [32] L. Dussopt, O. El Bouayadi, Y. Lamy, and G. Simon, Silicon inter-
[9] T. Reddy, Lindens Handbook of Batteries, 4th ed. New York, NY, USA: poser technology and antenna solutions for highly-integrated mmWave
McGraw-Hill, 2010. transceiver modules, in Proc. IEEE MTT-S Int. Microw. Symp. (IMS) ,
[10] M. K. Tsai, Cloud 2.0 clients and connectivityTechnology and chal- May 2016, pp. 136.
lenges, in IEEE Int. Solid-State Circuits Conf. (ISSCC) Dig. Tech. Papers, [33] J. Pang et al., A 128-QAM 60 GHz CMOS transceiver for IEEE 802.11ay
Feb. 2014, pp. 1519. with calibration of LO feedthrough and I/Q imbalance, in IEEE
[11] L. David, R. Bhandavat, U. Barrera, and G. Singh, Silicon oxycarbide Int. Solid-State Circuits Conf. (ISSCC) Dig. Tech. Papers, Feb. 2017,
glass-graphene composite paper electrode for long-cycle lithium-ion bat- pp. 13.
teries, Nature Commun., vol. 7, p. 10998, Mar. 2016. [34] D. Anzaldo. Navigate the AFE and Data-Converter Maze in Mobile
[12] S. Sutardja, Slowing of Moores law signals the beginning of smart Wireless Terminals, accessed on Dec. 20, 2016. [Online]. Available:
everything, in Proc. 44th Eur. Solid State Device Res. Conf., Sep. 2014, https://www.maximintegrated.com/en/appnotes/index.mvp/id/5519
pp. 78. [35] J. Liu, H. Minn, and A. Gatherer. (Jun. 2015). The Death of 5G
[13] L. Miller. Carrier Aggregation Fundamentals For Dummies, accessed Part 2: Will Analog be the Death of Massive MIMO? IEEE ComSoc
on Apr. 20, 2017. http://www.qorvo.com/resources/d/qorvo-carrier- Technology News, CTN Issue, accessed on Dec. 20, 2016. [Online].
aggregation-fundamentals-for-dummies-volume-1 Available: http://www.comsoc.org/ctn/death-5g-part-2-will-analog-be-
[14] C. Rowell and E. Y. Lam, Mobile-phone antenna design, IEEE Antennas death-massive-mimo
Propag. Mag., vol. 54, no. 4, pp. 1434, Aug. 2012. [36] J.-W. Nam, M. Hassanpourghadi, A. Zhang, and M. S.-W. Chen, A 12-bit
[15] K.-L. Wong and C.-Y. Tsai, IFA-based metal-frame antenna with- 1.6 GS/s interleaved SAR ADC with dual reference shifting and interpo-
out ground clearance for the LTE/WWAN operation in the metal- lation achieving 17.8 fJ/conv-step in 65 nm CMOS, in Proc. IEEE Symp.
casing tablet computer, IEEE Trans. Antennas Propag., vol. 64, no. 1, VLSI Circuits, Jun. 2016, pp. 12.
pp. 5360, Jan. 2016. [37] C. Lin, Y. Wei, and T.-C. Lee, A 10b 2.6 GS/s time-interleaved
[16] C. A. Balanis, Antenna Theory: Analysis and Design. Hoboken, NJ, USA: SAR ADC with background timing skew calibration, in IEEE Int.
Wiley, 2005. Solid-State Circuits Conf. (ISSCC) Dig. Tech. Papers, Jan./Feb. 2016,
[17] Aalto University, BUPT, CMCC, Nokia, NTT DOCOMO, New York
pp. 468469.
University, Ericsson, Qualcomm, Huawei, Samsung, Intel, University [38] J. Wu et al., A 4 GS/s 13 b pipelined ADC with capacitor and amplifier
of Bristol, KT Corporation, and University of Southern California, sharing in 16 nm CMOS, in IEEE Int. Solid-State Circuits Conf. (ISSCC)
5G Channel Model for bands up to 100 GHz, 2nd Workshop Dig. Tech. Papers, Jan./Feb. 2016, pp. 466467.
Mobile Commun. Higher Freq. Bands (MCHFB), San Diego, [39] S. Su and M. S.-W. Chen, A 12-Bit 2 GS/s dual-rate hybrid DAC
CA, USA, Tech. Rep. white paper version 1.0, Dec. 2015, with pulse-error pre-distortion and in-band noise cancellation achiev-
accessed on Apr. 22, 2017. [Online]. Available: http://www. ing > 74 dBc SFDR and < 80 dBc IM3 up to 1 GHz in 65 nm
5gworkshops.com/5GCM.html CMOS, IEEE J. Solid-State Circuits, vol. 51, no. 12, pp. 29632978,
[18] K. Haneda et al., 5G 3GPP-like channel models for outdoor urban Dec. 2016.
microcellular and macrocellular environments, in Proc. IEEE 83rd Veh. [40] R. H. Walden, Analog-to-digital converter survey and analysis, IEEE
Technol. Conf. (VTC-Spring), May 2016, pp. 17. J. Sel. Areas Commun., vol. 17, no. 4, pp. 539550, Apr. 1999.

14008 VOLUME 5, 2017

Y. Huo et al.: 5G Cellular User Equipment: From Theory to Practical Hardware Design

[41] B. Murmann. ADC Performance Survey 19972016, accessed on [64] B. Sadhu et al., A 28 GHz 32-element phased-array transceiver IC with
Dec. 20, 2016. [Online]. Available: http://web.stanford.edu/murmann/ concurrent dual polarized beams and 1.4 degree beam-steering resolution
adcsurvey.html for 5G communication, in IEEE Int. Solid-State Circuits Conf. (ISSCC)
[42] C. Byeon and C. S. Park, Design and analysis of the millimeter-wave Dig. Tech. Papers, Feb. 2017, pp. 13.
SPDT switch for TDD applications, IEEE Trans. Microw. Theory Techn., [65] UCSD. (Jan. 2017). TowerJazz and UC San Diego Demonstrate
vol. 61, no. 8, pp. 28582864, Aug. 2013. Best in Class 5G Mobile Transmit-Receive Chips With Greater
[43] Z. Yang and D. Peroulis, A 2040 GHz tunable MEMS band- Than 12 Gbps Data Rates, accessed on 2017. [Online]. Available:
pass filter with enhanced stability by gold-vanadium micro-corrugated http://jacobsschool.ucsd.edu/news/news_releases/release.sfe?id=2137
diaphragms, in IEEE MTT-S Int. Microw. Symp. Dig., May 2016, [66] Huawei. Huawei and DOCOMO Conduct Worlds First 5G Large Scale
pp. 13. Field Trial in the 4.5 GHz Band. accessed on 2017. [Online]. Available:
[44] M. Rumney, LTE and the Evolution to 4G Wireless: Design and Measure- http://www.huawei.com/en/news/2016/11/World-First-5G-Large-Scale-
ment Challenges. Hoboken, NJ, USA: Wiley, 2013. Field-Trial
[45] C. Johnson, Long Term Evolution in Bullets. North Charleston, SC, USA: [67] Ericsson. Ericsson T-Mobile and Ericsson Achieve Over 12
CreateSpace, 2012. Gbps on 5G Connection, accessed on 2017. [Online]. Available:
[46] N. Saito et al., A fully integrated 60-GHz CMOS transceiver chipset based https://www.ericsson.com/news/2043477
on WiGig/IEEE 802.11ad with built-in self calibration for mobile usage, [68] Nokia. SK Telecom and Nokia Networks Achieve 19.1 Gbps Over
IEEE J. Solid-State Circuits, vol. 48, no. 12, pp. 31463159, Dec. 2013. the Air in Joint 5G Trial, accessed on 2017. [Online]. Available:
[47] A. Tomkins et al., A 60 GHz 802.11ad/WiGig-compliant transceiver for http://www.nokia.com/en_int/news/releases/2015/10/30/sk-telecom-and-
infrastructure and mobile applications in 130 nm SiGe BiCMOS, IEEE nokia-networks-achieve-191-gbps-over-the-air-in-joint-5g-trial
J. Solid-State Circuits, vol. 50, no. 10, pp. 22392255, Oct. 2015. [69] Nokia. Video: World-First5G cmWave Technology 19.1
[48] D. Zhao and P. Reynaert, An E-band power amplifier with broadband Gbps Throughput, accessed on 2017. [Online]. Available:
parallel-series power combiner in 40-nm CMOS, IEEE Trans. Microw. https://networks.nokia.com/videos/world-first-nokia-5g-cm-wave-
Theory Techn., vol. 63, no. 2, pp. 683690, Feb. 2015. technology-19-1-gbps-over-the-air
[49] A. Larie, E. Kerherv, B. Martineau, L. Vogt, and D. Belot, A 60 GHz [70] National Instruments. NI Demonstrates Worlds First Real-Time
28 nm UTBB FD-SOI CMOS reconfigurable power amplifier with Overthe-Air Prototype for Verizon 5G at 28 GHz, accessed on
21% PAE, 18.2 dBm P1dB and 74 mW PDC, in IEEE Int. Solid-State 2017. [Online]. Available: http://www.ni.com/newsroom/release/
Circuits Conf. (ISSCC) Dig. Tech. Papers, Feb. 2015, pp. 13. ni-demonstrates-worlds-firstreal-time-over-the-air-prototype-for-verizon-
[50] A. Al-Dulaimi, S. Al-Rubaye, J. Cosmas, and A. Anpalagan, Planning 5g-at-28-ghz/en/
of ultra-dense wireless networks, IEEE Netw., vol. 31, no. 2, pp. 9096,
Mar./Apr. 2017.
[51] M. M. M. Ali and A.-R. Sebak, Dual band (28/38 GHz) CPW slot
directive antenna for future 5G cellular applications, in Proc. IEEE Int.
Symp. Antennas Propag. (APSURSI), Jun. 2016, pp. 399400.
[52] W. Zhai, V. Miraftab, M. Repeta, D. Wessel, and W. Tong, Dual-band
millimeter-wave interleaved antenna array exploiting low-cost PCB tech-
nology for high speed 5G communication, in IEEE MTT-S Int. Microw.
Symp. Dig., May 2016, pp. 13.
[53] S. Hu, F. Wang, and H. Wang, A 28 GHz/37 GHz/39 GHz multiband
linear Doherty power amplifier for 5G massive MIMO applications, in
IEEE Int. Solid-State Circuits Conf. (ISSCC) Dig. Tech. Papers, Feb. 2017,
pp. 13.
[54] Z. Popovic, Amping up the PA for 5G, IEEE Microw. Mag., vol. 18,
no. 3, pp. 137149, May 2017.
[55] D. F. Kimball, H. Kazeimi, J. J. Yan, P. T. Theilmann, I. Telleiz, and YIMING HUO (S08) received the B.Eng.
G. Collins, Envelope modulator & X-band MMICs on highly integrated
degree in information engineering from Southeast
3D tunable microcoax substrate, in Proc. IEEE Compound Semiconduc-
University, China, in 2006, and the M.Sc. degree
tor Integr. Circuit Symp., New Orleans, LA, USA, Oct. 2015, pp. 14.
[56] Samsung. Samsung Electronics Sets 5G Speed Record at 7.5 Gbps, in system-on-chip from Lund University, Sweden,
Over 30 Times Faster Than 4G LTE, accessed on 2017. [Online]. in 2010. He is currently pursuing the Ph.D. degree
Available: http://www.samsung.com/global/business/networks/insights/ in electrical engineering with the University of
news/samsung-electronics-sets-5g-speed-record-at-7-5gbps-over-30- Victoria, Canada. From 2009 to 2010, he stayed
times-faster-than-4g-lte with Ericsson and ST Ericsson, where he accom-
[57] Y. Kim et al., Feasibility of mobile cellular communications at millimeter plished his master thesis on multi-mode, wide-
wave frequency, IEEE J. Sel. Topics Signal Process., vol. 10, no. 3, band CMOS VCOs for cellular systems. Since the
pp. 589599, Apr. 2016. Ph.D. study, he has been involved in the a wide range of cross-disciplinary
[58] T. Obara, Y. Inoue, Y. Aoki, S. Suyama, J. Lee, and Y. Okumura, Experi-
topics of next-generation wireless communication systems design, and his
ment of 28 GHz band 5G super wideband transmission using beamforming
current research interests are in the physical layer design of 5G cellular user
and beam tracking in high mobility environment, in Proc. IEEE PIMRC,
Sep. 2016, pp. 15. equipment, and millimeter-wave circuits and systems.
[59] Qualcomm. (Oct. 2016). Qualcomm Showcases 5G Leadership by Between 2010 and 2011, he was with the Chinese Academy of
Announcing its First 5G Modem Solution, accessed on 2017. [Online]. Sciences (CAS) as Research Associate. From 2011 to 2012, he was with
Available: https://www.qualcomm.com/news/releases/2016/10/17/qual STMicroelectronics as RF Engineer for developing digital video broadcast-
comm-showcases-5g-leadership-announcing-its-first-5g-modem-solution ing systems. From 2015 to 2016, he spent a year conducting RF hardware
[60] Qualcomm. Snapdragon X50 5G Modem, accessed on 2017. and cellular design for various Projects in Apple Inc., Cupertino, CA, USA,
https://www.qualcomm.com/products/snapdragon/modems/5g/x50 as a Ph.D. Intern.
[61] R. Lai. Qualcomm-Powered 5G Devices May Arrive in First Mr. Huo was a recipient of the Best Student Paper Award of the 2016
Half of 2018, accessed on 2017. https://www.engadget.com/2016/
IEEE ICUWB, the Excellent Student Paper Award of the 2014 IEEE ICSICT,
10/17/qualcomm-snapdragon-x50-modem/
[62] Intel. (Jan. 2017). Intel Announces Worlds First Global 5G and the Bronze Leaf certificate of the 2010 IEEE PrimeAsia. He also
Modem, accessed on 2017. [Online]. Available: https://newsroom. received the University of Victoria Fellowship from 2012 to 2013, and
intel.com/newsroom/wp-content/uploads/sites/11/2017/01/5G-modem- ISSCC-STGA from the IEEE Solid-State Circuits Society (SSCS) in 2017.
fact-sheet.pdf He is a member of the SSCS, the CAS, the MTT-S, and the ComSoc.
[63] D. Goovaerts. (Jan. 2017). Intel Unveils 5G Modem With Sup- He has been serving as member of the Program Committee of the IEEE
port for Sub-6 GHz, 28 GHz at CES, accessed on 2017. [Online]. ICUWB 2017, and a technical reviewer for several IEEE conferences and
Available: https://www.wirelessweek.com/news/2017/01/intel-unveils-5g- journals, including the IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY, and
modem-support-sub-6-ghz-28-ghz-ces China Communications.

VOLUME 5, 2017 14009

 Y. Huo et al.: 5G Cellular User Equipment: From Theory to Practical Hardware Design

XIAODAI DONG (S97M00SM09) received WEI XU (S07M09SM15) received the B.Sc.

the B.Sc. degree in information and control engi- degree in electrical engineering and the M.S. and
neering from Xian Jiaotong University, Xian, Ph.D. degrees in communication and information
China, in 1992, the M.Sc. degree in electrical engi- engineering from Southeast University, Nanjing,
neering from the National University of Singapore, China, in 2003, 2006, and 2009, respectively. He is
Singapore, 1995, and the Ph.D. degree in electrical currently a Full Professor with the National Mobile
and computer engineering from Queens Univer- Communications Research Laboratory, Southeast
sity, Kingston, ON, Canada, in 2000, respectively. University. Between 2009 and 2010, he was a Post-
Since 2005, she has been with the University of Doctoral Research Fellow with the Department
Victoria, Victoria, BC, Canada, where she is cur- of Electrical and Computer Engineering, Univer-
rently a Professor and a Canada Research Chair (Tier II) with the Department sity of Victoria, Canada. He has authored over 100 refereed journal and
of Electrical and Computer Engineering. From 2002 to 2004, she was an conference papers in addition to six granted patents. His current research
Assistant Professor with the Department of Electrical and Computer Engi- interests include cooperative communications, information theory, and signal
neering, University of Alberta, Edmonton, AB, Canada. From 1999 to 2002, processing for wireless communications.
she was with Nortel Networks, Ottawa, ON, Canada, and worked on the base Dr. Xu is an Editor of the IEEE COMMUNICATIONS LETTERS. He has been
transceiver design of the third-generation mobile communication systems. involved in technical program committees for many international confer-
Her research interests include mobile communications, radio propagation, ences, including the IEEE Globecom, the IEEE ICC, and the IEEE WCNC.
ultrawideband radio, machine-to-machine communications, wireless secu- He received the Best Paper Awards of the IEEE MAPE in 2013, IEEE/CIC
rity, e-health, smart grid, nano-communications, and signal processing for ICCC in 2014, and the IEEE Globecom in 2014. He was Elected Core Team
communication applications. She is an Editor for the IEEE TRANSACTIONS ON Member with the Jiangsu Innovation Team in 2012. He was the co-recipient
VEHICULAR TECHNOLOGY. of the First Prize Award of Jiangsu Science and Technology Award in 2014.

14010 VOLUME 5, 2017