LIBRARY

FOCUS ON:

TECHNOLOGY
A compendium of articles
AND CHALLENGES
from the Editors of Microwaves & RF

Copyright © 2020 by Endeavor Media.

All rights reserved.
 5G delivers high throughput but
there are tradeoffs such as limited
reach and the need for massive
MIMO. Design is not easy but it is
manageable.
5G wireless technology promises
to deliver streaming of high definition
video as well as providing more
users with this bandwidth. It has
planned uses for automotive safety Bill Wong
and support as well as low latency Editor,
Senior Content
for IoT applications. There are even
Director
military applications that can take
advantage of 5G.
There are costs and tradeoffs
though. Small cell sizes will be the norm since the range
of 5G access points is lower than 4G. This means more
cells in addition to needing need equipment. 5G support
in the latest smartphones is useless unless there is a 5G

CONTENTS
tower nearby.
In this ebook we include sections that address these
issues and more such as massive MIMO.

Understanding
5G and IoT RF

02 11
Small Cells Help Systems with
Keep 5G Off-the-Shelf
Connected Components

05 16
Defining
The Building Massive
Blocks of 5G MIMO in a 5G

09 23
How 5G Could More
Impact the resources from
Military Microwaves & RF

☞ LEARN MORE @ mwrf.com | 1

 LIBRARY

CHAPTER 1:

Small Cells Help

Keep 5G Connected
TUAN NGUYEN, Product Line Director of Wireless Infrastructure Qorvo Inc. (TriQuint)

A
s more users tap into wireless-communications services, the demand on wireless
Key to the success of network capacity intensifies in both indoor and outdoor locations. Such network
small cells, which boost density or densification further pressures wireless carriers to keep pace with the
increased consumption of frequency bandwidth via voice, video, and data. It also
quality of service in drives those carriers to expand their cellular/wireless infrastructure with minimum
increases in cost or disruption of service to wireless customers. Thus, many are turning to
wireless networks, is small cells as a solution.
The rollout of fifth-generation (5G) wireless networks will address demands for
their high-frequency increased capacity and data, but these networks are still some years away. So, a more
practical answer that’s in keeping with today’s fourth-generation (4G) wireless networks is
components that draw to use small cells, which function as miniature base stations that are added to an existing
wireless network. They operate at relatively low power levels to fill any “holes” that exist in
upon a number of different wireless coverage, in both indoor and outdoor locations.
As an example of the magnitude of growing wireless user demands, fans attending the
technologies. National Football League’s (NFL) 2016 Super Bowl championship game in Santa Clara,
Calif., used more than 7 TB of data on the Verizon Wireless network alone—nearly three
times as much data used at the 2015 Super Bowl game. Fans connected to the network
via smartphones and many other unique wireless devices. They benefited from the
generous capacity provided by 4G Long Term Evolution (LTE) technology bolstered by the
use of small cells, macrocells, and mobile cell sites (as reported by Verizon Wireless in a
press release dated February 8, 2016).

Small Cells + DAS Solutions

The use of small cells and distributed antenna systems (DAS) are being employed

☞ LEARN MORE @ mwrf.com | 2

 LIBRARY CHAPTER 2: Small Cells help keep 5G Connected

The Evolution of 5G (Fig. 1) to achieve increased

2015 2016 2017 2018 2019 2020 data capacity in 5G wireless
networks while enhancing
90 to 95 GHz 2.9 GHz quality of service (QoS) in those
available
networks. But with densification
70 to 85 GHz 10 GHz
available comes interference and
mobility handover challenges
4 GHz
38 GHz
available between small cells and the
28 GHz
2 GHz macro network, requiring
Massive MIMO and available
careful network design and
Macro base station Densification digital beamforming Millimeter waves
management.
• Increasing frequency • Small cells • Up to 6 GHz • Ultra-wide bandwidth As mobile communications
beyond 2.7 GHz to 6 GHz • DAS • Array antenna • Ultra-high throughput
• Increasing bandwidth • LTE-U devices move from a macrocell
• Increasing efficiency environment to a small-cell
1. A number of technologies will be required to optimize performance in 5G wireless coverage area, the network will
networks, including distributed antenna systems (DAS), millimeter-wave technology, and instantaneously switch profiles.
small-cell basestations. The relatively low power levels
of small cells allow mobile
devices to be in close proximity to those small cells. They are able to gain network access
while conserving battery life, and they needn’t establish radio communications links with
more distant larger cell sites at transmission levels diminished by distance. Mobile-device
users will benefit from increased data speeds, mobility, and flexibility, too, since small
cells and DAS solutions support multiple standards, such as third-generation (3G) and 4G
cellular, and implement carrier aggregation with LTE Advanced (LTE-A) systems.
Antenna technologies such as multiple-user multiple-input, multiple-output (MIMO)
approaches, beam steering, and phased-array techniques help provide additional wireless
coverage and interoperability of multiple-band 3G and 4G systems. With massive MIMO
methods, wireless network operators can increase data rates and network capacity by
transmitting multiple, spatially separated data streams over the same frequency band,
using multiple antennas on the base station and the user’s device.
Massive MIMO base stations seek to provide from 16 to 256 channels, challenging
designers of those base stations to target smaller component sizes with high energy
efficiency and effective thermal management. Therefore, highly power-efficient
semiconductor technologies like gallium nitride (GaN) become attractive for such base
stations.
High levels of semiconductor integration are also instrumental in achieving the high
channel counts in these relatively small-sized base stations.
TABLE 1: SIZING UP SMALL CELLS Squeezing as many as 256 transmit channels into a single base
Output Cell radius
station requires subsystems that package power amplifiers (PAs),
Cell type Users Locations
power (W) (km) low-noise amplifiers (LNAs), and switches into compact modules,
Femtocell 0.001 to 0.25 0.010 to 0.1 1 to 30 Indoor and employing small-form-factor filter solutions.
30 to Indoor/
Picocell 0.25 to 1 0.1 to 0.2
100 outdoor Technology Potpourri
Microcell 1 to 10 0.2 to 2.0
100 to Indoor/ As Table 1 shows, small cells differ in output power levels,
2000 outdoor
coverage areas, and number of users served. For the best
Macrocell 10 to >50 8 to 30 >2000 Outdoor
performance and power efficiency, the subsystems used in small-

☞ LEARN MORE @ mwrf.com | 3

 LIBRARY CHAPTER 2: Small Cells help keep 5G Connected

cell basestations must combine components

based on different process technologies. For
example, GaN PAs may provide suitable output
power and power efficiency, but gallium-arsenide
(GaAs) technology may make more sense for
the LNAs needed to minimize receiver noise and
maximize receiver sensitivity.
Filters could require yet a third technology,
especially for operating conditions that may 2. The highly integrated model
experience extremes of temperature and humidity. TQP9218 is an example of a GaN PA
Temperature-stable bulk-acoustic-wave (BAW) for small-cell base stations. The space-
LowDrift filters from Qorvo provide a solution for saving device comes in a low-cost 7- ×
filtering high-power signals while also avoiding 7-mm SMT housing.
interference from adjacent frequency bands.
In addition to the various components required for wireless infrastructure designs,
including filters, switches, and LNAs, Qorvo developed a line of highly integrated PAs for
small-cell base stations. The PAs do not require linearization, and feature on-chip bias
control and temperature-compensation circuitry to further simplify the design of a small-
cell base station. They are available with +24- or +27-dBm average linear output power
when driving a 20-MHz-wide LTE signal (Table 2). The PAs also incorporate two stages of
amplifier gain in low-cost surface-mount-technology (SMT) packages.
For example, model TQP9218 is a 0.25-W (+24-dBm) GaN PA designed for small-cell
base stations operating from 1805 to 1880 MHz. It offers 31-dB small-signal gain across
its frequency range with internal impedance matching, on-chip
bias control circuitry, and temperature-compensation circuitry—all
TABLE 2: PAS FOR SMALL-CELL BASE STATIONS packed into a 7- × 7-mm RoHS-compliant SMT housing (Fig. 2).
Frequency range Average output The PA achieves 16% power-added efficiency (PAE) and draws just
PA model
(MHz) power (dBm) 240-mA quiescent current from a +4.5-V dc supply.
TQP9218 1805 to 1880 +24 Simply put, better wireless network coverage is needed to
TQP9418 1805 to 1880 +27 support the demand for more voice, video, and data among the
QPA9219 1930 to 2000 +24 growing number of wireless-communications users. Solutions for
QPA9419 1930 to 2000 +27 enhanced coverage include DAS and the addition of small-cell
TQP9221 2010 to 2170 +24 base stations to larger macro-cells. As technologies like automated
TQP9421 2010 to 2170 +27
vehicle control and safety along with literally millions of Internet of
Things (IoT) devices flood wireless operating bands, the need for
TQP9224 2300 to 2400 +24
small-cell base stations will only increase in the years to come.
TQP9424 2300 to 2400 +27

QORVO INC. (TRIQUINT), 2300 NE Brookwood Parkway,

Hillsboro, OR 97124; (844) 890-8163, www.qorvo.com.

to view this article online, ☞click here

☞ BACK TO TABLE OF CONTENTS

☞ LEARN MORE @ mwrf.com | 4

 LIBRARY

CHAPTER 2:

The Building Blocks of 5G

RICH FRY, Director of ICT Sales, TDK Corporation of America

W
ith every generation of mobile networks—from 1G to 4G—came a significant im-
The complexities and provement over the preceding system. Unlike these transitions, there won’t be
applications of 5G one massive step that moves us into the impending 5G standard. Many incre-
mental steps, or building blocks, comprise this transition. By working together to
networks will bring about address the various elements that make up 5G, engineers and telecom profes-
sionals will create a connected network with unprecedented speed and performance for
step-by-step innovations 2019 and beyond.
More than 30 years ago, 1G delivered analog voice, but it was limited due to the phones’
that will procedurally primitive ability to change frequencies between cellular towers. Phones had to work harder
to keep connected when moving between service towers, shifting from the frequency of
add unprecedented one to another using microprocessors without dropping the call.1
2G instantly rendered the processes of 1G obsolete by introducing digital voice. Time-
functionality to our public division multiple-access (TDMA) RF technology in the 1990s allowed mobile phones to
maintain connections to numerous frequencies simultaneously, eliminating the need to
spaces. reconnect from one signal at a time to hold a call. Network capacity also significantly grew,
so many users could be connected with little issue.2
By 2003, mobile web became feasible due to the widened bandwidth of 3G that allowed
for mobile data. While bandwidth during the 3G era wasn’t optimal for demanding loads
such as video, the breadth of possibilities for the platform was widening.3
4G ushered in the era of true mobile internet, breaking the speed barrier into megabytes
per second. 4G also gave way to a new IP-based, consolidated network. For carriers,
embracing 4G meant further digitizing voice by moving calls to VoIP. In the beginning,
typical 4G speeds began at around 4 Mb/s, with the latest LTE-A speeds reaching 200
or even 300 Mb/s.4,5 Mobile bandwidth that can support all standard internet practices
is now commonplace, with 5G set to open the floodgates to myriad new applications—
incrementally.
The FCC established a 5G standard in June 2018, marking a significant step in the

☞ LEARN MORE @ mwrf.com | 5

LIBRARY CHAPTER 2: The Building Blocks of 5G

adoption process. It later announced and held two auctions for high-band frequency
spectrum in the fall of 2018 in order to facilitate the deployment of 5G technology.6 To
design engineers, this means that the proliferation of 5G IoT devices is around the corner,
and the miniaturization of millimeter-wave (mmWave) components is beginning to become
more urgent.
Although many of today’s IoT devices don’t require significantly wider bands to operate,
5G networks will need to support massive numbers of connections, which will open up
possibilities for a swath of new applications for which these devices can serve. These
include autonomous-vehicle and connected-car technology, smart city sensors for
transport and security, as well as medical devices and AR/VR applications.

How Do We Achieve 5G? Why Aren’t We There Yet?

Although the 5G groundwork has begun to be laid down in multiple major U.S. cities, this
is but step one of a long process. In reality, the 5G rollout is still in its infancy, but major
mobile carriers would say it differently. The first flagship 5G networks might technically
be live, but they only exist in select American cities and are hotspot-based, serving like a
public Wi-Fi connection.7 Today’s handset companies are just now beginning to announce
5G-enabled phones with first releases of these phones in the second half of 2019. Most of
these phones will support <6-GHz technology. The first prototypes of mmWave technology
incorporated are currently being shown at technology seminars.  
Millimeter waves exist at the top of the radio spectrum and have been untapped until
now, leaving much opportunity for gigabit internet speeds to roam without bottlenecking.8
Since high-band signals by nature can’t travel as far as slower-speed connections, small-
cell technology has become a necessity for 5G. Small-cell base stations are miniaturized
towers that can be placed throughout metropolitan areas so that high-band signals don’t
drop. These will become imperative for dense 5G networks with millions of devices,
including connected cars, IoT sensor networks, and more.
Moreover, massive multiple-input, multiple-output (MIMO) technology will ensure that
plenty of 5G signal frequencies propagate for users so that they’re seamlessly connected
at all times. Beamforming will become more commonplace in 5G networks as well,
isolating the locations of devices and providing a direct line of signal service rather than
broadcasting to an inefficient, broad radius. Beamforming can also involve full duplex,
which is the simultaneous sending and receiving of data between devices.
5G offers networking companies an opportunity to continuously test new technologies
that will enable ultra-low-latency networks, providing users with an enhanced experience
while maintaining low costs. But due to the breadth of these technologies, rollout is
naturally going to be gradual.
Some current obstacles challenging design engineers, in terms of tech trends and
specific obstacles, include the following.

Rapid Rise of IoT Adoption

According to Gartner, by 2020, IoT technology will be in 95% of electronics for new
product designs. The rise in IoT adoption is considered a challenge for network providers
to truly achieve 5G. As more users go online, 4G networks will reach a limit of what they
can provide for the best possible user experience. Millions of IoT connections inevitably

☞ LEARN MORE @ mwrf.com | 6

LIBRARY CHAPTER 2: The Building Blocks of 5G

entering any network will not only be stretching their bandwidth capabilities, but will also
be testing the sheer number of devices that are able to be connected at a given time.
For example, as smart-city sensors begin to proliferate, they will be flooding networks
with location-specific data on transport and other environmental measurements such
as temperature, humidity, and more. As these sensors reach every street corner and
autonomous vehicles begin to fill streets, IoT bandwidth in its current form will be
antiquated—especially since millions of mobile phones will remain active in these networks.

Computing and Bringing AI to the Edge

5G will need to be able to understand what data/input is critical/not critical for decision-
making at the edge. Edge computing, a natural extension of cloud computing (and what
some are calling “fog”), will be taking the computing load off of small IoT devices within
5G networks.
Artificial intelligence and deep learning are critical to performing these processes
between devices. They will bring routing, storage, computing, and server functions to the
network edge in order to run mission-critical applications for both enterprises or individuals.
Engineers and networking professionals will be hard at work achieving ultra-low latency
for data-intensive technologies that can’t afford to have a delay, including autonomous
vehicles and medical devices.
By implementing 5G and enabling edge computing for connected cars, increased safety
and optimized traffic flow will be possible. Augmented-reality-enabled windshields could
give drivers real-time prompts or directions with 5G edge connectivity. 5G-enabled medical
devices will be able to deliver prompt warnings for any illnesses or emergencies, so that
patients can act on diagnoses as quickly as possible.

Conclusion
The building blocks that will gradually build out our future 5G networks will provide us
with procedural innovation in numerous industries. The good news is that this will ease
the general public into some truly groundbreaking technology. The bad news is that it will
not happen quickly.
OEMs of components, antennas, and sensors, as well as chipmakers and network
carriers, bear the responsibility of coming together for this greater purpose. Once this unity
is achieved across all of 5G’s challenges, connectivity will enable amazing unforeseen
advances in AR, VR, autonomous vehicles, and even simple innovations such as
download speeds 1,000X faster than 4G LTE. But first and foremost, design engineers
must continually and procedurally innovate at the RF and component level.

RICH FRY is the Director of ICT Sales at TDK. With over 25 years of experience
working in the information and communication electronics industry, Fry is considered an
influential leader within the engineering industry. Prior to his current role, Fry served as
the power-supply business development manager at TDK, where he was responsible
for executing business development strategy, with a primary focus on power supplies
for the communication industry, as well as a Global Account Manager for Various Tier 1
Communications Customers. Fry holds a Bachelor’s of Science in Electrical Engineering
from Rutgers University.

☞ LEARN MORE @ mwrf.com | 7

LIBRARY CHAPTER 2: The Building Blocks of 5G

to view this article online, ☞click here

☞ BACK TO TABLE OF CONTENTS
References:
1. https://www.forbes.com/sites/simonrockman1/2018/08/14/etacs-technology-is-1g/
2. https://www.forbes.com/sites/simonrockman1/2018/08/15/gsm-technology-is-2g/
3. https://www.forbes.com/sites/simonrockman1/2018/08/16/wcdma-is-3g/
4. https://www.engadget.com/2011/01/17/2g-3g-4g-and-everything-in-between-an-
engadget-wireless-prim/
5. https://www.pcmag.com/Fastest-Mobile-Networks
6. https://www.usatoday.com/story/tech/talkingtech/2018/07/11/faster-phones-smarter-
devices-come-fcc-5-g-spectrum-auctions/773270002/
7. https://www.pcmag.com/Fastest-Mobile-Networks/4
8. https://www.engadget.com/2018/07/23/how-5g-makes-use-of-millimeter-waves/

☞ LEARN MORE @ mwrf.com | 8

 LIBRARY

CHAPTER 3:

How 5G Could Impact the Military

JACK BROWNE, Contributing Editor

C
ommercial media attention to 5G cellular wireless communications systems has
Though jamming grown steadily in recent months, in anticipation of a wireless communications net-
interruption and signal work that will leave no stone unturned and possibly no citizen without a cellphone.
But what does such a significant technological introduction as 5G, with its massive
interception loom as infrastructure building on top of existing 3G and 4G Long Term Evolution (LTE) tow-
ers, mean to defense electronic technology?
potential problems, How well will critical defense electronic systems, such as terrestrial radar and
manpack portable radios, coexist with what promises to be the largest cellular wireless
5G could change the communications network ever built on Earth? Not to mention that future commercial
vehicles will be driving by those 5G networks aided by advanced driver-assistance
defense landscape with systems (ADAS), with their own radar signals bouncing from nearby signposts?
Military system designers, of course, will take advantage of available 5G hardware and
instantaneous transmission software for current and future systems, benefiting from such features as fast response
times and wide bandwidths that allow for, say, lightning-fast transmission and reception of
and reception of enormous images representing battlefield scenarios.
amounts of data. Perhaps the real question will be: Can all potential future 5G users coexist? Natural
limiting factors for military communications systems, such as mountainous terrain and
rainfall attenuation, will impact the effectiveness of 5G systems in all cases.
Achieving full performance of military 5G technologies with minimal interference from
operating environments and other users will depend on creative computer software
simulations. It will also require effective measurement strategies to predict the different
operating scenarios that an almost “unlimited” number of 5G users, commercial and
military, will face.
Military designers can take advantage of different, more ruggedized components and
packaging than used in more cost-conscious commercial radios, and perhaps unique
military nuances to the operating environment, such as portable, transportable 5G base
stations that can be moved around as needed. But they will face the usual obstacles to

☞ LEARN MORE @ mwrf.com | 9

LIBRARY CHAPTER 3: How 5G Could Impact the Military

overcome from opposing forces, such as high-power jamming signals. Because of the
wide bandwidths covered by 5G systems, jammers will have to follow into the millimeter-
wave range to jam close-range systems.
Whether for military or commercial users, 5G wireless networks will depend on higher-
frequency signals, such as 28 GHz and beyond, traveling shorter distances, compared to
4G networks that operate with longer-wavelength, lower-frequency signals at typically 3.5
GHz and below. The consumption of bandwidth at lower frequencies has quickly limited
the transmission speeds of 4G wireless networks to the Mb/s range. On the other hand,
5G system designers are hoping to reach transmission speeds of 1 Gb/s and beyond,
albeit at shorter distances than 4G systems.
Military users will no doubt benefit from the available bandwidths and transmission
speeds of 5G devices, along with the reduced lag times of those higher-frequency,
millimeter-wave signals. As always, the problems will come from an adversary’s use of the
same wireless communications technologies, which will be freely available to all.
Over-occupied bandwidth at lower frequencies was one of the motivating factors for
extending 5G wireless communications to the millimeter-wave frequency range—some
of those same 5G frequency bands may come into play in ADAS systems—or simply
because of military and aerospace use of those same higher-frequency bands. The 5G
spectrum landscape presents a complex map for system modelers, especially with the
constant threat of jamming interruption and signal interception by adversaries. But the 5G
rewards, for both military and civilian users, include almost instantaneous transmission
and reception of enormous amounts of data, resulting in highly informed and educated
users on both sides.
to view this article online, ☞click here
☞ BACK TO TABLE OF CONTENTS

☞ LEARN MORE @ mwrf.com | 10

 LIBRARY

CHAPTER 4:

Understanding 5G and IoT

RF Systems with Off-the-Shelf
Components
HOW-SIANG YAP, EEsof EDA Genesys Product Planner,
Keysight Technologies (www.keysight.com)

5
We explain how the process G does not mean 5 GHz. 5G is the upcoming fifth-generation wireless mobile net-
work, operating from 24 GHz and up to 95 GHz. It promises extremely high-data-rate
of architecting and building wireless connections such as 4K/8K ultra high-definition TV streaming. The Internet
of Things (IoT) is another fast-growing application of wireless technology. The IoT is
RF systems for 5G and the networking of things around us—from personal gadgets to industrial sensors and
freight tracking around the world. By 2020, over 50 billion IoT objects are forecasted to be
IoT applications can be in operation. This means tremendous work and pressure for RF/microwave engineers to
design and build 5G and IoT products quickly to compete for a share of the market.
performed efficiently
5G 28-GHz RF System Simulation
through accurate RF system Designing and building RF systems to operate at 24 GHz and above is challenging
due to the parasitics of interconnects, peripheral biasing, passive components, and the
simulation with real off- absence of simulation models of available system components. Calculating with spread-

the-shelf components and sheets and then bread boarding with actual hardware is very costly in terms of time, instru-
mentation, and effort for all ensuing iterations.
fast multi-stage impedance A more efficient approach to design, prototype, and realize RF systems in one pass is
now possible and is validated in the example below.
matching synthesis on a Figure 1 shows the block diagram of a 5G system with a 28-GHz RF input and two
down-conversion local oscillators (LOs) at 22 GHz and 7 GHz, respectively. The system
printed circuit board layout. produces a 1-GHz intermediate frequency (IF).

☞ LEARN MORE @ mwrf.com | 11

 LIBRARY CHAPTER 4: Understanding 5G and IoT RF Systems with Off-the-Shelf Components

RF: 28 GHz, BW: 3 GHz IF1: 6 GHz, BW: 9 GHz The block diagram is simulat-
MLI_0832LSM_2
ed using the Keysight Genesys
Spectrasys system simulator
Source
XM-A2T6-0404D XM-A123-0404D
RCAT_01A RCAT_01B
XM-A2F6-0404D XM-A121-0404DA with system blocks modeled
as:
X-Microwave
Keysight Sys-Parameters • X-Parameters for nonlinear
×N
X-Parameters
S-Parameters
circuit data
XM-A398-0404D XM-A234-0404D • Sys-Parameters for system
datasheet behavioral data
IF2: 1 GHz, BW: 40 MHz
with frequency, bias, and
SIM_852MH temperature dependence
• S-Parameters for linear cir-
RCAT_02A RCAT_02B Transition
Out cuit data
XM-A278-1204D XM-A1A1-0204D XM-A111-0404D
• Behavioral equation-based
models
LO2
Behavioral RF system simulation tech-
XM-A121-0404DB nology has come a long way
since the use of spreadsheets.
Fig 1. Shown is a block Improvements in accuracy and diagnostic capabilities are significant. An example of
diagram of a 28-GHz 5G RF these improvements includes identifying the component origin and frequency equation
receiver system with dual of nonlinear intermods. This includes knowing the system blocks and their specs in the
downconversion to 1 GHz. The system line up and how they contribute to performance degradation such as error vector
system was simulated with the magnitude (EVM), bit error rate (BER), and adjacent channel power ratio (ACPR) under
Keysight Genesys RF system digital modulated RF stimulus.
simulator. Figure 2 shows a budget analysis of EVM error versus the system component lineup.
This analysis instantly identifies the major contributors of EVM degradation as LO phase

Fig 2. EVM budget analysis can diagnose which components in the RF system lineup are causing degradation to digital
modulated RF signals in the system path to prevent wasteful hardware prototype iterations.

☞ LEARN MORE @ mwrf.com | 12

LIBRARY CHAPTER 4: Understanding 5G and IoT RF Systems with Off-the-Shelf Components

Fig 3. Shown is the

hardware prototype
of the 28-GHz
receiver system
using X-Blocks
from X-Microwave.
Discrepancies caused
by interconnect
parasitics or
inaccurate system
models are eliminated,
allowing for “what-
you-simulate-is-what-
you-get” accuracy.

noise and linearity of the mixer and amplifiers. Clicking on the culprit components enables
their offending specs to be adjusted to improve the system performance. This enables
proper selection of the system component lineup without over- or under-specifying them
to achieve the best performance with minimum cost.
Specifying the behavioral specs during the design process and then trying to find real
components that have such specs during realization is a common but inefficient approach
that will inevitably result in multiple iterations. Sys-Parameters representing simulatable
datasheets of real off-the-shelf components or X-Parameters of measured nonlinear
components can be used directly in RF system simulation. This way, when the design is
complete, parts have also been already specified and verified to work in the system chain.
The system is now ready for hardware implementation.

5G System Realization with Off-the-Shelf Parts

Off-the-shelf parts from vendors such as Mini-Circuits, Analog Devices, Qorvo, Marki
Microwave, and Broadcom implemented into modular tiles by X-Microwave were used to
realize the 28-GHz 5G RF receiver system (Fig. 3). Each modular tile, called an X-Block,
includes all the biasing and peripheral passive components for the active device such as
LOs, mixers, and amplifiers. They are characterized by measured X-Parameters or Sys-
Parameters at their co-planar interconnect reference planes for simulation to accurately
model how they are being used in the actual system hardware.
The parts are connected by a flipped, co-planar laminate that spans the small gap
between the X-Blocks. These laminates are held down by compression without soldering
to reliably work up to 67 GHz. The 1.9-mm test launchers are also held down by compres-
sion so that the X-Blocks can be reused with no damage. When the prototype is finalized,
the same composite layout can be used directly for production, since they are all built on
the same laminate material.
When the system was measured, the agreement with the simulated result was unex-
pectedly close, and is within the uncertainty error of the vector signal analyzer (VSA; see
the table).

☞ LEARN MORE @ mwrf.com | 13

 LIBRARY CHAPTER 4: Understanding 5G and IoT RF Systems with Off-the-Shelf Components

Input Power Measured EVM Estimated EVM

Designing for the IoT
(dBm) (%rms) (%rms) Multiple standards are emerg-
-50 1.8 2.2 ing for IoT radios based on range
-60 3.1 3.5 of coverage, data bandwidth, and
-70 9.3 9.3 operating frequencies. The IoT fre-
-75 17.9 16.3 quencies can be broadly divided
into two categories: Sub-1-GHz
-78 20.0 22.9
and those above—namely around
the 2.4- and 5.8-GHz industrial-sci-
entific-medical (ISM) bands. From the perspective of designing IoT physical radio links
that work at these frequency bands, the focus should be on impedance matching the IoT
chipset to the antenna. For a longer range, amplifiers may be inserted in between the
chipset and the antenna.
Ideally, the impedance matching network must be compact and economical to build.
Multi-stage impedance matching over a broad bandwidth (30% or more) to complex fre-
quency-dependent impedances such as an antenna; measured S-parameters of an IoT
chipset; or an unstable non-unilateral discrete transistor amplifier is extremely difficult and
tedious using traditional Smith chart or benchtop cut-and-try techniques.
A more efficient and optimal approach involves using automatic impedance matching
synthesis. This technique employs multiple algorithms from simple L-sections to the Real-
Frequency-Technique to address the increasingly difficult aforementioned impedance
matching problems. Because synthesis can accomplish difficult simultaneous multi-stage
matching in seconds with distributed and/or lumped networks, the IoT radio designer can
quickly experiment with multiple matching topologies to select the one that is most eco-
nomical to build.
Figure 4 shows a three-stage matching network used to match an antenna to a low-
noise stabilized transistor amplifier circuit. The measured S-parameters of a chipset power
amplifier (PA) are shown, demonstrating a return loss of -20 dB and a gain of 35 dB from

2 MONTH.2013 HYDRAULICS & PN

Fig 4. Shown is the impedance matching synthesis and microstrip layout of a three-stage matching network from 2 to 3 GHz
to achieve -20 dB of return loss and 35 dB of gain. The process was completed in less than one hour.

☞ LEARN MORE @ mwrf.com | 14

LIBRARY CHAPTER 4: Understanding 5G and IoT RF Systems with Off-the-Shelf Components

2 to 3 GHz. The microstrip layout dimensions were also synthesized with the automatic
insertion of discontinuities, such as tees and open stubs. The entire process was complet-
ed within one hour.

Summary
RF systems for 5G and IoT applications can now be efficiently simulated, prototyped,
and produced with off-the-shelf system components thanks to breakthrough diagnostic
capabilities that can pinpoint any insufficiently performing components in the system line-
up. Accurate X- and Sys-Parameter simulation models of off-the-shelf RF system compo-
nents enables “what-you-simulate-is-what-you-get” efficiency in going from design to pro-
totype and production with no iterations. Impedance matching synthesis replaces tedious
manual design and optimization with an instant selection of various suitable matching
topologies for the most economical realization.
To learn more visit www.keysight.com/find/eesof-genesys-info and www.xmicrowave.
com.
to view this article online, ☞click here
☞ BACK TO TABLE OF CONTENTS

☞ LEARN MORE @ mwrf.com | 15

 LIBRARY

CHAPTER 5:

Defining Massive MIMO in a 5G

CLAIRE MASTERSON, Analog Devices www.analog.com

T
he insatiable demand for high-speed mobile data is creating a series of pressing de-
Massive MIMO technology sign challenges as today’s cellular base stations strain to handle an increasingly satu-
and 5G communications rated RF spectrum. In many dense urban areas, our ability to continuously accelerate
transmit and receive data speeds is under threat.
are often mentioned in the One path forward is to deploy base stations with large numbers of antennas that
simultaneously communicate with multiple spatially separated user terminals over the
same sentence. This article same frequency resource and exploit multipath propagation. Often referred to as massive
multiple-input, multiple-output (MIMO), this technology is also described as beamforming
takes an in-depth look with a large number of antennas. But this raises the question: What is beamforming?

at the former, and how it Beamforming vs. Massive MIMO

Beamforming is a word that means different things to different people. According
applies to the latter. to its basic definition, it is the ability to adapt the radiation pattern of the antenna
array to a particular scenario. In the cellular
communications space, many people think of
beamforming as steering a lobe of power in
a particular direction toward a user (Fig. 1).
Relative amplitude and phase shifts are applied
to each antenna element to allow for the output
signals from the antenna array to coherently
add together for a particular transmit/receive
angle and destructively cancel each other
out for other signals. The spatial environment
that the array and user are in is not generally
considered. This is indeed beamforming, but is
1. This figure illustrates just one specific implementation of it.
traditional beamforming. Massive MIMO can be considered as a form

☞ LEARN MORE @ mwrf.com | 16

LIBRARY CHAPTER 5: Defining Massive MIMO in a 5G

of beamforming in the more general sense

of the term, but is quite removed from the
traditional form. “Massive” simply refers to
the large number of antennas in the base-
station antenna array. “MIMO” refers to the
fact that multiple spatially separated users
are catered for by the antenna array in the
same time and frequency resource.
Massive MIMO also acknowledges that
in real-world systems, data transmitted
between an antenna and a user terminal—
and vice versa—undergoes filtering from
2. Shown is a multipath environment between the
the surrounding environment. The signal
antenna array and user.
may be reflected off buildings and other
obstacles, and these reflections will have
h1, 1 an associated delay, attenuation, and
1 direction of arrival (Fig. 2). There may not
1 h1, 2
even be a direct line-of-sight between the
antenna and the user terminal. It turns out
hM, 2 2
hM, 1 that these non-direct transmission paths
can be harnessed as a power for good.
h1, K In order to take advantage of the
M
multiple paths, the spatial channel
hM, K
between antenna elements and user
K terminals needs to be characterized. In
the literature, this response is generally
h1, 1 h1, K referred to as channel state information
(CSI). This CSI is effectively a collection
H=
of the spatial transfer functions between
hM, 1 hM, K
each antenna and each user terminal.
Such spatial information is gathered in a
3. Channel state information is needed to characterize
matrix (Fig. 3). The next section looks at
a massive MIMO system.
the concept of CSI and how it is collected
in more detail. The CSI is used to digitally
encode and decode the data transmitted
from and received by the antenna array.

Characterizing the Spatial Channel

Between Base Station and User
An interesting analogy is to consider
a balloon being popped at one location,
and the sound of this pop (or impulse)

4. An audio analogy demonstrates the spatial

characterization of a channel.

☞ LEARN MORE @ mwrf.com | 17

LIBRARY CHAPTER 5: Defining Massive MIMO in a 5G

being recorded at another (Fig. 4). The sound recorded at the microphone position is a
spatial impulse response that contains information unique to the particular position of both
the balloon and the microphone in the
surrounding environment. The sound that
is reflected off obstacles is attenuated and
delayed compared to the direct path.
If we expand the analogy to compare to
the antenna array/user terminal case, we
need more balloons (Fig. 5). Note that in
order to characterize the channel between
each balloon and the microphone, we
need to burst each balloon at a separate
time so the microphone doesn’t record
the reflections for different balloons
overlapping. The other direction also
5. This figure is an illustration of an audio analogy for needs to be characterized (Fig. 6). In
downlink channel characterization. this instance, all the recordings can be
done simultaneously when the balloon is
popped at the user terminal position. This
is clearly a lot less time-consuming!
In the RF space, pilot signals are used
for characterizing the spatial channels
(Fig. 7). The over-the-air transmission
channels between antennas and user
terminals are reciprocal, meaning the
channel is the same in both directions.
This is contingent on the system operating
in time division duplex (TDD) mode as
opposed to frequency division duplex
(FDD) mode.
6. Shown is an audio analogy for uplink channel In TDD mode, uplink and downlink
characterization. transmissions use the same frequency
resource. The reciprocity assumption
means the channel only needs to be
characterized in one direction. The uplink
channel is the obvious choice, as just one
pilot signal needs to be sent from the user
terminal and is received by all antenna
elements.
The complexity of the channel estimation
is proportional to the number of user
terminals, not the number of antennas in
the array. This is of critical importance
given the user terminals may be moving,
7. This figure illustrates how each user terminal and hence the channel estimation
transmits an orthogonal pilot symbol. will need to be performed frequently.

☞ LEARN MORE @ mwrf.com | 18

 LIBRARY CHAPTER 5: Defining Massive MIMO in a 5G

Another significant advantage of uplink-based

characterization is that all the heavy duty channel
estimation and signal processing is done at the
M>>K
base station, and not at the user end.
So now that the concept of collecting CSI has
been established, how is this information applied
y1 to data signals to allow for spatial multiplexing?
ŝ1 Filtering is designed based on the CSI to precode
Detection
ŝ2
Matrix
the data transmitted from the antenna array so
K×M that multipath signals will coherently add at the
ŝK
yM
user terminals position. Such filtering can also be
K Signals
used to linearly combine the data received by the
antenna array RF paths so that the data streams
Channel State
M Antennas Information H from different users can be detected. The following
K Users
section addresses this in more detail.
DETECTION TYPE
Maximum ratio (MR) ŝ = HHy The Signal Processing that Enables
Zero forcing (ZR) ŝ = (HHH)-1HHy Massive MIMO File: Callouts_41Q

NMSE or RZF ŝ = (HHH + βℓ)-1HHy In the previous section we described howFigure 3the
CSI (denoted by the matrix H) is estimated.
8. Shown is a depiction of uplink signal processing. H denotes the Detection and precoding matrices are calculated
conjugate transpose. based on H, and there are a number of methods
for calculating these matrices. This article focuses
on linear schemes. Examples of linear precoding/detection methods are maximum ratio,
zero forcing, and minimum mean-square error. Full derivations of the precoding/detection
filters from the CSI are not provided in this article, but the criteria they optimize for—as
well as the advantages and disadvantages of each method—are discussed. Additional
references provide a more detailed treatment of these topics. 1, 2, 3
Figures 8 and 9 give a description of how the signal processing works in the uplink
and downlink, respectively, for the
three linear methods previously
mentioned. For precoding, there
Figure 9

M>>K may also be some scaling matrix

to normalize the power across theType
Precoding
Maximum ratio (MR)
x1 array that has been omitted for (ZR)
Zero forcing
NMSE or RZF
s1 simplicity.
s2 Precoding
Matrix Maximum ratio filtering, as the
M×K
sK xM name suggests, aims to maximize
K Signals
the signal-to-noise ratio (SNR).
Channel State It is the simplest approach from
M Antennas Information HT K Users
Figure 10
TBS
PRECODING TYPE M Antennas
Maximum ratio (MR) x = H*s 9. This figure illustrates downlink
Channel State Informatio
RUE
signal processing. T denotes the
K Users
Zero forcing (ZR) x = H*(HTH*)-1s
transpose, while * denotes the
NMSE or RZF x = H*(HTH* + βℓ)-1s
conjugate. Figure 11
Digital Baseband
SYSREF
REFCLK
SYSREF
☞ LEARN MORE @ mwrf.com | 19 REFCLK
Clock generation
AD9528
 LIBRARY CHAPTER 5: Defining Massive MIMO in a 5G

a signal processing viewpoint, as the detection/precoding matrix is just the conjugate

transpose or conjugate of the CSI matrix, H. The big downside of this method is that inter-
user interference is ignored.
Zero forcing precoding attempts to address the inter-user interference problem by
designing the optimization criteria to minimize for it. The detection/precoding matrix is the
pseudoinverse of the CSI matrix. Calculating the pseudoinverse is more computationally
expensive than the complex conjugate, as in the maximum ratio case. However, by
focusing so intently on minimizing the interference, the received power at the user suffers.
Minimum mean-square error tries to strike a balance between getting the most signal
amplification and reducing the interference. This holistic view comes with signal processing
complexity as a price tag. The minimum mean-square error approach introduces a
regularization term to the optimization—denoted as β in Figs. 8 and 9—that allows for a
balance to be found between the noise covariance and the transmit power. It is sometimes
also referred to in literature as regularized zero forcing (RZF).
This is not an exhaustive list of precoding/detection techniques, but gives an overview of
the main linear approaches. Nonlinear signal processing techniques, such as dirty paper
coding and successive interference cancellation, can also be applied. Although these
techniques offer optimal capacity, they are very complex and difficult to implement.
The linear approaches described are generally sufficient for massive MIMO, where the
number of antennas gets large. The choice of a precoding/detection technique will depend
on the computational resources, the number of antennas, the number of users, and the
diversity of the particular environment the system is in. For large antenna arrays where the
number of antennas is significantly greater than the number of users, the maximum ratio
approach may well be sufficient.

The Practical Obstacles Real-World Systems Present to Massive MIMO

When massive MIMO is implemented in a real-world scenario, there are further
practical considerations to be taken into account. As an example, consider an antenna
array with 32 transmit (Tx) and 32 receive (Rx) channels operating in the 3.5-GHz
band. There are 64 RF signal chains to be put in place, and the spacing between the
antennas is approximately 4.2 cm given the operating frequency. That’s a lot of hardware
to pack into a small space. It also means there is a lot of power being dissipated, which
brings inevitable temperature concerns. Analog
Devices’ integrated transceivers offer a highly
TBS RUE
effective solution to many of these issues. The
AD9371 will be discussed in more detail in the
next section.
Previously in this article, the application of
reciprocity to the system to drastically cut
the channel estimation and signal processing
overheads were discussed. Figure 10 shows
the downlink channel in a real-world system.
It is split into three components: the over-the-
Channel State air channel (H), the hardware response of the
M Antennas Information H K Users
base station transmit RF paths (TBS), and the
10. This figure shows a real-world downlink channel. hardware response of the user receive RF

☞ LEARN MORE @ mwrf.com | 20

 LIBRARY CHAPTER 5: Defining Massive MIMO in a 5G

paths (RUE).
The uplink is the opposite, as RBS characterizes the base station receive hardware
RF paths. TUE characterizes the user transmit hardware RF paths. While the reciprocity
assumption holds for the over-the-air interface, it does not for the hardware paths. The
RF signal chains introduce inaccuracies into the system due to mismatched traces, poor
synchronization between the RF paths, and temperature-related phase drift.
Using a common synchronized reference clock for all local-oscillator (LO) PLLs in the RF
paths, and synchronized SYSREFs for the baseband digital JESD204B signals will help
address latency concerns between the RF paths. However, there will still be some arbitrary
phase mismatch between the RF paths at system startup. Temperature-related phase drift
contributes further to this issue, and it is clear that calibration is required in the field when
the system is initialized and periodically thereafter. Calibration allows for the advantages
of reciprocity such as maintaining the signal processing complexity at the base station and
keeping the uplink only channel characterization. It can generally be simplified so that only
the base station RF paths (TBS and RBS) need to be considered.
There are a number of approaches one can chose from in calibrating these systems.
One is to use a reference antenna positioned carefully in front of the antenna array to
calibrate both the receive and transmit RF channels. It’s questionable whether having an
antenna placed in front of the array in this way is suited to practical base-station calibration
11. Shown is a block in the field. Another is to use mutual coupling between the existing antennas in the array
diagram of a 32-Tx, 32-Rx as the calibration mechanism. This may well be feasible.
massive MIMO radio head The most straightforward approach is probably to add passive coupling paths just before
that features the AD9371 the antennas in the base station. This adds more complexity in the hardware domain,
transceivers. but should provide a robust calibration mechanism. To fully calibrate the system, a signal
is sent from one designated calibration transmit channel, which
is received by all RF receive paths through the passive coupled
connection. Each transmit RF path then sends a signal in sequence
AD9371#1 that is picked up at the passive coupling point before each antenna,
relayed back to a combiner, and then to a designated calibration
receive path. Temperature related effects are generally slow to
change, so this calibration does not have to be performed very
frequently, unlike the channel characterization.

Digital
Baseband Analog Devices’ Transceivers and Massive MIMO
AD9371#15
Analog Devices’ range of integrated transceiver products are
particularly suited to applications where there is a high density
of RF signal chains required. The AD9371 features two transmit
paths, two receive paths, and an observation receiver, as well as
three fractional-N PLLs for RF LO generation in a 12-mm × 12-mm
AD9371#16 package. This level of integration enables manufacturers to create
complex systems in a timely and cost-effective manner.
A possible system implementation featuring multiple AD9371
SYSREF REFCLK
SYSREF REFCLK transceivers is shown in Fig. 11. This is a 32-transmit, 32-receive
Clock generation
system with 16 AD9371 transceivers. Three AD9528 clock generators
AD9528
AD9528 provide the PLL reference clocks and JESD204B SYSREFs to the
AD9528 32:1 Combiner/Splitter
system.

☞ LEARN MORE @ mwrf.com | 21

LIBRARY CHAPTER 5: Defining Massive MIMO in a 5G

The AD9528 is a two-stage PLL with 14 LVDS/HSTL outputs and an integrated

JESD204B SYSREF generator for multiple device synchronization. The AD9528s are
arranged in a fanout buffer configuration with one acting as a master device, with some of
its outputs used to drive the clock inputs and the SYSREF inputs of the slave devices. A
possible passive calibration mechanism is included—shown in green and orange—where
a dedicated transmit and receive channel are used to calibrate all the receive and transmit
signal paths through a splitter/combiner, as discussed in the previous section.

Conclusion
Massive MIMO spatial multiplexing has the potential to become a game-changing
technology in the cellular communications space, allowing for increased cellular capacity
and efficiency in high-traffic urban areas. The diversity that multipath propagation
introduces is exploited to allow for data transfer between a base station and multiple users
in the same time and frequency resource. Due to reciprocity of the channel between the
base station antennas and the users, all the signal processing complexity can be kept
at the base station, and the channel characterization can be done in the uplink. Analog
Devices’ RadioVerse family of integrated transceiver products allow for a high density of
RF paths in a small space, making them well suited for massive MIMO applications.

References
1. Xiang Gao, Massive MIMO in Real Propagation Environments, Lund University, 2016.
2. Michael Joham, Josef A. Nossek, and Wolfgang Utschick, “Linear Transmit Processing
in MIMO Communications Systems,” IEEE Transactions on Signal Processing, Vol. 53,
Issue 8, Aug. 2005.
3. Hien Quoc Ngo, Massive MIMO: Fundamentals and System Designs, Linköping
University, 2015.
to view this article online, ☞click here
☞ BACK TO TABLE OF CONTENTS

☞ LEARN MORE @ mwrf.com | 22

CHECK OUT THESE RESOURCES FROM
MICROWAVES & RF AND OUR SISTER BRANDS
WEBSITES

MAGAZINES RESOURCES
You’ll find a goldmine of resources on the mwrf.
You can also apply for a Inside Track with Smart Homes
THE SMART HOME BUTLER
The Real Smart IoT

A collection
of services

Boosting Public

com site including: Webcasts, white papers,

Automated Automated Open/close doors
temperature control grocery service and windows

Klaus Werner, RF Delivering Safety via

Turn on/ Smart medical Control home
off lights service appliances

Detect leaks and Turn on/off

turn off water entertainment system

Energy Alliance|24 Sustainability|33 Communications|45

subscription to one of our

impedance tuners

eBooks, RF Classics, What’s the Difference Between

traditional magazines available
march 2017 trusted eNgiNeeriNg resource for oVer 50 years www.mwrf.com

Impedance Tuners series and more. Check it out now at mwrf.com

in both print and digital formats. in control 100 p|

MICROWAVES & RF –
complimentary internationally- Corner Ad.pdf 1 4/19/2016 1:37:18 PM

☞ Subscribe Now C

NEWSLETTERS
Non-qualified: you will be prompted to
M

Y
march 2017 • Vol. 56 •, No. 3

FAST
CM

MY

pay based on your location. Stay current with

CY

CMY

$10.00 Powered by Periodicals Postage Paid • USPS 100 Approved Poly

newsletters created by
ELECTRONIC DESIGN –
engineers and editors with
Electronic Design

complimentary in US and Canada -

☞ Subscribe Now the authority on
emerging technologies
for design solutions
inside connections! You’ll
Non-qualified or Outside the US and find a coverage on a wide
APRIL 2017 electronicdesign.com

Canada: you will be prompted to pay build, buy, range of technologies for
or both?
consumer electronics

based on your location.

RF engineers including Test
•
$10 computing Price Point
•
trends in Poe

& Measurement, Defense

Consider these factors
when designing an
embedded system. p12

and more. ☞ Click Here

to check out what thousands of engineers are
reading now.
APRIL 2017 • Vol. 62, no. 11

$10.00 Periodicals Postage Paid • USPS 100 Approved Poly

161107_5Mill_ELECDES_US_Snipe.indd 1 11/7/16 11:33 AM

ABOUT US
A trusted industry resource since 1962, Microwaves & RF is the premier independent information
source for engineers specializing in high-frequency electronic design. We provide engineers with
“need-to-know” information and technical tutorials on the latest microwave technology, as well as
applications for design engineers. Our community of engineers span many markets – from
aerospace and defense to medical, automotive, Ics and semiconductors, aviation, computers,
broadcast and test equipment.

☞ LEARN MORE @ mwrf.com | 23