Shri Mata Vaishno Devi University: 5G Network Survey and Technologies

This essay presents a detailed survey of 5G network architecture and emerging technologies, focusing on the advancements needed to meet the increasing demands for capacity, data rates, and quality of service. The paper explores the evolution of wireless technologies from 1G to 5G, highlighting key technologies such as massive MIMO and device-to-device communication (D2D). It also examines emerging technologies including interference management, spectrum sharing, ultra-dense networks, and cloud technologies. The essay proposes a general 5G cellular network architecture and includes a review of current research projects. The survey emphasizes the challenges and facilitators of 5G, such as higher capacity, data rates, and reduced latency, along with the role of IEEE 802.11 standards. The paper provides a comprehensive overview of the progression of wireless technologies and their impact on modern communication, as well as the evolution of 5G cellular network architecture.

SPECIAL SECTION ON RECENT ADVANCES IN SOFTWARE DEFIN
NETWORKING FOR 5G NETWORKS
Received July 11, 2015, accepted July 22, 2015, date of publication July 28, 2015, date of current version August 7, 2015.
Digital Object Identifier 10.1109/ACCESS.2015.2461602
A Survey of 5G Network: Architecture and
Emerging Technologies
AKHIL GUPTA, (Student Member, IEEE), AND RAKESH KUMAR JHA, (Senior Member, IEEE)
School of Electronics and Communication Engineering, Shri Mata Vaishno Devi University, Katra 182320, India
Corresponding author: A. Gupta ([e-mail])
ABSTRACTIn the near future,i.e.,beyond 4G,some of the prime objectives or demands that need to
be addressed are increased capacity,improved data rate,decreased latency,and better quality of service.
To meet these demands, drastic improvements need to be made in cellular network architecture. This pape
presents the results of a detailed survey on the fifth generation (5G) cellular network architecture and som
of the key emerging technologies that are helpful in improving the architecture and meeting the demands
users. In this detailed survey, the prime focus is on the 5G cellular network architecture, massive multiple
input multiple output technology, and device-to-device communication (D2D). Along with this, some of the
emerging technologies that are addressed in this paper include interference management, spectrum sharin
with cognitive radio,ultra-dense networks,multi-radio access technology association,full duplex radios,
millimeter wave solutions for 5G cellular networks, and cloud technologies for 5G radio access networks
and software defined networks. In this paper, a general probable 5G cellular network architecture is propos
which shows that D2D, small cell access points, network cloud, and the Internet of Things can be a part of
5G cellular network architecture.A detailed survey is included regarding current research projects being
conducted in different countries by research groups and institutions that are working on 5G technologies.
INDEX TERMS 5G, cloud, D2D, massive MIMO, mm-wave, relay, small-cell.
I. INTRODUCTION
Today and in the recentfuture,to fulfillthe presumptions
and challenges of the near future,the wireless based net-
works of today will have to advance in various ways. Recent
technology constituent like high-speed packet access (HSPA)
and long-term evolution(LTE) will be launchedas a
segmentof the advancementof currentwirelessbased
technologies.Nevertheless,auxiliary components may also
constitute future new wireless based technologies,which
may adjunctthe evolved technologies.Specimen ofthese
new technology components are different ways of accessing
spectrum and considerably higherfrequency ranges,the
instigation of massive antenna configurations, direct device-
to-device communication, and ultra-dense deployments [1].
Since itsinitiation in the late 1970s,mobile wireless
communication has come across from analog voice calls to
currentmodern technologies adeptof providing high qual-
ity mobile broadband services with end-user data rates of
severalmegabits per second overwide areas and tens,or
even hundreds, of megabits per second locally. The extensive
improvements in terms of potentiality of mobile communica-
tion networks, along with the initiation of new types of mobile
devices such as smart phones and tablets, have produced an
eruption of new applications which will be used in cases
mobile connectivity and a resultantexponentialgrowth in
network traffic. This paper presents our view on the futu
wireless communication for 2020 and beyond. In this pa
we describe the key challenges that will be encountered
future wireless communication while enabling the netwo
society. Along with this, some technology routes that ma
taken to fulfill these challenges [1].
The imagination of our future is a networked society w
unbounded access to information and sharing of data wh
is accessible everywhere and every time for everyone an
everything. To realize this imagination, new technology
ponents need to be examined for the evolution of existin
wireless based technologies.Presentwireless based tech-
nologies, like the 3rd Generation Partnership Project (3G
LTE technology, HSPA and Wi-Fi, will be incorporating ne
technology components that will be helping to meet the
of the future. Nevertheless, there may be certain scenar
cannot be adequately addressed along with the evolutio
ongoing existing technologies. The instigation of comple
new wireless based technologies will complement the cu
technologies which are needed for the long term realiza
of the networked society [2].
1206
2169-3536 2015 IEEE. Translations and content mining are permitted for academic research only.
Personal use is also permitted, but republication/redistribution requires IEEE permission.
See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
VOLUME 3, 2015

A. Gupta, R. K. Jha:Survey of 5G Network:Architecture and Emerging Technologies
The remainderof the paperis organized asfollows:
In SectionII, we presentthe evolutionof wireless
technologies.Section IIIgives the detailed description of
the proposedgeneral5G cellular networkarchitecture.
Section IV comprisesof the detailed explanation ofthe
emerging technologies for 5G wireless networks.We con-
clude ourpaperin Section V.A list of currentresearch
projects based on 5G technologies is shown in the appendix.
II. EVOLUTION OF WIRELESS TECHNOLOGIES
G. Marconi,an Italian inventor,unlocksthe path of
recent day wireless communications by communicating the
letter ‘S’ along a distance of 3Km in the form of three dot
Morse code with the help of electromagnetic waves.After
this inception,wireless communications have become an
importantpartof presentday society.Since satellite com-
munication,television and radio transmission has advanced
to pervasive mobile telephone, wireless communications has
transformed the style in which society runs.The evolution
of wireless begins here [2] and is shown in Fig. 1. It shows
the evolving generations of wireless technologies in terms of
data rate,mobility,coverage and spectral efficiency.As the
wireless technologies are growing,the data rate,mobility,
coverage and spectralefficiency increases.It also shows
that the 1G and 2G technologies use circuit switching while
2.5G and 3G uses both circuitand packetswitching and
the nextgenerations from 3.5G to now i.e.5G are using
packetswitching.Along with these factors,it also differ-
entiate between licensed spectrum and unlicensed spectrum.
All the evolving generations use the licensed spectrum while
the WiFi,Bluetooth and WiMAX are using the unlicensed
spectrum.An overview aboutthe evolving wireless
technologies is below:
FIGURE 1.Evolution of wireless technologies.
A. 1G
The 1st generationwas announcedin initial 1980’s.
It has a data rate up to 2.4kbps.Major subscribers were
Advanced Mobile Phone System (AMPS),Nordic Mobile
Telephone(NMT), and Total Access Communication
System (TACS).It has a lotof disadvantages like below
parcapacity,reckless handoff,inferiorvoice associations,
and with no security, since voice calls were stored and p
in radio towers due to which vulnerability of these calls f
unwanted eavesdropping by third party increases [7].
B. 2G
The 2nd generationwas introducedin late 1990’s.
Digitaltechnology is used in 2nd generation mobile tele-
phones. Global Systems for Mobile communications (GSM
was the first2nd generation system,chiefly used for voice
communicationand havinga datarate up to 64kbps.
2G mobile handset battery lasts longer because of the r
signals having low power. It also provides services like S
Message Service (SMS) and e-mail. Vital eminent techno
gies were GSM,Code Division Multiple Access (CDMA),
and IS-95 [3], [7].
C. 2.5G
It generally subscribesa 2nd generation cellularsystem
merged with GeneralPacketRadio Services (GPRS)and
otheramenitiesdoesn’tcommonly endow in 2G or1G
networks.A 2.5G system generallyuses 2G system
frameworks,but it appliespacketswitching along with
circuit switching. It can assist data rate up to 144kbps. T
main 2.5G technologies were GPRS,Enhanced Data Rate
for GSM Evolution (EDGE),and Code Division Multiple
Access (CDMA) 2000 [3], [7].
D. 3G
The 3rd generation was established in late 2000.It imparts
transmission rateup to 2Mbps. Third generation (3G)
systems merge high speed mobile access to services ba
on InternetProtocol(IP). Aside from transmission rate,
unconventional improvement was made for maintaining
Additional amenities like global roaming and improved v
quality made 3G as a remarkable generation.The major
disadvantage for3G handsetsis that,they require more
power than most2G models.Along with this 3G network
plans are more expensivethan 2G [3], [7]. Since
3G involves the introduction and utilization ofWideband
Code Division Multiple Access (WCDMA), Universal
Mobile TelecommunicationsSystems(UMTS) and Code
Division Multiple Access (CDMA) 2000 technologies,the
evolving technologieslike High Speed Uplink/Downlink
Packet Access (HSUPA/HSDPA) and Evolution-Data
Optimized (EVDO)has madean intermediatewireless
generation between 3G and 4G named as 3.5G with imp
data rate of 5-30 Mbps [3].
E. 3.75G
Long-Term Evolution technology(LTE) and Fixed
Worldwide Interoperability for Microwave Access (WIMAX
is the future of mobile data services. LTE and Fixed WIM
has the potential to supplement the capacity of the netw
and provides a substantialnumber of users the facility to
access a broad range of high speed services like on dem
video, peer to peer file sharing and composite Web serv
VOLUME 3, 2015 1207

A. Gupta, R. K. Jha:Survey of 5G Network:Architecture and Emerging Technologies
Along with this,a supplementary spectrum is accessible
which accredit operators manage their network very compli-
antly and offers better coverage with improved performance
for less cost [4]–[7].
F. 4G
4G is generally referred as the descendant of the 3G and 2G
standards.3rd GenerationPartnershipProject (3GPP)
is presentlystandardizingLong Term Evolution(LTE)
Advanced as forthcoming 4G standard along with Mobile
Worldwide Interoperability for Microwave Access (WIMAX).
A 4G system improvesthe prevailingcommunication
networks by imparting a complete and reliable solution based
on IP. Amenities like voice,data and multimedia willbe
imparted to subscribers on every time and everywhere basis
and at quite higher data rates as related to earlier generations.
Applications thatare being made to use a 4G network are
MultimediaMessagingService(MMS), Digital Video
Broadcasting (DVB),and video chat,High Definition TV
content and mobile TV [2], [4]–[6].
G. 5G
With an exponentialincrease in the demand of the users,
4G will now be easily replacedwith 5G with an
advanced access technology named Beam Division Multiple
Access (BDMA) and Non- and quasi-orthogonalor Filter
Bank multicarrier(FBMC) multiple access.The concept
behind BDMA technique is explained by considering the case
of the base station communicating with the mobile stations.
In this communication,an orthogonalbeam is allocated to
each mobile station and BDMA technique willdivide that
antenna beam according to locations of the mobile stations
for giving multiple accesses to the mobile stations,which
correspondingly increase the capacity ofthe system [8].
An idea to shift towards 5G is based on current drifts,it is
commonly assumed that 5G cellular networks must address
six challenges that are not effectively addressed by 4G i.e.
higher capacity, higher data rate, lower End to End latency,
massive device connectivity,reduced costand consistent
Quality of Experienceprovisioning[22], [23]. These
challengesare conciselyshownin Fig. 2 along with
some potentialfacilitators to address them.An overview
of the challenges,facilitators,and corresponding design
fundamentalsfor 5G is shown in Fig.2 [20]. Recently
introduced IEEE 802.11ac, 802.11ad and 802.11af standards
are very helpfuland actas a building blocks in the road
towards 5G [9]–[13]. The technical comparison between these
standards is shown in table 1 and the detailed comparison of
wireless generations is shown in table 2.
III. 5G CELLULAR NETWORK ARCHITECTURE
To contemplate 5G network in the market now, it is evident
thatthe multipleaccesstechniquesin the network are
almost at a still and requires sudden improvement.Current
technologieslike OFDMA will work at leastfor next
50 years.Moreover,there is no need to have a change in
the wireless setup which had come aboutfrom 1G to 4G.
Alternatively,there could be only the addition of an appli-
cation or amelioration done atthe fundamentalnetwork to
please userrequirements.This will provoke the package
providers to drift for a 5G network as early as 4G is com
mercially setup [8].To meetthe demands of the user and
to overcome the challenges that has been put forward in
5G system,a drastic change in the strategy ofdesigning
the 5G wireless cellular architecture is needed.A general
observation of the researchers has shown in [14] that m
the wireless users stay inside for approximately 80 perc
time and outside for approximately 20 percent of the tim
In presentwireless cellular architecture,for a mobile user
to communicate whether inside or outside,an outside base
station present in the middle of a cell helps in communic
So for inside users to communicate with the outside bas
station,the signals will have to travel through the walls of
the indoors, and this will result in very high penetration
which correspondingly costs with reduced spectral effici
data rate, and energy efficiency of wireless communicat
To overcome this challenge,a new idea or designing tech-
niquethathas comein to existencefor scheming the
5G cellulararchitecture isto distinctoutside and inside
setups [8]. With this designing technique, the penetratio
through the walls of the building willbe slightly reduced.
This idea will be supported with the help of massive MIM
technology [15],in which geographically dispersed array
of antenna’s are deployed which have tens or hundreds
antenna units. Since present MIMO systems are using ei
two or four antennas, but the idea of massive MIMO syst
has come up with the idea of utilizing the advantages of
array antenna elements in terms of huge capacity gains
To build orconstructa large massive MIMO network,
firstly the outside base stations willbe fitted with large
antenna arrays and among them some are dispersed aro
the hexagonalcell and linked to the base station through
optical fiber cables, aided with massive MIMO technolog
The mobile users presentoutside are usually fitted with a
certain number of antenna units but with cooperation a
virtual antenna array can be constructed, which togethe
antenna arrays of base station form virtual massive MIM
links.Secondly,every building willbe installed with large
antenna arrays from outside,to communicate with outdoor
base stations with the help ofline of sightcomponents.
The wireless access points inside the building are conne
with the large antenna arrays through cables for commu
cating with indoorusers.This will significantly improves
the energy efficiency, cell average throughput, data rate
spectral efficiency of the cellular system but at the expe
of increased infrastructure cost.With the introduction of
such an architecture,the inside userswill only have to
connect or communicate with inside wireless access poi
while larger antenna arrays remained installed outside t
buildings [8].For indoor communication,certain technolo-
gies like WiFi, Small cell, ultra wideband, millimeter wav
communications [16], and visible light communications
1208 VOLUME 3, 2015

A. Gupta, R. K. Jha:Survey of 5G Network:Architecture and Emerging Technologies
FIGURE 2.5G challenge, facilitators, and design fundamental [20].
are useful for small range communications having large data
rates. But technologies like millimeter wave and visible light
communication are utilizing higher frequencies which are not
conventionally used for cellular communications.But it is
not an efficient idea to use these high frequency waves
outside and long distance applications because these wa
VOLUME 3, 2015 1209

A. Gupta, R. K. Jha:Survey of 5G Network:Architecture and Emerging Technologies
TABLE 1.Technical comparison between recent 802.11 standards.
will notinfiltrate from dense materials efficiently and can
easily be dispersed by rain droplets, gases, and flora. Though,
millimeter waves and visible light communications technolo-
gies can enhance the transmission data rate for indoor setups
because they have come up with large bandwidth.Along
with the introduction of new spectrum,which is notbeing
conventionally used for wireless communication, there is one
more method to solve the spectrum shortage problem by
improving the spectrum utilization of currentradio spectra
through cognitive radio (CR) networks [18].
Since the 5G cellular architecture is heterogeneous,so it
must include macrocells, microcells, small cells, and relays.
A mobile small cell concept is an integral part of 5G wireless
cellular network and partially comprises of mobile relay and
smallcell concepts [19].It is being introduced to putup
high mobility users,which are inside the automobiles and
high speed trains. Mobile small cells are positioned inside the
moving automobiles to communicate with the users inside
the automobile,while the massive MIMO unitconsisting
of large antenna arrays is placed outside the automobile to
communicate with the outside base station.According to
user’s opinion,a mobile smallcell is realized as a regular
base station and its allied users are all observed as a single
unit to the base station which provesthe above idea of
splitting indoorand outdoorsetups.Mobile small cell
users [19] have a high data rate for data rate services with
considerably reduced signaling overhead,as shown in [8].
As the 5G wireless cellularnetwork architecture consists
of only two logicallayers:a radio network and a network
cloud.Differenttypes of components performing different
functions are constituting the radio network.The network
function virtualization (NFV) cloud consists of a User plane
entity (UPE)and a Controlplane entity (CPE)thatper-
form higherlayerfunctionalities related to the Userand
Control plane, respectively. Special network functionalit
a service (XaaS) will provide service as per need,resource
pooling is one ofthe examples.XaaS is the connection
between a radio network and a network cloud [20].
The 5G cellular networkarchitectureis explained
in [8] and [20].It has equalimportance in terms of front
end and backhaulnetwork respectively.In this paper,a
general 5G cellular network architecture has been propo
as shownin Fig. 3. It describesthe interconnectivity
among the differentemerging technologieslike Massive
MIMO network, CognitiveRadio network,mobileand
static small-cellnetworks.This proposed architecture also
explains the role of network function virtualization (NFV)
cloud in the 5G cellular network architecture.The concept
of Device to Device (D2D) communication, small cell acc
points and Internet of things (IoT) has also been incorpo
in this proposed 5G cellular network architecture. In gen
this proposed 5G cellular network architecture may prov
a good platform for future 5G standardization network.
But there are several issues that need to be addressed
order to realize the wireless network architecture in part
ular,and 5G networks in general.Some of these issues are
summarized in Table. 3 [20].
IV. EMERGING TECHNOLOGIES FOR
5G WIRELESS NETWORKS
It is expected thatmobile and wireless traffic volume will
increase a thousand-fold overthe nextdecade which will
be driven by the expected 50 billion connected devices
nected to the cloud by 2020 and all need to access and
data, anywhere and anytime. With a rapid increase in th
ber of connected devices, some challenges appear whic
be responded by increasing capacity and by improving e
efficiency, cost and spectrum utilization as well as provi
1210 VOLUME 3, 2015

A. Gupta, R. K. Jha:Survey of 5G Network:Architecture and Emerging Technologies
TABLE 2.Evolution of wireless technologies.
better scalability for handling the increasing number of the
connected devices. For the vision of all-communicating world
relative to today’s network,the overalltechnicalaim is to
provide a system idea that supports [21]:
• 1000 times increased data volume per area
• 10 to 100 times increased number of connected devices
• 10 to 100 times increased typical user data rate
• 10 times extended battery life for low power Massive
Machine Communication (MMC) devices
• 5 times reduced End-to-End (E2E) latency
In this paper,we willcover a wide area of technologies
with a lotof technicalchallenges arises due to a variety
of applications and requirements of the user.To provide a
common connected platform for a variety of application
requirements for 5G, we will research the below technol
components [21]:
• Radio-links, includes the development of new trans
sion waveforms and new approaches of multiple acc
control and radio resource management.
• Multi-node and multi-antenna transmissions, in
designing of multi-antenna transmission/reception te
nologies based on massive antenna configurations a
developing advanced inter-node coordination schem
and multi-hop technologies.
VOLUME 3, 2015 1211

A. Gupta, R. K. Jha:Survey of 5G Network:Architecture and Emerging Technologies
FIGURE 3.A general 5G cellular network architecture.
• Network dimension, includes considering the demand,
traffic and mobility management, and novel approaches
for efficientinterferencemanagementin complex
heterogeneous deployments.
• Spectrum usage,includes consideringextended
spectrum band of operation, as well as operation in new
spectrum regimes to provide a complete system concept
for new spectrum regimes that carefully addresses the
needs of each usage scenario.
Now the topicswhich will integratea subsetof the
technology components and provides the solution of some of
the goals which are identified earlier are [21]:
• Device-to-Device(D2D)communicationsrefersto
directcommunication between devices allowing local
exchange of user plane traffic without going through a
network infrastructure.
• Massive Machine Communications (MMC) will form
the basis of the Internetof Things with a wide range
of application fields including the automotive industry,
public safety, emergencyservices and medical
field.
• Moving Networks(MN) will enhanceand extend
linking together potentially large populations of jointly
moving communication devices.
• Ultra-dense Networks (UDN) will be the main driver
whose goals are to increase capacity,increase energy
efficiency of radio links, and enable better exploitation
of under-utilized spectrum.
• Ultra-reliableNetworks(URN) will enablehigh
degrees of availability.
In this section, we identify several technologies, ranke
perceived importance, which will be crucial in future wir
standards.
A. MASSIVE MIMO
Massive MIMO is an evolving technology thathas been
upgraded from the current MIMO technology. The Massiv
MIMO system uses arrays of antenna containing few hun
antennas which are at the same time in one time, freque
slot serving many tens of user terminals. The main objec
of Massive MIMO technology is to extractall the benefits
of MIMO but on a larger scale. In general, massive MIMO
is an evolving technology ofNext generation networks,
which is energy efficient,robust,and secure and spectrum
efficient [24].
Massive MIMO depends on spatialmultiplexing,which
furtherdepends on the base station to have channelstate
information, both on the uplink as well as on the downlin
In case of downlink,it is noteasy,butin case of uplink,
it is easy,as the terminalssend pilots.On the basisof
pilots,the channelresponse of each terminalis estimated.
In conventionalMIMO systems,the base station sends the
pilotwaveforms to the terminals and based on these,the
terminal estimate the channel, quantize it and feedback
to the base station.This processis not viable formas-
sive MIMO systems,especially in high mobility conditions
because of two reasons. Firstly the downlink pilots from
base station mustbe orthogonalamong the antennas,due
to which the requirementof time,frequency slots for the
downlink pilots increases with the increase in the numbe
1212 VOLUME 3, 2015

A. Gupta, R. K. Jha:Survey of 5G Network:Architecture and Emerging Technologies
TABLE 3.Small cell setup options and concern [20].
of antennas. So Massive MIMO systems would now require
a large numberof similarslots as compared to the con-
ventionalMIMO system.Secondly,as the number of base
station antennas increases the number of the channelesti-
mates also increases for each terminal which in turn needed
hundred times more uplink slots to feedback the channel
responses to the base station. A general solution to this prob-
lem is to work in Time Division Duplexing (TDD) mode
and depend on the reciprocity amid the uplink and downlink
channels [25].
Massive MIMO technology depends on phase coherent
signals from allthe antennas atthe base station,butthe
computational processing of these signals is simple.Below
are certain positives of a massive MIMO system [24]:
1) MASSIVE MIMO HAS THE CAPABILITY THAT IT CAN
IMPROVE THE RADIATED ENERGY EFFICIENCY BY
100 TIMES AND AT THE SAME TIME, INCREASES
THE CAPACITY OF THE ORDER OF 10 OR MORE
The positiveof increasein capacity isbecauseof the
spatial multiplexing technique used in Massive
MIMO systems. Regarding the improvement in the radiated
energy efficiency, it is because of the increase in the number
of antennas,the energy can now be concentrated in small
regions in the space. It is based on the principle of coherent
superposition of wave fronts.After transmitting the shaped
signals from the antennas,the base station has no role to
play by confirming thatall the wave fronts thathave been
emitted from the antennas possibly will add constructive
the intended terminal’s locations and destructively elsew
Zero forcing isused to suppressthe remaining interfer-
ence between the terminals, but at the expense of incre
transmitted power [24].
The desirability of maximum ratio combining (MRC) is
more as related to Zero forcing (ZF) because of its com-
putationalease i.e.received signals are multiplied by their
conjugate channel responses and due to the reason that
executed in a dispersed mode, autonomously at every a
element. Though ZF also works equally well for an ortho
MIMO system which MRC normally does not.The main
reason behind the efficientuse of the MRC with massive
MIMO involving large number of base station antennas,
channel responses allied with different terminals tend to
almost orthogonal.
With the use ofMRC receiver,we are operating in a
noise restricted system.MRC in Massive MIMO system
will scale down the power to an extent possible deprived
really upsetting the overall spectral efficiency and multi
interference,butthe effects ofhardware deficiencies are
likely to be overcome by the thermal noise. But the inte
behind the overall10 times higherspectralefficiency as
compared to conventional MIMO is because 10 times mo
terminals are served concurrently in the same time freq
resource [26].
VOLUME 3, 2015 1213

A. Gupta, R. K. Jha:Survey of 5G Network:Architecture and Emerging Technologies
2) MASSIVE MIMO SYSTEMS CAN BE PUT TOGETHER
WITH THE HELP OF LOW POWER AND
LESS COSTLY COMPONENTS
Massive MIMO has come up with a changewith
respectto concept,schemesand execution.Massive
MIMO systems use hundreds of less expensive amplifiers in
respect to expensive ultra-linear 50 Watt amplifiers because
earlier are having an outputpower in the milliwattrange,
which is much betterthan the latterwhich are generally
being used in conventional systems.It is dissimilar to con-
ventional array schemes, as it will use only a little antenna’s
that are being fed from high power amplifiers but having a
notable impact.The most significant improvement is about
the removal of a large number of expensive and massive items
like large coaxial cables [24].
With the use of a large number of antennas in massive
MIMO technology the noise,fading and hardware deficits
will be averaged because signals from a large numberof
antennas are combined together in the free space. It condenses
the limits on precision and linearity of every single amplifier
and radio frequency chain and altogetherwhatmatters is
their collective action.This willincrease the robustness of
massive MIMO againstfading and failure ofone of the
antenna elements.
A massive MIMO system has degrees of freedom in excess.
For example,with 100 antennas,10 terminals are showing
presence while the remaining 90 degrees offreedom are
still available.These available degrees of freedom can be
exploited by using them for signalshaping which willbe
hardware friendly. Specifically, each antenna with the use of
very cheap and power proficient radio frequency amplifiers
can transmit signals having small peak to average ratio [27]
and constant envelope [28] at a modest price of increased total
radiated power. With the help of constant envelope multiuser
precoding,the signals transmitted from each antenna are
neither being formed in terms of beam nor by weighing of
a symbol. Rather, a wave field is created and sampled with
respectto the location ofthe terminals and they can see
precisely the signals whatwe intended to make them see.
Massive MIMO has a vital property which makes it possible.
The massive MIMO channel has large null spaces in which
nearly everything can be engaged withoutdisturbing the
terminals.Precisely modules can be placed into this null
spacethatmakesthe transmitted waveformsfulfill the
preferred envelope restraints.Nevertheless,the operative
channels amid the base station and every terminal,can be
proceeded without the involvement of PSK type modulation
and can take any signal constellation as input [24].
The considerable improvementin the energy efficiency
facilitates massive MIMO systems to work two steps of lower
magnitude than with existing technology on the total output
RF power. This is important because the cellular base stations
are consuming a lot of power and it is an area of concern.
In addition, if base stations that consume less power could be
driven by renewable resources like solar or wind and therefore
it is helpfulto deploy base stations to the places where
electricity is notavailable.Along with this,the increased
concerns of electromagnetic exposure willbe considerably
less.
3) MASSIVE MIMO PERMITS A SUBSTANTIAL DECREASE
IN LATENCY ON THE AIR INTERFACE
Latency is the prime area of concern in the next generat
networks.In wireless communication,the main cause of
latency is fading.This phenomenon occurs amid the base
station and terminal, i.e. when the signal is transmitted
the base station,it travels through differentmultiple paths
because of the phenomenon’s like scattering, reflection
diffraction before itreaches the terminal.When the signal
through these multiple paths reaches the terminal it will
fere either constructively or destructively, and the case
following waves from these multiple paths interfere dest
tively, the received signal strength reduces to a conside
low point. If the terminal is caught in a fading dip, then i
to wait for the transmission channel to change until any
can be received.Massive MIMO,due to a large number of
antennas and with the idea of beam forming can avoid f
dips and now latency cannot be further decreased [24].
4) MASSIVE MIMO MAKES THE MULTIPLE
ACCESS LAYER SIMPLE
With the arrivalof Massive MIMO,the channelstrength-
ens and now frequency domain scheduling is notenough.
OFDM provides, each subcarrier in a massive MIMO syst
with considerably the same channel gain due to which e
and every terminal can be provided with complete band
which reduces most of the physical layer control signalin
terminated [24].
5) MASSIVE MIMO INCREASES THE STRENGTH EQUALLY
AGAINST UNINTENDED MAN MADE INTERFERENCE
AND INTENDED JAMMING
Jammingof the wirelesssystemsof the civilian is a
prime area of concern and poses a serious threatto cyber
security.Owing to limited bandwidth,the distribution of
information overfrequency justis not possible.Massive
MIMO offers the methodsof improving robustnessof
wireless communications with the help of multiple anten
It provides with an excess of degrees of freedom thatcan
be useful for canceling the signals from intended jamme
If massive MIMO systems use joint channel estimation a
decoding instead ofuplink pilots forchannelestimation,
then the problem from the intended jammers is conside
reduced [24].
The advantagesof massiveMIMO systems can be
reviewed from an information theoreticpoint of view.
Massive MIMO systems can obtain the promising multi-
plexing gain ofmassive pointto pointMIMO systems,
while eliminating problems due to unfavorable propagat
environments [29].
1214 VOLUME 3, 2015

A. Gupta, R. K. Jha:Survey of 5G Network:Architecture and Emerging Technologies
Let us study a massive MIMO system having L cells, where
every cell has K attended single antenna users and one base
station with N antennas. hi,k ,l,nrepresent the channel coeffi-
cient from the k-th user in the l-th cell to the n-th antenna of
the i-th base station, which is equivalent to a complex small
scale fading factor time an amplitude factor that interprets for
geometric attenuation and large-scale fading:
hi,k ,l,n= gi,k ,l,n
p di,k ,l (1)
Where gi,k ,l,nand di,k ,lrepresent complex small scale fad-
ing and large scale fading coefficients, respectively. The small
scale fading coefficients are implicit to be diverse for diverse
users or for diverse antennas atevery base station though
the large scale fading coefficients are the same for diverse
antennas atthe same base station,butare user dependent.
Then, the channel matrix from all K users in the l-th cell to
the i-th base station can be expressed as
Hi,l =



hi,1,l,1 · · · hi,K ,l,1
... ... ...
hi,1,l,N · · · hi,K ,l,N


 = Gi,l D1/2
i,l (2)
Where
Gi,l =



gi,1,l,1 · · · gi,K ,l,1
... ... ...
gi,1,l,N · · · gi.K ,l,N


 (3)
Di,c =



di,1,l · · · . . .
... ... ...
. . . · · · di.K ,l


 (4)
Let us study a single cell (L = 1) massive MIMO system
with K singled antennausersand a basestation with
N antennas. For ease, the cell and the base station indices are
plunged when single cell systems are deliberated [29].
a: UPLINK
The received signal vector at a single base station for uplink
signal transmission is denoted as yu ∈ CN ∗1, can be stated as:
yu = √ ρuHxu + nu (5)
where xu ∈ C K ∗1is the signalvectorfrom allusers,
H ∈ C N∗K is the uplink channel matrix defined in (2) by
reducing the cell and the base station indices, nu ∈ CN ∗1is
a zero mean noise vector with complex Gaussian distribution
and identity covariance matrix, and ρu is the uplink transmit
power. The transmitted symbol from the k-th user, xu
k, is the
k-th element of xu = [xu
1, . . . ., xu
K ]T with [|xu
k|2] = 1.
The column channel vectors from diverse users are asymp-
totically orthogonalas the number of antennas atthe base
station,N, grows to infinity by supposing thatthe small
scale fading coefficients for diverse users is independent [30].
Then, we have
HH H = D1/2GH GD1/2 ≈ ND1/2IK D1/2 = ND (6)
An exhaustive debate about this result can be seen in
Centered on the result in (6), the overall achievable rate
users come to be
C = log2 det(I + ρuHH H )
≈ log2 det(I + N ρuD)
=
KX
k=1
log2 (1 + Nρudk)
bits
s
Hz (7)
Capacity in (7)can be achieved atthe base station by
simple MF processing.When MF processing is used,the
base station processes the signalvector by multiplying the
conjugate transpose of the channel, as
HH yu = HH √ ρuHxu + nu
≈ N√ ρuDxu + HH nu (8)
where (6) is used. Note that the channel vectors are a
totically orthogonal when the number of antennas at the
station grows to infinity.So,HH does not shade the noise.
Since D is a diagonal matrix,the MF processing splits the
signals from diverse users into diverse streams and ther
asymptotically no inter user interference. So now the sig
transmission can be treated as a SISO channel transmis
for each user. From (8), the signal to noise ratio (SNR) fo
k-th user is N ρudk. Subsequently, the attainable rate by usi
MF is similar as the limit in (7), which indicates that simp
MF processing at the base station is best when the num
antennas at the base station, N, grows to infinity.
b: DOWNLINK
yd ∈ CK ∗1can be denoted as the received signal vector a
all K users. Massive MIMO works properly in time division
duplexing (TDD) mode as discussed in [29], where the d
link channelis the transpose of the uplink channelmatrix.
Then, the received signal vector can be expressed as
yd = √ ρdHT xd + nd (9)
where xd ∈ CN ∗1is the signal vector transmitted by the
base station,nd ∈ CK ∗1is an additive noise and ρd is the
transmit power of the downlink. Let us assume, E[|xd|2] = 1
for normalizing transmitting power.
As discussed in [29], the base station usually has chan
state information equivalent to all users based on uplink
transmission. So, it is likely for the base station to do po
allocation for maximizing the sum transmission rate. The
capacity of the system with power allocation is [32]
C = max
p log2 det(IN + ρdHPHH )
≈ max
p log2 det(IK + ρdNPD)
bits
s
Hz (10)
where (6) is used and P is a positive diagonal matrix w
the power allocations (p1, . . . .., pk) as its diagonal elements
andP K
k=1pk = 1
If the MF precoder is used, the transmitted signal vect
xd = H ∗ D−1/2P1/2sd (11)
VOLUME 3, 2015 1215

A. Gupta, R. K. Jha:Survey of 5G Network:Architecture and Emerging Technologies
where sd ∈ CK ∗1is the source information vector. Then,
the received signal vector at all K users is
yd = √ ρdHT H∗D−1/2P1/2sd + nd
≈ √ ρdND1/2P1/2sd + nd (12)
where the second line of (12) is for the case when the
number of antennas at the base station, N, grows to infinity,
and (6) is used. Since P and D are both diagonal matrices so
the signal transmission from the base station to every user can
be treated as if initiating from a SISO transmission which thus
inhibited the inter user interference.The overallattainable
data rate in (12) can be maximized by proper choice of the
power allocation as in (10), which validates that the capacity
can be attained using the simple MF precoder.
According to the auspicious propagation assumption of (6),
the simple MF precoder or detector can attain the capacity of
a massive MIMO system when the number of antennas at the
base station, N, is much larger than the number of users, K,
and grows to infinity,i.e.,N K and N → ∞. Another
scenario assumption is that both the number of antennas at
the base station and the number of users grows large while
their ratio is bounded, i.e., N/K = c as N, K → ∞, where c
is a constant, are different [35].
The main area ofconcern in today’swirelesscellular
network is on energy efficiency and poweroptimization.
So a lot of researchers are working on to increase the energy
efficiency and optimizing the power.The work done on
poweroptimization in [33]has been realized and shown
in Fig. 4.
Fig. 4 clearly shows thatif we increase the number of
antennas at the base station as well as on the small cell access
point, the total power per subcarrier decreases to 10 fold as
compare to the case of no antenna at small cell access point.
FIGURE 4.Average total power consumption in the scenario containing
small cell access points.
However,there are saturation points where extra hardwar
will not decrease the total power anymore.
With the introduction of the concept of small cell acce
point, it will fulfill the need of self organizing network (SO
technologyfor minimizinghumaninterventionin the
networking processes as given in [36] and [37]. While a
summary of the work done on the massive MIMO techno
to increase the energy efficiency and optimizing the pow
the wireless cellular network is shown in Table 4.
B. INTERFERENCE MANAGEMENT
For efficientutilization oflimited resources,reuse is one
of the conceptthatis being used by many specifications
of cellularwirelesscommunication systems.Along with
this, for improved trafficcapacity and userthroughput
TABLE 4.Effect of massive MIMO technology on energy efficiency of the wireless cellular network.
1216 VOLUME 3, 2015

A. Gupta, R. K. Jha:Survey of 5G Network:Architecture and Emerging Technologies
densification of the network is one of the key aspect. So with
the introduction ofreuse and densification concept,there
will be an additional enhancement in terms of efficient load
sharing between macro cells and local access networks. But
all these advantages have come up with a problem that the
density and load of the network have increased considerably
and correspondingly receiver terminals in the network suffer
from increased co-channel interference, mainly at the bound-
aries of cells.Thus co-channelinterference poses a threat
which is inhibiting the further improvementof 4G cellular
systems.Hence the need for efficient interference
management schemes is vital. Below are the two interference
management techniques [38]:
1) ADVANCED RECEIVER
Modern day and growing cellular system, interference grow
as a big threat,so to mitigate or manage interference,an
appropriate interference management technique is the need of
the hour. Advanced interference management at the receiver,
or an advanced receiver is the technique which will somewhat
help in interference management.It will detectand even
try to decode the symbols of the interference signal within
the modulation constellation,coding scheme,channel,and
resource allocation.Then based on the detector output,the
interference signals can be reconstructed and cancelled from
the received signalso as to improve the anticipated signal
decoding performance [38].
Advanced receivers not only limits to inter cell interference
atthe cellboundaries,butalso intra cellinterference as in
the case ofmassive MIMO.According to LTE-Advanced
Release 10, every base station transmitter has been equipped
with up to eightantennaswhich will call for intra cell
interference, as the number of antenna’s increases. [38].
2) JOINT SCHEDULING
In LTE standard,Releases 8 and 9,interference random-
ization through scrambling oftransmitting signalsis the
only interferencemanagementstrategiesthatwerecon-
sidered and there were no advanced co-channelinterfer-
ence managementstrategies.But in 3GPP LTE-Advanced,
Release 10 and 11,through probability readings,it was
realized thatthere was a space for additionalperformance
improvement at the cell edges with the help of synchronized
transmission among multiple transmittersdispersed over
different cell sites [38].
For calibrating the development, some typical coordinated
multipoint schemes,like to coordinatescheduling,
coordinated beam forming, dynamic point selection, and joint
transmission, were normally conferred [38].
In the article [38], joint scheduling is broadly used to refer
advanced interference management of cellular systems and
link variation from the network side.But as in coordinated
multipointschemes,the transmission rates and schemes of
multiple cells are not autonomously determined. In the case
of fastnetwork distribution and interoperability,advanced
interference management schemes by joint scheduling from
the network side need to be stated in detail in the 5G sy
withoutseparating itentirely as an employmentissue.For
attracting maximum coordination,the user equipmentand
network side,advanced interference managementmustbe
deliberated instantaneously [38].
C. SPECTRUM SHARING
To apprehend theperformancetargetsof futuremobile
broadband systems [22], [39], there is a need of conside
more spectrum and wider bandwidths as compared to th
currentavailable spectrum forrealizing the performance.
So to overcome this difficulty, spectrum will be made av
able under horizontal or vertical spectrum sharing syste
The significanceof spectrum sharing isprobableto
increase,dedicated licensed spectrum access is expected
remain the baseline approach for mobile broadband whi
providesreliability and investmentcertainty forcellular
mobile broadband systems. Network components using
spectrum are likely to play a balancing role [40].
There are mainly two spectrum sharingtechniques
thatenable mobile broadband systems to share spectrum
and are classified as distributed solutions and
centralized solutions[40]. In a distributed solution the
systems coordinate amid each other on an equal basis w
in a centralized solution each system coordinates discre
with a central unit and the systems do not directly intera
with each other.
1) DISTRIBUTED SPECTRUM SHARING TECHNIQUES
Distributed spectrum sharing techniques is more efficien
it can take place in a localframework.Its principle is to
only manage those transmissions that really create inter
ence amid systems. Distributed coordination can be ent
included into standards and thus they can work without
need for commercial contracts between operators [40].
The managementof horizontalspectrum sharing
happens through the clear exchange of messages unsw
ingly between the sharing systems through a distinct int
in a peer to peer coexistence protocol. This protocol des
the performance ofthe nodes on the receiving ofcertain
messages or taking place of certain events.An example of
this is explained in [41].
The systemsfrequently transmitgenerally understood
signals that will show presence, activity factor and the t
when they will transmit in a coexistence beacon based s
tions. The information that is available openly can be us
the other systems to adjust their spectrum access perfo
for providing fairspectrum sharing.Coexistence beacons
are possibly the solution forboth,horizontaland vertical
sharing setups.An example ofits implementation is the
802.22.1 standard [42].
MAC behavior based schemes uses a MAC protocol wh
is designed to allow horizontal spectrum sharing. Blueto
using frequency hopping and WLAN systems using reque
to send/clear to send functionality are some of the exam
ples. For an even horizontal coexistence with Wi-Fi syste
a Wi-Fi coexistence mode is adapted. The MAC protocol
VOLUME 3, 2015 1217

A. Gupta, R. K. Jha:Survey of 5G Network:Architecture and Emerging Technologies
leave silent periods for Wi-Fi systems to operate and use a
listen before talk method which allows Wi-Fi systems to gain
access to the channel.
In Spectrum sensing and dynamic frequency selection,
operating frequency range is dynamically selected on the
basis of measurement results like energy detection or feature
detection. To detect the aforementioned coexistence beacons,
feature detection is highly useful.Due to a hidden node
problem, this method is not considered as a very dependable
method [40].
2) CENTRALIZED SPECTRUM SHARING TECHNIQUES
The Centralized spectrum sharing technique is useful for the
systems that have granularity of spectrum sharing on a higher
levelthan the actualradio resource allocation granularity.
This technique has some restraints, as it is conservative and
possibly separate users on orthogonal resources without com-
plete information on whether they would actually interfere or
not. While the benefits are in terms of reliability, certainty and
control.
Geo-locationdatabasemethodis an exampleof a
centralized sharing technique which involves the querying of
a database to obtain information about the resources available
at a particular location [43].This is the required classical
verticalsharing solution foraccessing the locally unused
TV bands [44].
The spectrum broker approach is one of the example of a
centralized sharing technique in which horizontally sharing
systems negotiate with a central resource management unit
for getting short term grants to use spectrum resources on a
limited basis [45].
Both the Geo-locationdatabaseand the spectrum
broker approach may additionally support horizontal sharing
between unlicensed systems [40].
However,along with the two above spectrum sharing
techniques most easily usable spectrum bands have also been
allocated, but various studies have revealed that these bands
are significantly underutilized.These concerns have driven
the researchers to innovate a new radio technology which
will encounter with the upcoming demands both in terms of
spectrum efficiency and performance ofcertain applica-
tions.To encounter the demand of the future,a disruptive
technology revolution that will empower the future wireless
world is Cognitive Radio.Cognitive radios are completely
programmable wireless devices and has an extensive adap-
tation property for achieving better network and application
performance. It can sense the environment and dynamically
performs adaptation in the networking protocols,spectrum
utilization methods,channelaccess methods and transmis-
sion waveform used. It is expected that cognitive radio tech-
nology will soon arise as a general purpose programmable
radio. Similar to the role of microprocessors in the computa-
tion, cognitive radio will also serve as a universal platform for
wireless system expansion. But the task of successfully build-
ing and large scale deployment of cognitive radio networks
to dynamically improve spectrum use is an intricate task.
It is an area of concern that the academic researchers a
industry in this area has reached a pointof fading returns.
Its future will now depend on the multi institutional rese
teams that are working on a new approach with real wor
experimental deployments of cognitive radio networks [
D. DEVICE TO DEVICE COMMUNICATION SYSTEM
Device to Device Communication system can be explain
by visualizing a two level5G cellular network and named
them as macro cell level and device level. The macro ce
comprises of the base station to device communications
an orthodox cellular system.The device level comprises of
device to device communications. If a device links the ce
network through a base station,then it will be operating in
the macro cell level and if a device links directly to anot
device or apprehends its transmission through the supp
other devices, then it will be on the device level. In thes
of systems, the base stations will persist to attend the d
as usual. But in the congested areas and at the cell edge
ad hoc mesh network is created and devices will be perm
to communicate with each other [47].
In the insightof device levelcommunications,the base
station either have fullor partialcontrolover the resource
allocation amid source,destination,and relaying devices,
or not have any control.Thus, we can describethe
subsequent four main types of device-level communicat
(Figs. 5-8) [47]:
FIGURE 5.Device relaying communication with base station controlled
link formation.
1) DEVICE RELAYING WITH BASE STATION
CONTROLLED LINK FORMATION
This type of communication is applicable for a device wh
is at the edge ofa cell, i.e. in the coverage area which
1218 VOLUME 3, 2015

A. Gupta, R. K. Jha:Survey of 5G Network:Architecture and Emerging Technologies
FIGURE 6.Direct device to device communication with base station
controlled link formation.
FIGURE 7.Device relaying communication with device controlled link
formation.
have poor signal strength. In this type of communication, the
devices will communicate with the base station by relaying
their information through other devices.
This type of communication willbe helpfulfor the
device to attain a higher quality of service and respective
increasedbatterylife. For partialor full controllink
formation,the base station communicates with the relaying
devices.
FIGURE 8.Direct device to device communication with device controlled
link formation.
2) DIRECT DEVICE TO DEVICE COMMUNICATION WITH
BASE STATION CONTROLLED LINK FORMATION
In this type of communication,the source and destination
devices are exchanging data with each otherwithoutthe
involvement of a base station, but they are supported b
base station for link formation.
3) DEVICE RELAYING WITH DEVICE
CONTROLLED LINK FORMATION
In this type ofcommunication,a base station isneither
involved in link formation nor for communication purpos
So, source and destination devices are totally responsib
synchronizing communication using relays amid each ot
4) DIRECT DEVICE TO DEVICE COMMUNICATION
WITH DEVICE CONTROLLED LINK FORMATION
In this type of communication,the source and destination
devices have direct communication with each other and
link formation is controlled itself by the devices without
assistance from the base station. Hence, the resource sh
be utilized by the source and destination devices in a wa
to certify limited interference with other devices in the s
level and the macro cell level.
For a substantialadvancementin excessof traditional
cellular system architecture, a dualistic cellular system s
be designed. For introducing the concept of device to de
communication, some technical issues needs to be addr
like security and interference management issues [47].
As in device to device communication, the routing of u
data is through the devices of the other users,so the main
area of concern is aboutsecurity because the privacy need
to be maintained.Closed access willensure their security
for the devices thatwantto operate in the device level.
VOLUME 3, 2015 1219

A. Gupta, R. K. Jha:Survey of 5G Network:Architecture and Emerging Technologies
In closed access, a device has a list of certain reliable devices,
like the users in the close vicinity or office to whom you are
familiar with, otherwise the users that have been legitimated
through a reliable party like an association, can unswervingly
communicate with each other, sustaining a level of discretion,
whereas the devices noton this listneed to use the macro
cell level to communicate with it. Also to prevent divulging
of theirinformation to otherdevices in a group,one can
set an appropriate encryption amongst one another.Instead
of this,in open access,each device can turn in to relay for
otherdevices deprived ofany limits.Meanwhile,in such
an instance security is an open research problem.Security
problems in device to device communication contain the
empathy of possible attacks,threats,and weakness of the
system. To discourse security problems in open access device
to device, the research on the security problems of machine
to machine communication [48]–[52] can be utilized.
Second technical issue of a dualistic system that need to be
addressed is of interference management. In device relaying
communication with the base station controller and direct
device to device communication with base station controlled,
the base station can execute the resource allocation and call
setup process.So, the base station,to a certain degree can
ease the problem of interference management by using cen-
tralized methods. But in device relaying communication with
device controller and direct device to device communication
with device controller,resource allocation between devices
will not be supervised by the centralized unit.Devices will
unavoidably effect macro cell users because they are working
in the same licensed band. So to confirm the nominal effect
on the performance of prevailing macro cell base stations, a
dualistic network needs to be considered that involves differ-
ent interference management techniques and resource alloca-
tion schemes. In addition to the interference amid the macro
cell and device levels, interference amid users at the device
level is also of prime concern.For performing the resource
allocation in this type of communication, different algorithms
as shown in table 4 and methods like resource pooling [53],
non-cooperative game [54] or bargaining game,admission
controland power allocation [55],cluster partitioning,and
relay selection [56] can be engaged.
In device relaying communication with the base station
controller,as shown in Fig.5, since the base station is one
of the communicating units,so the aforementioned chal-
lenges can be addressed with the help of the base station
like authenticating the relaying devices through encryption
for maintaining adequate privacy of the information of the
devices [57]. The challenge of spectrum allocation amid the
relaying devices to prevent them from interfering with other
devices will also be managed by the base station.
In direct device to device communication with base station
controlled, shown in Fig. 6, the devices communicate directly
with each other, but the base station controls the formation of
links between them. Precisely, the work of the base station is
to authenticate the access, control the connection formation,
resource allocation, and also deals with financial interaction
amid devices. Basically the base station has complete co
over the device to device connections, like connection s
and maintenance,and resource allocation.Since device to
device connections share the cellular licensed band in th
device level with the regular cellular connections in the
cell level. So for assigning resources to every device to d
connection,the network can either assign resources in an
identicalmanner as a regular cellular connection or in the
form of a dedicated resource poolto alldevices to device
connections [47].
In device relaying communication with device controll
and direct device to device communication with device
troller, there is no base station to control the communic
amid devices. As shown in Figs. 7 and 8, several devices
communicating with each other by using supportive or n
supportive communication by playing the role of relays
the other devices. Since there is no centralized supervis
the relaying, so distributed methods will be used for pro
like connection setup, interference management, and re
allocation. In this type of communication, two devices ne
to find each other and the neighboring relays first by pe
cally broadcasting their identity information. This will aw
the other devices of their presence and then they will de
whether or notto starta device to device director device
relaying communication [53].
Now to know the effect of relay’s,let us study a system
model for relay aided device to device communication [5
as shown in Fig.9. For studying it,let us considerthat
the cellular user equipmenteNodeB links are unfavorable
for direct communication and need the assistance of rel
The device to device user equipment’s are also supporte
the relay nodes due to long distance or poor link conditi
between peers.
FIGURE 9.A single cell with multiple relay nodes.
a: NETWORK MODEL
Let us consider a device to device enabled cellular netw
with multiple relays as shown in Fig.9. A relay node in
1220 VOLUME 3, 2015

A. Gupta, R. K. Jha:Survey of 5G Network:Architecture and Emerging Technologies
4G (LTE-Advanced) is connected between the radio access
network and both the cellular and devices to the device user
equipment’s through a donor eNodeB with a wireless connec-
tion. Let L = {1, 2, . . . , L} represents the set of fixed location
relays [57] in the network. The system bandwidth is divided
into N resource blocks denoted by N = {1, 2, . . . ., N }. Relay
node can be used for scheduling and resource allocation for
the device to device user equipment’s, when the link condition
between two devices to device user equipment’s is too poor
for direct communication. In addition, the direct communica-
tion between two devices to the device user equipment’s also
requires the aid of a relay node. Both cellular and device to
device user equipment’s assisted by relay ` are denoted by u`.
The set of user equipment’s assisted by relay ` is U` such that
U` ⊆ {C ∪ D}, ∀` ∈ L,
S
` U` = {C ∪ D}, and
T
` U` = ∅.
In the second step of communication,there could be mul-
tiple relays communicating to their related device to device
user equipment’s. According to our assumed system model,
relays are useful for scheduling and resource allocation for
the user equipment’s to reduce the computational load at the
eNodeB [58].
b: RADIO PROPAGATION MODEL
For realizingand exhibitingthe propagationchannel,
distance dependent path loss and shadow fading are consid-
ered and assumed that the channel is experiencing Rayleigh
fading.Particularly,3GPP propagation environmentpre-
sented in [60] is considered. For example, link between user
equipment and relay or between relays and device to device
follows the following path loss equation
PLul ,l (l )[dB] = 103.8 + 20.9 log (l) + Lsu+ 10 log (ζ )(13)
Where l is the distance between user equipment and relay
in kilometer,Lsu is interpreted asshadow fading and is
demonstrated as a log normal random variable,and ζ is an
exponentially distributed random variable which denotes the
rayleigh fading channel power gain. In the same way, the path
loss equation for the relay and eNodeB link is expressed as
PLl.eNodeB(l )[dB] = 100.7 + 23.5 log (l) + Lsr + 10 log (ζ )
(14)
Where Lsr is a log normal random variable accounting for
shadow fading.Hence,given the distance l,the link gain
between any pairof network nodes i, jcan be calculated
as 10−(PLi,j(l)/10).
c: REALIZABLE DATA RATE
h(n)
i,j can be denoted as the directlink gain between node
i and j overresource block n.The interference link gain
between relay (user equipment) i and a user equipment (relay)
j over resource block n is denoted by g
(n)
i,j where user equip-
ment (relay) j is not associated with relay (user equipment) i.
The unit power SINR for the link between user equipment
ul ∈ Ul and relay l using resource block n in the first hop is
given by
γ(n)
ul ,l,1 = h(n)
ul ,l
P
∀uj∈Uj,j6=l,j∈L P(n)
uj,jg(n)
uj,l + σ2 (15)
The unitpower SINR for the link between relay land
eNodeB for cellular user equipment ul (i.e.,ul ∈ {C ∩ Ul )
in the second hop is as follows:
γ(n)
l,ul ,2 = h(n)
l,eNodeB
P
∀uj∈{D∩Uj},j6=l,j∈L P(n)
j,uj g(n)
j,eNodeB+ σ2 (16)
In the same way, the unit power SINR for the link betw
relay l and receiving device to device user equipment fo
device to device user equipment’s ul (i.e., ul ∈{D ∩ Ul ) in the
second hop can be written as
γ(n)
l,ul ,2 = h(n)
l,ul
P
∀uj∈Uj,j6=l,j∈L P(n)
j,uj g(n)
j,ul + σ2 (17)
In (15)–(17), P
(n)
i,j is the transmit power in the link betwee
i and jover resource block n, σ2 = N0BRB, where BRB is
bandwidth of an resource block, and N0 denote thermal noise.
h(n)
l,eNodeBis the gain in the relay and eNodeB link and h
(n)
l,ul is
the gain in the link between relay land receiving device to
device user equipment corresponding to the device to d
transmitter user equipment’s ul .
The attainable data rate forul in the firsthop can be
expressed as
r(n)
ul ,1 = BRB log2 (1 + P
(n)
ul ,l γ(n)
ul ,l,1)
In the same way, the attainable data rate in the secon
is given by
r(n)
ul ,2 = BRB log2 (1 + P
(n)
l,ul γ(n)
l,ul ,2)
Since we are considering atwo hop communication
approach, the end to end data rate for ul on resource block n
is the halfof the minimum attainable data rate overtwo
hops, i.e.,
R(n)
ul = 1
2min{r
(n)
ul ,1, r(n)
ul ,2 (18)
The ongoing problem in device to device communicati
is aboutResource allocation.So a lot of researchers are
workingon to proposean optimalresourceallocation
algorithm.Table 5 willprovide a briefsummary on the
proposed algorithms.
E. ULTRA DENSE NETWORKS
To meet the increasing traffic demands due to the incre
number of users,densification of the infrastructure willbe
the prior aspectof 5G communications.But for achieving
ultra-dense,heterogeneous networks will play an importan
role.With the introduction of moving networks and ad-ho
social networks,the heterogeneous networks are becoming
more dynamic.Though dense and dynamic heterogeneous
VOLUME 3, 2015 1221

A. Gupta, R. K. Jha:Survey of 5G Network:Architecture and Emerging Technologies
TABLE 5.Summary of proposed algorithms for optimal resource allocation in device to device communication.
networkswill give rise to new challengesin termsof
interference,mobility and backhauling.To overcome these
challenges,there arisesa requirementof designing new
network layer functionalities for maximizing the performance
farther from the design of the existing physical layer.
In presentnetworkslike Long Term Evolution (LTE),
there exists interference mitigation techniques like enhanced
Inter-CellInterference Coordination and autonomous com-
ponentcarrier selection.But these techniques are applica-
ble only to nomadic and dense small cell deployments and
have limited flexibility. So for 5G networks, the interference
mitigation techniques should be more flexible and open to
the variations as changes in the traffic and deployment are
expected to occur more rapidly than existing networks [66].
With the introduction ofsmartwirelessdevices,the
interaction between these devices and with the environm
are destined to increase.To meetthe challenges thathave
arisen because of the increasing density of nodes and in
changing connectivity options,there arises a need ofthe
userindependentalgorithms.So future smartdevices are
designed in such a way thatwith the help of the context
information,they will learn and decide how to manage the
connectivity.Contextualinformation possibly willbe the
approaching service profile, battery position of a device
complete data acquired through either in built sensors, c
servers or serving base station. For example, to enable f
initialization of direct Device-to-Device communications
native multicast group making, context information abo
1222 VOLUME 3, 2015

A. Gupta, R. K. Jha:Survey of 5G Network:Architecture and Emerging Technologies
social networking will be very helpful as it will decrease the
signaling overhead in the network. Context information can
also provide sustenance for the network to decrease energy
consumption in base stations because of the switching of cells
by improvingthe mobility and traffic management
procedures and local handover strictures [66].
In short,future smartdevices and smallcell networks
will be capable of providing the bestwireless connectivity
with minimum interference and less powerconsumption.
Along with this, they shouldbe rapidlyadaptableto
the changing requirementsof devicesand radio access
network.
F. MULTI RADIO ACCESS TECHNOLOGY ASSOCIATION
As we are heading towards 5G, the networks are becoming
more heterogeneous. The main aspect that has attracted many,
is the integration among different radio access technologies.
A distinctive 5G aided device should be manufactured whose
radios notonly supporta new 5G standard like millimeter
wave frequencies, but also 3G, various releases of 4G LTE,
numerous types of WiFi, and possibly direct device to device
communication, all across the different spectral bands [67].
So, defining of standards and utilization of spectrum to which
base station or users willbe a really intricate job for the
network [68].
Defining ofthe optimaluserassociation isthe prime
area ofconcern which depends on the signalto interfer-
ence and noise ratio from every single user to every single
base station,the selections ofother users in the network,
the load atevery single base station,and the prerequisite
to apply the same base station and standard in both uplink
and downlinkfor simplifyingthe operationof control
channels forresource allocation and feedback [69],[70].
So, certain procedures mustbe implemented to overcome
these issues.
To increase edge rates by as much as 500%,a simple,
apparently highly suboptimalassociation method centered
on aggressive butstatic biasing towardssmallcells and
blanking about half of the macrocell transmissions has been
shown in [71]. The combined problem of user association and
resource allocation in two tier heterogeneous networks, with
adaptive tuning of the biasing and blanking in each cell,is
considered in [69], [70], and [72]–[77]. A model of hotspot
traffic shows that the optimal cell association is done by rate
ratio bias,instead of power level bias [73]–[75].An active
modelof cellrange extension as shown in [79],the traffic
arrives as a Poisson process in time and at the possible arrival
rates, for which a steadying scheduling policy subsists. With
massive MIMO atthe base stations,userassociation and
load balancing in a heterogeneous networks,is considered
in [79]. An exciting game theoretic approach is used in [80]
for the problem of radio access technology selection, in which
union to Nash equilibria and the Pareto-efficiency of these
equilibria are deliberated [67].
In conclusion, there is a vast scope for modeling, exploring
and optimizing base station-user associations in 5G [81].
G. FULL DUPLEX RADIOS
For a long duration of communication period, it is assum
the wireless system design that radios have to operate i
duplex mode.It means that it will not transmit and receive
simultaneously on the same channel.Many scholars,aca-
demics and researchers at different universities and rese
groups have tried to undermine this assumption by prop
many designs to build in-band full-duplex radios.
But the realization to build full duplex radio has a lot o
implications. The cellular networks will have to reduce th
spectrum demands to half as only a single channel is us
for achieving the same performance.As in LTE, for both
uplink and downlink,it uses equal width separate channels
for empowering radios to realize full duplex.
For communicating in the fullduplex mode,the self-
interference results from its own transmission to the rec
signal has to be completely removed. Let us consider th
of WiFi signals which are transmitting at 20dBm (100mW
average powerwith the noise floorof around −90dBm.
So the transmitself-interference need to be canceled by
110dB (20dBm-(−90dBm)) to achieve the similar level a
the noise floor and reduce it to insignificant. If any resid
self-interference is not completely canceled, then it will
as noise to the received signal,which in turn reduces SNR
and subsequently throughput [82].
H. A MILLIMETER WAVE SOLUTION
FOR 5G CELLULAR NETWORK
The Wireless industry has been growing day by day and
spite of the efforts by the industrial researchers for crea
the proficientwireless technologies,the wireless industry
continuously facing the overpowering capacity demands
its current technologies. Recent innovations in computin
communications and the arrival of smart handsets along
the need to access the internet poses new anxieties in f
of the wireless industry.These demands and anxieties will
grow in the approaching years for 4G LTE and indicates
at some point around 2020, there will arise a problem of
gestion in wireless networks. It will be must for the resea
industry to implement new technologies and architectur
meeting the increasing demands of the users.The ongoing
work plans a wireless future in which data rates increase
the multigigabitper second range.These high data rates
can be attainable with the help of steerable antennas an
millimeter wave spectrum and at the same time will sup
mobile communications and backhaul networks [83].
Recentresearcheshave put forwardthat mm-wave
frequencies of 2.6 GHz radio spectrum possibly willsup-
plement the presently saturated 700 MHz band for wirel
communications [84]. Feasibility of millimeter wave wire
communications is supported by the fact that the use of
gain,steerable antennas at the mobile and base station a
cost effective CMOS technology can now operate well in
the millimeter wave frequency bands [85]–[87]. Addition
with the use of millimeterwave carrierfrequencies,
VOLUME 3, 2015 1223

A. Gupta, R. K. Jha:Survey of 5G Network:Architecture and Emerging Technologies
larger bandwidth allocations will come up with higher data
transfer rates and service providers that are presently using
20 MHz channels for 4G customers willnow significantly
expand the channelbandwidths [87].With the increase in
bandwidth, capacity will also get increased, while the latency
will getdecreased,which give rise to better internetbased
access and applications like realtime streaming.Since the
wavelength of millimeter wave frequencies are very small,
so it will utilize polarization and different spatial processing
techniqueslike massiveMIMO and adaptivebeam-
forming [15].With the significantincrease in bandwidth,
the data links to densely populated areas willnow handle
greater capacity than present 4G networks. Likewise the base
stations are constantly reducing the coverage areas of the cell
for spatial reuse, cooperative MIMO, relays and interference
mitigation between base stations. Since the base stations are
abundant and more densely dispersed in urban areas, which
will reduce the cost per base station. Spectrum distributions
of over 1 GHz of bandwidth are currently being utilized in
the 28 GHz and 38 GHz bands.
By far as for the concern ofbuilding a prototype,the
antenna is essentially being positioned in very close vicinity
to the 28 GHz Radio Frequency Integrated Circuit and the
front end module because there will be high signal attenua-
tion at 28 GHz.Realizing the antenna array directly on the
printed circuit board of the 5G cellular device will minimize
the insertion loss between the antenna and Radio Frequency
Integrated Circuit.This infers thatan employmentof the
Radio Frequency blocks in the 5G architecture before the
intermediate frequency stage will be reliant on the placement
of the 28 GHz antenna array in the cellular phone.Taking
this concept into a thought,a minimum set of two 28 GHz
antenna arrays is proposed for millimeter wave 5G cellular
applications in [88],the two antenna arrays are employed
in the top and bottom partof the cellulardevice.The
28 GHz antenna array configuration for 5G cellular mobile
terminals and its comparison with the 4G standard is given
in table 6.
TABLE 6.28 GHz antenna array configuration for 5G cellular mobile
terminals and its comparison with the 4G standard.
The millimeter wave spectrum is under-utilized and is left
idle until present years.The main reason behind the under-
utilization is its unsuitability forcellularcommunications
because of unfriendly channel conditions like path loss e
absorption due to atmosphere and rain, small diffraction
penetration about obstacles and through objects respec
There is one more reason of unsuitability is due to stro
phase noise and excessive apparatus costs. But the prev
reason is that the large unlicensed band around 60 GHz
were appropriate primarily forvery shortrange transmis-
sion [90].So, the emphasis had been given to both fixed
wireless applications in the 28,38,71–76 and 81–86 GHz
and WiFi with the 802.11ad standard in the 60 GHz band
Semiconductors are also evolving,as their costs and power
consumption values are decreasing rapidly due to the gr
of the abovementioned short range standards. The main
agation issues regarding millimeter wave propagation fo
cellular communication are [67]:
1) PATH LOSS
The free space path loss is dependent on the carrier freq
as the size of the antennas is kept constant which is me
by the wavelength λ = c/fc, where fc is the carrier frequency.
Now as the carrier frequency increases, the size of the a
nas gotreduced and their effective aperture increases wi
the factor ofλ2
4π , while the free space path loss between a
transmitter and a receiver antenna grows with f2
c . So,if we
increase the carrier frequency fc from 3 to 30 GHz,it will
correspondingly add 20 dB of power loss irrespective of
transmitter-receiver distance. But for increased frequen
the antenna aperture at one end of the link is kept cons
then the free-space path loss remains unchanged. Addit
ally,if both the transmitter and receiver antenna apertur
are keptconstant,then the free space path loss decreases
with f2
c [67].
2) BLOCKING
Microwave signals are less prone to blockages butit dete-
riorates due to diffraction.In the contrary,millimeter wave
signals suffer less diffraction than the microwave signals
exhibit specular propagation, which makes them much m
vulnerable to blockages. This will fallout as nearly bimod
channelsubjectto the existence or lack of Line of Sight.
Recent studies in [84] and [91] reveals that, with the inc
in the transmitter and receiver distance the path loss inc
to 20 dB/decade under Line of sight propagation, but de
to 40 dB/decade plus an added blocking loss of 15–40 dB
non-line of sight [67].
So due to the presence of blockages, the set connectio
promptly shift from usable to unusable which will results
large scale impediments that cannot be avoided with ty
small scale diversity countermeasures.
3) ATMOSPHERIC AND RAIN ABSORPTION
Within the unlicensed 60-GHz band,the absorption due to
rain and air particularly the 15 dB/km oxygen absorption
more perceptible. But these absorptions are insignifican
the urban cellular deployments, where base station spac
1224 VOLUME 3, 2015

A. Gupta, R. K. Jha:Survey of 5G Network:Architecture and Emerging Technologies
mightbe on the order of 200 m.But actually,these type
of absorptions are usefulas itwill efficiently increase the
segregation of each cell by further attenuating the background
interference from more distant base stations [67].
So from the above explanation, it is inferred that the propa-
gation losses for millimeter wave frequencies are resolvable,
but only by steering the beam energy with the help of large
antenna arrays and then collect it coherently. But for practical
viability, the concept of narrow beam communication is fresh
for cellular communications and poses problems like:
a: LINK ACQUISITION
The main problem thatthe narrow beams are facing is in
establishing links amid users and base stations for both initial
access and handoff.The userand base stations willhave
to locate each other by scanning lots of angular positions
where the possibility of a narrow beam is high. This problem
poses an important research challenge predominantly in the
perspective of high mobility [67].
b: NEED OF NEW TRANSCEIVER ARCHITECTURES
Wirelessmillimeterwave systemshave gone through
significantimprovementbutstill there are some hardware
issues which will affect the designing of the communication
systems.The analog to digitaland digitalto analog con-
verters needed for large bandwidths are the prime cause of
power consumption. A prime reason of power consumption is
because of the use of large antenna arrays. Along with these,
high receiver sensitivities are needed to deal with the path loss
because it is not feasible that each antenna will be provided
with normal fully digital beam formers [67].
I. CLOUD TECHNOLOGIES FOR FLEXIBLE
5G RADIO ACCESS NETWORKS
1) MOBILE CLOUD COMPUTING
In the recentyears,mobile cloud computing has earned a
lot of admiration as itis a coalition ofmany computing
fields. It offers computing, storage, services, and applications
over the Internet.It also reduces cost,disconnectservices
from the existing technology, and offers flexibility in terms
of resource provisioning.So mobile cloud computing can
be defined as an incorporation ofcloud computing tech-
nology with mobile devices.This integration will make the
mobile devices resource full in terms of computational power,
memory, storage, energy, and context awareness [92]. Mobile
cloud computing can also be explained with differentcon-
cepts of the mobile cloud [93].
In the first method,let us consider that the other mobile
deviceswill also actas resourceprovidersas in [95].
So the combined resources of the numerous mobile devices
and other available stationary devices in the local area will
be exploited as shown in Fig.10. This method supports
user mobility and identifies the potential of mobile clouds to
perform collective sensing.
The cloudletconceptproposed in [96]is the second
method of mobile cloud computing. This method is explained
FIGURE 10.Virtual resource cloud made up of mobile devices in the
vicinity.
FIGURE 11.A cloudlet enabling mobile devices to bypass latency and
bandwidth issues while benefitting from its resources.
FIGURE 12.Flexible functional split [103].
in Fig. 11, where a local cloudlet encompassed by nume
multi core computers with connectivity to the remote clo
servers is used by the mobile device to relieve from its w
load. Plug Computers having form factor, diversity and l
power consumption can be considered as good contend
cloudlet servers. But these computers are ideal for smal
VOLUME 3, 2015 1225

A. Gupta, R. K. Jha:Survey of 5G Network:Architecture and Emerging Technologies
FIGURE 13.Characteristics of a radio access network as a service
implementation.
FIGURE 14.Number of Machine to Machine (M2M) connections in
mobile [105].
servers installed in the public organization because they have
the similar general architecture as a normal computer and are
less powerful, smaller, and less costly. Hence, these cloudlets
should be installed in public areas like restaurants so that
mobile devices can connect directly with the cloudlet instead
of a remote cloud server to remove latency and bandwidth
problems [93].
Mobile cloud computing follows the basic concepts of
cloud computing. There are some specific requirements that
need to be encountered in a cloud like adaptability, scalability,
availability and self-awareness as discussed in [94].
So mobile cloud computing should also fulfillthese
requirements.For example,a mobilecomputingcloud
should be cognizant of its availability and dynamically plug
themselves in, depending on the requirements and workload.
An appropriate technique of self pretentious one’s own qual-
ity is desirable for mobile users to proficiently take advantage
of the cloud, as the internal status and the external environ-
mentis subjectto change.Others facets like mobility,low
connectivity and limited source of power also needed to be
considered [93].
FIGURE 15.Machine to Machine traffic to increase 40-fold from
2010 to 2015.
FIGURE 16.Cell spectral efficiency in 5G networks [105].
FIGURE 17.Demand to delay in control and user planes for
4G/5G networks [105].
2) RADIO ACCESS NETWORK AS A SERVICE
Centralization is the prime objective of 5G mobile netwo
because processing and management will need to be fle
and adapted to the actualservice requirements.This will
lead to a compromise between the decentralized today’
work and fully centralized cloud radio access network. T
compromise is addressed by the radio access network a
1226 VOLUME 3, 2015

A. Gupta, R. K. Jha:Survey of 5G Network:Architecture and Emerging Technologies
TABLE 7.5G related activities in Europe [109].
service concept, which partly centralizes functionalities of the
radio access network depending on the needs and character-
istics of the network. The Radio access network as a service
is an application of the software as a service paradigm [97],
so every function may be packed and distributed in the
form of a service within a cloud platform.This willcause
increased data storage and processing capabilities,as pro-
vided by a cloud platform accommodated in data centers. The
design of radio access network as a service based on cloud
enables flexibility and adaptability from differentpercep-
tions. Recent advances in Cloud radio access network is given
in [98]–[102].
There is a flexible functionalsplitof the radio protocol
stacks as shown in Fig. 12 is present in the central element
of radio access network as a service between the central
radio access network as a service platform and the local radio
access points. With the introduction of this functional split,
degrees of freedom increases.
The left side demonstrates a traditional LTE employment
in which all functionalities up to admission/congestion con-
trol are locally employed at the base station.The right side
illustrates the cloud radio access network approach in which
only the radio front-end is locally employed, and all the rest
functionality is centralized.But radio access network as a
service does notfully centralize allradio access network
functionalities [103].
Functional split realization poses a serious challenge f
the radio access network.Theoretically,the functional split
occur on every protocol layer or on the interface amid e
layer.Presentarchitecture involves restraints on the func-
tions between discrete protocol layers.So with a restrained
backhaul, most of the radio protocol stack and radio res
management will accomplished locally, while functions w
less restrained requirements like bearer management a
balancing are placed in the radio access network as a se
platform.So when a high capacity backhaulis available,
lower-layer functions like PHY and MAC are shifted for a
higher degree of centralization into the radio access net
as a service platform [103].
The following listas shown in Fig.13 condenses major
characteristicsof a radio accessnetworkas a service
implementation similar to the basic characteristics of a c
computing platform and is explained in [103].
3) JOINT RADIO ACCESS NETWORK BACKHAUL OPERATION
The main reliability factor of 5G wireless networks is den
spread small cell layer which necessitates to be connect
the radio access network as a service platform.Though,
VOLUME 3, 2015 1227

A. Gupta, R. K. Jha:Survey of 5G Network:Architecture and Emerging Technologies
TABLE 8.5G related activities in America [109].
the need of deployment of small cells is in the places where
the line of sight centered microwave solutions are either hard
or too costly to deploy for backhaul. Hence, the need to con-
nect small cells at diverse locations made backhaul network a
critical part of the infrastructure. In particular, there is a
of flexible centralization for dynamic adaptation of netw
routes.The degree ofradio access network centralization
depends on available backhaul resources.
1228 VOLUME 3, 2015

A. Gupta, R. K. Jha:Survey of 5G Network:Architecture and Emerging Technologies
So there is a need of a refined transportnetwork design
that can convey the data headed towards the central unit free
of the degree of centralization. This is an important necessity
for maximum flexibility when the introduction of the new
functionalities to the network is taking place.
But the complicationsincreasesin routing and
classification ofdatapacketsaccording to theirquality
of service.On the otherhand,software defined network
provides quicker reaction to link/node letdowns, higher uti-
lization of the accessible resources,and faster deployment
of new updates with ease.These advantages have come up
with a centralized controlexample,which streamlines the
arrangement and management, but with increased computa-
tional efforts, as algorithmic complexity increases [103]. Also
for spectrum utilization,software defined radio (SDR) and
software defined networks (SDN) are the optimum solution
and the study in [104] revealed that the co-existence of SDR
and SDN is essential, and the optimal results can be attained
only by co-existence and joint compliments.
J. TRENDS AND QUALITY OF SERVICE
MANAGEMENT IN 5G
5G technologies are likely to appear in the market in 2020.
It is expected to significantly improve customers Quality of
Service in the context of increasing growth of data volume
in mobile networks and the growth of wireless devices with
variety ofservices provided.Some generaltrends related
to 5G can be explained in terms ofmachine to machine
traffic and number of machine to machine connections in
mobile [105].
Based on the projections as shown in Fig. 14, in 2018 the
number of machine to machine (M2M) connections in the
networks of mobile operators will surpass 15 billion [108],
which is 2 times more than the presentrate,and in 2022
mobile operators will have more than 26 billion machine to
machine connections.
At the same time the stake of machine to machine con-
nections of the totalnumber of connections in the mobile
operator’s networks willrise from the present5% to 15%
in 2018 and to 22% in 2022 [105].
A key trend relates to mobility, as broadband mobile usage,
with more than 2.4 billion users globally (as of June 2012)
is expected to be dominant over the coming years. For data
traffic and machine to machine communications, an expected
40-fold increase between 2010 and 2015 is shown in Fig. 15
and a 1000 fold increase is predicted over a decade.This
level of growth force the network operators to provide global
broadband access to all types of heterogeneous and modified
Internet based services and applications [105].
While the Quality of Service managementin 5G can be
realized in terms of cell spectral efficiency and latency.
Demand to the cell’s spectral efficiency in 5G networks for
diverse transmission channels are shown in Fig. 16. Increased
spectralefficiency of5G networks can be attained using
non-orthogonal access methods in radio access networks and
by using non-orthogonal signals [107]. Comparison of these
demands with the same demands to 4G networks shows
progress of spectral efficiency by 3-5 times [105].
Assessmentof demandsto delay in controland user
planes for signaling traffic and user traffic respectively is
shown in Fig.17.This figure depicts thatthe demands to
5G networkswill be twice more firm fortraffic in the
user plane and 10 times more firm in the subscriber traffi
plane [106].
V. CONCLUSION
In this paper,a detailed survey has been done on the per-
formance requirements of 5G wireless cellular communi
tion systems thathave been defined in terms of capacity,
data rate, spectral efficiency, latency, energy efficiency,
Quality of service.A 5G wireless network architecture has
been explained in this paper with massive MIMO techno
network function virtualization (NFV) cloud and device to
device communication.Certain short range communication
technologies, like WiFi, Small cell, Visible light communi
tion, and millimeter wave communication technologies,
been explained, which provides a promising future in ter
of better quality and increased data rate for inside users
at the equivalent time reduces the pressure from the ou
base stations.Some key emerging technologies have also
been discussed that can be used in 5G wireless systems
fulfill the probable performance desires, like massive MI
and Device to Device communication in particular and in
ference management, spectrum sharing with cognitive r
ultra dense networks,multiradio access technology,full
duplex radios,millimeter wave communication and Cloud
Technologies in general with radio access networks and
ware defined networks.This paper may be giving a good
platform to motivate the researchers for better outcome
different types of problems in next generation networks
APPENDIX
A list of current research projects based on 5G technolog
are given in Table 7, 8 and 9.
TABLE 9.5G related activities in Asia [109].
REFERENCES
[1] R. Baldemair et al., ‘‘Evolving wireless communications: Addressing
challenges and expectations of the future,’’ IEEE Veh.Technol.Mag.,
vol. 8, no. 1, pp. 24–30, Mar. 2013.
[2] T. Rappaport,WirelessCommunications:Principlesand Practice.
Englewood Cliffs, NJ, USA: Prentice-Hall, 1996.
[3] T. Halonen,J. Romero,and J.Melero,Eds.,GSM, GPRS and EDGE
Performance: Evolution Towards 3G/UMTS. New York, NY, USA: Wiley
2003.
VOLUME 3, 2015 1229

A. Gupta, R. K. Jha:Survey of 5G Network:Architecture and Emerging Technologies
[4] J. G. Andrews, A. Ghosh, and R. Muhamed, Fundamentals of WiMAX.
Englewood Cliffs, NJ, USA: Prentice-Hall, 2007.
[5] B. Furht and S. A. Ahson, Eds., Long Term Evolution: 3GPP LTE Radio
and Cellular Technology. Boca Raton, FL, USA: CRC Press, 2009, ch. 12,
pp. 441–443.
[6] S. Sesia,I. Toufik,and M.Baker,Eds.,LTE: The UMTS Long Term
Evolution. New York, NY, USA: Wiley, 2009.
[7] K. R. Santhi, V. K. Srivastava, G. SenthilKumaran, and A. Butare, ‘‘Goals
of true broad band’s wireless next wave (4G–5G),’’ in Proc. IEEE 58th
Veh. Technol. Conf., vol. 4. Oct. 2003, pp. 2317–2321.
[8] C.-X. Wang etal., ‘‘Cellular architecture and key technologies for 5G
wireless communication networks,’’ IEEE Commun. Mag., vol. 52, no. 2,
pp. 122–130, Feb. 2014.
[9] E. Perahia and R. Stacey,Next GenerationWirelessLANs:
Throughput,Robustness,and Reliabilityin 802.11n.Cambridge,
U.K.: Cambridge Univ. Press, 2008.
[10] E. H. Ong, J. Kneckt, O. Alanen, Z. Chang, T. Huovinen, and T. Nihtila,
‘‘IEEE 802.11ac:Enhancements forvery high throughputWLANs,’’
in Proc.IEEE 22nd Pers.Indoor Mobile Radio Commun.,Sep.2011,
pp. 849–853.
[11] E. Perahia and M.X. Gong,‘‘Gigabitwireless LANs:An overview
of IEEE 802.11ac and 802.11ad,’’ ACM SIGMOBILE Mobile Comput.
Commun. Rev., vol. 15, no. 3, pp. 23–33, Jul. 2011.
[12] E. Perahia,C. Cordeiro,M. Park,and L.L. Yang,‘‘IEEE 802.11ad:
Defining the nextgeneration multi-Gbps Wi-Fi,’’in Proc. 7th IEEE
Consum. Commun. Netw. Conf., Jan. 2010, pp. 1–5.
[13] A. B. Flores,R. Guerra,E. W. Knightly,P. Ecclesine,and S.Pandey,
‘‘IEEE 802.11af: A standard for TV white space spectrum sharing,’’ IEEE
Commun. Mag., vol. 51, no. 10, pp. 92–100, Oct. 2013.
[14] V. Chandrasekhar, J. G. Andrews, and A. Gatherer, ‘‘Femtocell networks:
A survey,’’ IEEE Commun. Mag., vol. 46, no. 9, pp. 59–67, Sep. 2008.
[15] F. Rusek et al., ‘‘Scaling up MIMO: Opportunities and challenges with
very large arrays,’’ IEEE Signal Process. Mag., vol. 30, no. 1, pp. 40–60,
Jan. 2013.
[16] A. Bleicher,‘‘Millimeterwavesmay be the future of5G phones,’’
Samsung’s millimeter-wave transceiver technology could enable ultrafast
mobile broadband by 2020, Jun. 2013
[17] H. Haas. (Aug. 2011). Wireless Data From Every Light Bulb. [Online].
Available: http://bit.ly/tedvlc
[18] X. Hong, C.-X. Wang, H.-H. Chen, and Y. Zhang, ‘‘Secondary spectrum
access networks,’’ IEEE Veh.Technol.Mag.,vol.4, no.2, pp.36–43,
Jun. 2009.
[19] F. Haideret al., ‘‘Spectralefficiency analysisof mobile Femtocell
based cellular systems,’’ in Proc. IEEE ICCT, Jinan, China, Sep. 2011,
pp. 347–351.
[20] P. Agyapong,M. Iwamura,D. Staehle,W. Kiess,and A.Benjebbour,
‘‘Design considerations for a 5G network architecture,’’ IEEE Commun.
Mag., vol. 52, no. 11, pp. 65–75, Nov. 2014.
[21] A. Osseiran et al., ‘‘Scenarios for 5G mobile and wireless communica-
tions: The vision of the METIS project,’’ IEEE Commun. Mag., vol. 52,
no. 5, pp. 26–35, May 2014.
[22] M. Fallgren et al., Scenarios, Requirements and KPIs for 5G Mobile and
Wireless System, document ICT-317669-METIS/D1.1, Apr. 2013.
[23] IndustryProposal for a Public Private Partnership(PPP) in
Horizon 2020 (Draft Version2.1), Horizon 2020 Advanced5G
NetworkInfrastructurefor the Future InternetPPP. [Online].
Available: http://www.networks-etp-eu/fileadmin/user_upload/Home/
draft-PPP-proposal.pdf
[24] E. G. Larsson,F. Tufvesson,O. Edfors,and T.L. Marzetta,‘‘Mas-
sive MIMO for nextgeneration wirelesssystems,’’IEEE Commun.
Mag.,vol. 52, no. 2, pp. 186–195,Feb.2014.[Online].Available:
http://arxiv.org/pdf/1304.6690v1.pdf
[25] J. Nam, J.-Y. Ahn, A. Adhikary, and G. Caire, ‘‘Joint spatial division and
multiplexing: Realizing massive MIMO gains with limited channel state
information,’’ in Proc. IEEE 46th Annu. Conf. Inf. Sci. Syst., Mar. 2012,
pp. 1–6.
[26] A. Pitarokoilis, S. K. Mohammed, and E. G. Larsson, ‘‘On the optimal-
ity of single-carrier transmission in large-scale antenna systems,’’ IEEE
Wireless Commun. Lett., vol. 1, no. 4, pp. 276–279, Aug. 2012.
[27] C. Studer and E. G. Larsson, ‘‘PAR-aware large-scale multi-user MIMO-
OFDM downlink,’’IEEE J. Sel. AreasCommun.,vol. 31, no. 2,
pp. 303–313, Feb. 2013.
[28] S. K. Mohammed and E.G. Larsson,‘‘Per-antenna constantenvelope
precoding for large multi-user MIMO systems,’’ IEEE Trans. Commun
vol. 61, no. 3, pp. 1059–1071, Mar. 2013.
[29] L. Lu, G. Y. Li, A. L. Swindlehurst, A. Ashikhmin, and R. Zhang, ‘‘An
overview ofmassive MIMO:Benefits and challenges,’’IEEE J. Sel.
Topics Signal Process., vol. 8, no. 5, pp. 742–758, Oct. 2014.
[30] T. L. Marzetta,‘‘Noncooperative cellular wireless with unlimited num-
bers of base station antennas,’’ IEEE Trans.Wireless Commun.,vol.9,
no. 11, pp. 3590–3600, Nov. 2010.
[31] M. Matthaiou,M. R. McKay,P. J. Smith,and J.A. Nossek,‘‘On the
condition number distribution of complex Wishart matrices,’’ IEEE T
Commun., vol. 58, no. 6, pp. 1705–1717, Jun. 2010.
[32] S. Vishwanath, N. Jindal, and A. Goldsmith, ‘‘Duality, achievable rate
and sum-rate capacity of Gaussian MIMO broadcastchannels,’’ IEEE
Trans. Inf. Theory, vol. 49, no. 10, pp. 2658–2668, Oct. 2003.
[33] E. Björnson, M. Kountouris, and M. Debbah, ‘‘Massive MIMO and sma
cells: Improving energy efficiency by optimal soft-cell coordination,’
Proc. 20th Int. Conf. Telecommun. (ICT), May 2013, pp. 1–5.
[34] E. Björnson, L. Sanguinetti, J. Hoydis, and M. Debbah, ‘‘Optimal desi
of energy-efficientmulti-userMIMO systems:Is massive MIMO the
answer?’’ IEEE Trans. Wireless Commun., vol. 14, no. 6, pp. 3059–30
Jun. 2015. [Online]. Available: http://arxiv.org/abs/1403.6150
[35] Y. Xu, G. Yue, and S. Mao, ‘‘User grouping for massive MIMO in FDD
systems: New design methods and analysis,’’ IEEE Access, vol. 2, no
pp. 947–959, Sep. 2014.
[36] M. Peng,D. Liang,Y. Wei, J.Li, and H.-H.Chen,‘‘Self-configuration
and self-optimization in LTE-advanced heterogeneous networks,’’ IE
Commun. Mag., vol. 51, no. 5, pp. 36–45, May 2013.
[37] M. Peng,C. Wang,J. Li, H. Xiang,and V.Lau,‘‘Recentadvances in
underlay heterogeneous networks: Interference control, resource al
tion, and self-organization,’’ IEEE Commun. Surveys Tuts., vol. 17, n
pp. 700–729, Secondquarter 2015.
[38] W. Nam, D. Bai, J. Lee, and I. Kang, ‘‘Advanced interference manage
mentfor 5G cellular networks,’’ IEEE Commun.Mag.,vol.52,no.5,
pp. 52–60, May 2014.
[39] 5G Radio Access—Research and Vision, Ericsson, Stockholm, Swede
Jun. 2013.
[40] T. Irnich, J. Kronander, Y. Selen, and G. Li, ‘‘Spectrum sharing scena
and resulting technical requirements for 5G systems,’’ in Proc. IEEE
PIMRC, London, U.K., Sep. 2013, pp. 127–132.
[41] M. Goldhammer,802.16h Main Concepts, documentIEEE
802.19-05/0051r0, 2005.
[42] G. J. Buchwald,S. L. Kuffner,L. M. Ecklund,M. Brown,and
E. H. Callaway,‘‘The design and operation of the IEEE 802.22.1 dis-
abling beacon for the protection of TV whitespace incumbents,’’ in P
3rd IEEE DySPAN, Oct. 2008, pp. 1–6.
[43] ECC REPORT 159,‘‘Technicaland OperationalRequirements for the
Possible Operation of Cognitive Radio Systems in the ‘White Spaces
of the Frequency Band 470–790 MHz,’’ CEPT,Cardiff,Wales,Tech.
Rep. 159, Jan. 2011.
[44] FCC. FCC12-36 Third Memorandum Opinion and Order in the Matter
of Unlicensed Operation in the TV Broadcast Bands (ET Docket No. 0
186) and Additional Spectrum for Unlicensed Devices Below 900 MH
and in the 3 GHz Band (ET Docket No. 02-380). Washington, DC, US
Federal Communications Commission, 2012.
[45] J. Huang,R. A. Berry,and M. L. Honig,‘‘Auction-based spectrum
sharing,’’ Mobile Netw. Appl., vol. 11, no. 3, pp. 405–418, Jun. 2006.
[46] P. Steenkiste,D. Sicker,G. Minden,and D. Raychaudhuri,‘‘Future
directions in cognitive radio network research,’’ in Proc. NSF Worksh
Rep., Mar. 2007.
[47] M. N. Tehrani,M. Uysal,and H.Yanikomeroglu,‘‘Device-to-device
communication in 5G cellular networks: Challenges, solutions, and f
directions,’’ IEEE Commun. Mag., vol. 52, no. 5, pp. 86–92, May 2014
[48] I. Cha,Y. Shah,A. U. Schmidt,A. Leicher,and M. V. Meyerstein,
‘‘Trust in M2M communication,’’ IEEE Veh. Technol. Mag., vol. 4, no.
pp. 69–75, Sep. 2009.
[49] J. Yue,C. Ma, H.Yu, and W. Zhou, ‘‘Secrecy-based access control for
device-to-device communication underlaying cellular networks,’’ IEE
Commun. Lett., vol. 17, no. 11, pp. 2068–2071, Nov. 2013.
[50] A. Perrig,J. Stankovic,and D.Wagner,‘‘Security in wireless sensor
networks,’’ Commun. ACM, vol. 47, no. 6, pp. 53–57, Jun. 2004.
[51] Y. Zhou,Y. Fang,and Y.Zhang,‘‘Securing wireless sensor networks:
A survey,’’IEEE Commun.Surveys Tuts.,vol. 10, no. 3, pp. 6–28,
Third Quarter 2008.
1230 VOLUME 3, 2015

A. Gupta, R. K. Jha:Survey of 5G Network:Architecture and Emerging Technologies
[52] R. Muraleedharan and L. Osadciw, ‘‘Jamming attack detection and coun-
termeasures in wireless sensor network using ant system,’’ in Proc. Symp.
Defense Secur., Orlando, FL, USA, Apr. 2006.
[53] G. Fodor etal., ‘‘Design aspects of network assisted device-to-device
communications,’’ IEEE Commun.Mag.,vol.50,no.3, pp.170–177,
Mar. 2012.
[54] E. Hossain,D. Niyato, and Z. Han, DynamicSpectrum Access
and Managementin CognitiveRadio Networks.Cambridge,U.K.:
Cambridge Univ. Press, 2009.
[55] D. Feng,L. Lu, Y. Yuan-Wu,G. Y. Li, G. Feng,and S.Li, ‘‘Device-
to-device communications underlaying cellular networks,’’ IEEE Trans.
Commun., vol. 61, no. 8, pp. 3541–3551, Aug. 2013.
[56] B. Zhou, H. Hu, S.-Q. Huang, and H.-H. Chen, ‘‘Intracluster device-to-
device relay algorithm with optimal resource utilization,’’ IEEE Trans.
Veh. Technol., vol. 62, no. 5, pp. 2315–2326, Jun. 2013.
[57] L. Lei, Z. Zhong,C. Lin, and X.Shen,‘‘Operator controlled device-
to-device communications in LTE-advanced networks,’’ IEEE Wireless
Commun., vol. 19, no. 3, pp. 96–104, Jun. 2012.
[58] Q. Ye, M. Al-Shalash,C. Caramanis,and J.G. Andrews,‘‘Distributed
resource allocation in device-to-device enhanced cellularnetworks,’’
IEEE Trans. Commun., vol. 63, no. 2, pp. 441–454, Feb. 2015.
[59] D. I. Kim, W. Choi, H. Seo, and B.-H. Kim, ‘‘Partial information relaying
and relaying in 3GPP LTE,’’ in Cooperative Cellular Wireless Networks.
Cambridge, U.K.: Cambridge Univ. Press, 2011.
[60] Y. Yuan,‘‘LTE-A relay scenariosand evaluation methodology,’’in
LTE-Advanced Relay Technology and Standardization. Berlin, Germany:
Springer-Verlag, 2013, pp. 9–38.
[61] G. Yu, L. Xu, D. Feng,R. Yin, G. Y. Li, and Y.Jiang,‘‘Jointmode
selection and resource allocation for device-to-device communications,’’
IEEE Trans. Commun., vol. 62, no. 11, pp. 3814–3824, Nov. 2014.
[62] L. Wang and H.Wu, ‘‘Fastpairing of device-to-device link underlay
for spectrum sharing with cellular users,’’ IEEE Commun. Lett., vol. 18,
no. 10, pp. 1803–1806, Oct. 2014.
[63] Z. Zhou, M. Dong, K. Ota, J. Wu, and T. Sato. (2014). ‘‘Energy efficiency
and spectral efficiency tradeoff in device-to-device (D2D) communica-
tions.’’ [Online]. Available: http://arxiv.org/abs/1407.1556
[64] N. Naderializadeh and A. S. Avestimehr, ‘‘ITLinQ: A new approach for
spectrum sharing in device-to-device communication systems,’’ IEEE J.
Sel. Areas Commun., vol. 32, no. 6, pp. 1139–1151, Jun. 2014.
[65] H. Chen,D. Wu, and Y. Cai, ‘‘Coalition formation game forgreen
resource management in D2D communications,’’ IEEE Commun.Lett.,
vol. 18, no. 8, pp. 1395–1398, Aug. 2014.
[66] T. Ihalainen etal., ‘‘Flexible scalable solutions fordense smallcell
networks,’’ in Proc. WWRF, vol. 30.Oulu, Finland, Apr. 2013.
[67] J. G. Andrews et al., ‘‘What will 5G be?’’ IEEE J. Sel. Areas Commun.,
vol. 32, no. 6, pp. 1065–1082, Jun. 2014.
[68] O. Galinina etal., ‘‘Capturing spatialrandomness ofheterogeneous
cellular/WLAN deployments with dynamic traffic,’’ IEEE J.Sel. Areas
Commun., vol. 32, no. 6, pp. 1083–1099, Jun. 2014.
[69] Q. Ye, B. Rong, Y. Chen, M. Al-Shalash,C. Caramanis,and
J. G. Andrews,‘‘User association for load balancing in heterogeneous
cellularnetworks,’’IEEE Trans. WirelessCommun.,vol. 12, no. 6,
pp. 2706–2716, Jun. 2013.
[70] S. Corroy, L. Falconetti, and R. Mathar, ‘‘Cell association in small het-
erogeneous networks: Downlink sum rate and min rate maximization,’’ in
Proc. IEEE Wireless Netw. Commun. Conf., Apr. 2012, pp. 888–892.
[71] S. Singh and J. G. Andrews, ‘‘Joint resource partitioning and offloading
in heterogeneous cellular networks,’’ IEEE Trans.Wireless Commun.,
vol. 13, no. 2, pp. 888–901, Feb. 2014.
[72] Q. Ye, M. Al-Shalash, C. Caramanis, and J. G. Andrews, ‘‘On/off macro-
cells and load balancing in heterogeneous cellular networks,’’ in Proc.
IEEE GLOBECOM, Atlanta, GA, USA, Dec. 2013, pp. 3814–3819.
[73] D. Fooladivanda and C. Rosenberg, ‘‘Joint resource allocation and user
association for heterogeneous wireless cellular networks,’’ IEEE Trans.
Commun., vol. 12, no. 1, pp. 248–257, Jan. 2013.
[74] A. Bedekar and R.Agrawal,‘‘Optimalmuting and load balancing for
eICIC,’’ in Proc. Int. Symp. Modeling Optim. Mobile, Ad Hoc, Wireless
Netw. (WiOpt), May 2013, pp. 280–287.
[75] S. Borst, S. Hanly, and P. Whiting, ‘‘Throughput utility optimization in
HetNets,’’ in Proc. IEEE Veh. Technol. Conf., Jun. 2013, pp. 1–5.
[76] S. Borst,S. Hanly,and P.Whiting,‘‘Optimalresource allocation in
HetNets,’’ in Proc. IEEE Int. Conf. Commun., Jun. 2013, pp. 5437–5441.
[77] S. Deb, P. Monogioudis, J. Miernik, and J. P. Seymour, ‘‘Algorithms fo
enhanced inter-cell interference coordination (eICIC) in LTE HetNets
IEEE/ACM Trans. Netw., vol. 22, no. 1, pp. 137–150, Feb. 2014.
[78] B. Soret, K. I. Pedersen, T. Kolding, H. Kroener, and I. Maniatis, ‘‘Fas
muting adaptation for LTE-A HetNets with remote radio heads,’’ in P
IEEE GLOBECOM, Dec. 2013, pp. 3790–3795.
[79] S. V. Hanly and P. A. Whiting, ‘‘On the capacity of HetNets,’’ in Proc.
ITA, Feb. 2014, pp. 1–9.
[80] D. Bethanabhotla, O. Y. Bursalioglu, H. C. Papadopoulos, and G. Cair
‘‘User association and load balancing for cellular massive MIMO,’’ in
Proc. ITA, Feb. 2014, pp. 1–10.
[81] T. A. Levanen,J. Pirskanen,T. Koskela,J. Talvitie,and M.Valkama,
‘‘Radio interface evolution towards 5G and enhanced local area com
nications,’’ IEEE Access, vol. 2, pp. 1005–1029, 2014.
[82] D. Bharadia,E. McMilin, and S.Katti,‘‘Full duplex radios,’’ in Proc.
ACM SIGCOMM Conf. SIGCOMM (SIGCOMM), 2013, pp. 375–386.
[83] T. S. Rappaport et al., ‘‘Millimeter wave mobile communications for
cellular: It will work!’’ IEEE Access, vol. 1, pp. 335–349, 2013.
[84] Z. Pi and F.Khan,‘‘An introduction to millimeter-wave mobile broad-
band systems,’’IEEE Commun.Mag.,vol. 49, no. 6, pp. 101–107,
Jun. 2011.
[85] F. Gutierrez,S. Agarwal, K. Parrish, and T. S. Rappaport,
‘‘On-chip integrated antenna structures in CMOS for 60 GHz WPAN
systems,’’ IEEE J.Sel.Areas Commun.,vol. 27,no. 8,pp.1367–1378,
Oct. 2009.
[86] T. S. Rappaport,E. Ben-Dor,J. N. Murdock,and Y.Qiao,‘‘38 GHz
and 60 GHz angle-dependentpropagation forcellular& peer-to-peer
wireless communications,’’ in Proc. IEEE Int. Conf. Commun., Jun. 20
pp. 4568–4573.
[87] T. S. Rappaport,J. N. Murdock,and F.Gutierrez,‘‘State of the artin
60-GHz integrated circuits and systems for wireless communications
Proc. IEEE, vol. 99, no. 8, pp. 1390–1436, Aug. 2011.
[88] W. Hong, K.-H. Baek, Y. Lee, Y. Kim, and S.-T. Ko, ‘‘Study and proto-
typing of practically large-scale mmWave antenna systems for 5G c
devices,’’ IEEE Commun. Mag., vol. 52, no. 9, pp. 63–69, Sep. 2014.
[89] Estimated Spectrum Bandwidth Requirements for the Future Develo
ment of IMT-2000 and IMT-Advanced, document Rep. M.2078, ITU-R
Geneva, Switzerland, 2006.
[90] T. Baykas et al., ‘‘IEEE 802.15.3c: The first IEEE wireless standard fo
data rates over 1 Gb/s,’’ IEEE Commun. Mag., vol. 49, no. 7, pp. 114
Jul. 2011.
[91] W. Roh et al., ‘‘Millimeter-wave beamforming as an enabling techno
for 5G cellular communications:Theoreticalfeasibility and prototype
results,’’ IEEE Commun. Mag., vol. 52, no. 2, pp. 106–113, Feb. 2014
[92] A. R. Khan,M. Othman,S. A. Madani,and S.U. Khan,‘‘A survey of
mobile cloud computing application models,’’ IEEE Commun.Surveys
Tuts., vol. 16, no. 1, pp. 393–413, First Quarter 2014.
[93] N. Fernando,S. W. Loke,and W.Rahayu,‘‘Mobile cloud computing:
A survey,’’ Future Generat.Comput.Syst.,vol.29,no.1, pp.84–106,
2012.
[94] L. Mei, W. K. Chan,and T.H. Tse,‘‘A tale of clouds: Paradigm com-
parisons and some thoughts on research issues,’’ in Proc.Asia-Pacific
Services Comput. Conf. (APSCC), Dec. 2008, pp. 464–469.
[95] E. E. Marinelli,‘‘Hyrax:Cloud computing on mobile devices using
MapReduce,’’ M.S. thesis, School Comput. Sci., Carnegie Mellon Univ
Pittsburgh, PA, USA, 2009.
[96] M. Satyanarayanan,P. Bahl,R. Caceres,and N.Davies,‘‘The case for
VM-based cloudlets in mobile computing,’’ IEEE Pervasive Comput.,
vol. 8, no. 4, pp. 14–23, Oct./Dec. 2009.
[97] P. Mell and T. Grance.(Sep.2011).The NIST definition ofcloud
computing.NIST. [Online].Available:http://csrc.nist.gov/publications/
nistpubs/800-145/SP800-145.pdf, accessed Aug. 2013.
[98] M. Peng,Y. Li, Z. Zhao,and C.Wang,‘‘System architecture and key
technologies for 5G heterogeneous cloud radio access networks,’’ IE
Netw., vol. 29, no. 2, pp. 6–14, Mar./Apr. 2015.
[99] M. Peng,X. Xie, Q. Hu, J. Zhang,and H.V. Poor,‘‘Contract-based
interference coordination in heterogeneous cloud radio access netw
IEEE J. Sel. Areas Commun., vol. 33, no. 6, pp. 1140–1153, Jun. 2015
[100] M. Peng,C. Wang,V. Lau, and H.V. Poor,‘‘Fronthaul-constrained
cloud radio access networks:Insights and challenges,’’ IEEE Wireless
Commun., vol. 22, no. 2, pp. 152–160, Apr. 2015.
[101] M. Peng,C. I, C. Tan,and C.Huang,‘‘IEEE ACCESS special section
editorial: Recent advances in cloud radio access networks,’’ IEEE Ac
vol. 2, pp. 1683–1685, Dec. 2014.
VOLUME 3, 2015 1231

A. Gupta, R. K. Jha:Survey of 5G Network:Architecture and Emerging Technologies
[102] M. Peng, Y. Li, J. Jiang, J. Li, and C. Wang, ‘‘Heterogeneous cloud radio
access networks: A new perspective for enhancing spectral and energy
efficiencies,’’IEEE Wireless Commun.,vol. 21, no. 6, pp. 126–135,
Dec. 2014.
[103] P. Rost et al., ‘‘Cloud technologies for flexible 5G radio access networks,’’
IEEE Commun. Mag., vol. 52, no. 5, pp. 68–76, May 2014.
[104] H.-H. Cho, C.-F. Lai, T. K. Shih, and H.-C. Chao, ‘‘Integration of SDR
and SDN for 5G,’’ IEEE Access, vol. 2, pp. 1196–1204, 2014.
[105] V. Tikhvinskiy and G. Bochechka, ‘‘Perspectives and quality of service
requirements in 5G networks,’’J. Telecommun.Inf. Technol.,no. 1,
pp. 23–26, 2015.
[106] Y. Park, 5G Vision and Requirements, 5G Forum, Korea, Feb. 2014.
[107] 5GNOW, (5th Generation Non-Orthogonal Waveforms for Asynchronous
Signaling) is a European collaborative Research Project Supported by
the European Commission Within FP7 ICT Call 8, 2015.
[108] M. Hatton,The Global M2M Market in 2013.London,U.K.: Machina
Research White Paper, Jan. 2013
[109] 4G America’s Summary of Global 5G Initiatives, A One-Time Overview
of Global 5G Initiatives as of the First Quarter of 2014, Jun. 2014.
[110] G. Wunder et al., ‘‘5GNOW: Non-orthogonal, asynchronous waveforms
for future mobile applications,’’ IEEE Commun.Mag.,vol. 52,no.2,
pp. 97–105, Feb. 2014.
[111] G. Wunder etal., ‘‘5GNOW:Challenging the LTE design paradigms
of orthogonality and synchronicity,’’ in Proc.IEEE 77th Veh.Technol.
Conf. (VTC Spring), Jun. 2013, pp. 1–5.
[112] K. Samdanis,A. Ripke,and D. Kutscher,‘‘Enhancing userQoE in
multi-carrier LTE dense networks via multi-path support,’’ in Proc. Eur.
Wireless, 20th Eur. Wireless Conf., May 2014, pp. 1–5.
[113] H. Tullberg et al., ‘‘Towards the METIS 5G concept: First view on hor-
izontal topics concepts,’’ in Proc. Eur. Conf. Netw. Commun. (EuCNC),
Jun. 2014, pp. 1–5.
[114] J. F. Monserratet al., ‘‘Rethinking the mobile and wireless network
architecture: The METIS research into 5G,’’ in Proc.Eur.Conf.Netw.
Commun. (EuCNC), Jun. 2014, pp. 1–5.
[115] J.-C. Belfiore, ‘‘Codes for wireless wiretap channels,’’ in Proc. IEEE Inf.
Theory Workshop (ITW), Nov. 2014, pp. 307–308.
[116] M. Olsson,C. Cavdar,P. Frenger,S. Tombaz,D. Sabella,and
R. Jantti, ‘‘5GrEEn: Towards Green 5G mobile networks,’’ in Proc. 9th
Int. Conf. Wireless Mobile Comput., Netw. Commun. (WiMob), Oct. 2013,
pp. 212–216.
AKHIL GUPTA (S’15) received the
B.E. degree in electronicsand communication
engineering from Jammu University,Jammu and
Kashmir, India, in 2010, and the M.Tech. degree in
electronics and communication engineering from
the Jaypee University of Information Technology,
Waknaghat, India, in 2013. He is currently pursu-
ing the Ph.D. degree in electronics and communi-
cation engineering with Shri Mata Vaishno Devi
University, Jammu and Kashmir.
He is currently involved in research work on massive MIMO and device-
to-device communication. He is also working on the security issues of next-
generation networks,and OPNET simulation,MATLAB, and NS3 tools
for wireless communication.His research interests include the emerging
technologies of 5G wireless communication network.
Mr. Gupta received the Teaching Assistantship at the Ministry of Human
Resource Development from 2011 to 2013. He is a member of the Interna-
tional Association of Engineers and the Universal Association of Computer
and Electronics Engineers.
RAKESH KUMAR JHA (S’10–M’13–SM–15)
received the B.Tech.degree in electronicsand
communication engineering in Bhopal, India, the
M.Tech. degree from NIT Jalandhar, India, and the
Ph.D. degree from NIT Surat,India,in 2013.
He is currently an AssistantProfessorwith the
School of Electronics and Communication Engi-
neering,Shri Mata VaishnoDevi University,
Jammu and Kashmir,India.He is carrying out
his research on wireless communication,power
optimizations, wireless security issues, and optical communications.
He has authored over30 internationaljournalpapers and more than
20 international conference papers. His area of interest is wireless com
cation, optical fiber communication, computer networks, and security i
Dr. Jha’s concept related to router of wireless communication has bee
accepted by the InternationalTelecommunication Union (ITU)in 2010.
He received the Young Scientist Author Award from ITU in 2010, the AP
Fellowship in 2011 and 2012, and the Student Travel Grant from COMS
in 2012. He is a Senior Member of the Global ICT Standardization Forum
India,Society forIndustrialand Applied Mathematics,the International
Association of Engineers, and the Advance Computing and Communica
Society.
1232 VOLUME 3, 2015