See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/342455193

The Evolution of Radio Access Network Towards Open-RAN: Challenges and

Opportunities

Conference Paper · April 2020

DOI: 10.1109/WCNCW48565.2020.9124820

CITATIONS READS
143 4,606

3 authors:

Sameer Kumar Singh Rohit Singh

Indian Institute of Technology Ropar Indian Institute of Technology Ropar
9 PUBLICATIONS 158 CITATIONS 28 PUBLICATIONS 367 CITATIONS

SEE PROFILE SEE PROFILE

Brijesh Kumbhani
Indian Institute of Technology Ropar
57 PUBLICATIONS 424 CITATIONS

SEE PROFILE

All content following this page was uploaded by Sameer Kumar Singh on 16 March 2021.

The user has requested enhancement of the downloaded file.

 The Evolution of Radio Access Network Towards
Open-RAN: Challenges And Opportunities
Sameer Kumar Singh, Rohit Singh and Brijesh Kumbhani
Electrical Engineering Department
Indian Institute of Technology Ropar, India
{2018eey0007, 2017eez0007, brijesh} @iitrpr.ac.in

Abstract—The coexistence of massive Internet of Things (IoT)

network and modern technologies (e.g., high speed gaming and
self driving vehicles) requires a versatile network which can
provide support to all such applications. Since the Quality of
Service (QoS) requirement of each application is different from
one another, the existing Radio Access Network(RAN) is unable
to support such diverse applications. Consequently, Open Radio
Access Network(O-RAN) is being considered as the most viable
solution for next generation RAN. In this paper, we present
the evolution of RAN along with the possible architecture and
features of the most promising next generation RAN (i.e., O-
RAN). This work mainly discusses architectural and functional
Fig. 1. QoS requirement of various set of applications.
advancement of the RAN in each generation. In addition, we dis-
cuss various challenges associated with O-RAN implementation
and possible opportunities created with the advent by O-RAN.
Index Terms—Open-RAN, IoT, Cloud-RAN software oriented infrastructure which enables the network
to behave differently according to the QoS requirement of
I. I NTRODUCTION the processed application. From the market view point, O-
RAN creates a chance for the small vendors and operators to
In the past fifty years wireless communication technology start their own services and to increase their market revenue
has gone through several transformations [1]- [7]. Specifically, [28]- [30]. Though the advent of O-RAN may provide several
past few decades have witnessed a remarkable growth in wire- benefits, there are a number of challenges associated with it.
less communication framework due to the advent of massive Some of those challenges are as follow:
IoT and modern real time applications such as high speed
• Due to different QoS requirements, it is difficult to design
video gaming, self driving vehicles, etc [8]- [9]. However, as
depicted in Fig. 1, the QoS requirement of each application a stand alone service oriented architecture.
• The network should be flexible to support further upgra-
is different from the other. For instance, connected vehicles
demand high speed communication with high degree of relia- dations and must be compatible with the existing devices.
• The network should not exert extreme burden on the
bility [8]. On the other hand, some applications (e.g., IoT) seek
low throughput requirement but excellent coverage with low backhaul and must posses low computational complexity.
power consumption [9]. In contrast, some applications require In this paper, we present the evolution of RAN along with the
low latency and real time data processing. Consequently, the possible architecture and features of the most promising next
co-existence of such diverse variety of applications require a generation RAN (i.e., O-RAN). This work mainly discusses
versatile network which posses all features. Unfortunately, all architectural and functional advancement of the RAN in each
these targets cannot be achieved by existing/previous RAN generation. In addition, we discuss the challenges associated
which creates the demand of network up-gradation. One way to O-RAN implementation and possible opportunities created
to support such connectivity is to design separate network with the advent of O-RAN.
for different set of applications (shown in Fig. 1 ). However,
II. RAN OVERVIEW AND E VOLUTION
it is not a feasible solution from economics and operators
respectives. As a result, both the academia and industries RAN is the major part of the wireless communication
are trying to make the mobile network more software driven, system as it connects the user equipment(UE) to the core
virtualized, flexible, intelligent and energy efficient [10]- [11]. network by radio connectivity [12]- [13] as shown in Fig.
Moreover, the network has to be cost efficient and reliable. 2 . The basic functionality of RAN is to manage the radio
Another possible way to fulfil all the given requirements is resources [12]- [13]. Thus, typical RAN involves two major
to split the RAN into various parts based on the functionality. unit namely Radio Unit(RU) and Processing Unit(PU) as
Splitting can make the architecture smarter and versatile. shown in Fig. 2 .
This new architecture is known as Open RAN (O-RAN). • Radio Unit: Radio unit contains transceiver antennas and
Specifically, the advent of O-RAN is a step towards the it is responsible for transmission and reception.

‹,(((
Authorized licensed use limited to: Indian Institute of Technology (Ropar). Downloaded on August 02,2020 at 05:47:10 UTC from IEEE Xplore. Restrictions apply.
 TABLE I
A BBREVIATIONS USED IN THE PAPER .

AI Artificial Intelligence
BS Base Station
CoMP Coordinated Multipoint
CP Central Processor
CRAN Cloud Radio Access Network
CU Control Unit
DU Distributed Unit
G Generation
IOT Internet of Things
IP Internet Protocol Fig. 3. Different generation of RAN condition
MAC Medium Access Control
MEC Mobile Edge Computing
ML Machine Learning demand of millimeter wave (mm-wave) has been initiated
MIMO Multi Input Multi Output
MME Mobility Management Entity which subsequently raise the demand of connected framework
NVF Network Function Virtualization given in Fig. 3(c). The scenario given in Fig. 3(c) is also
OA Orchestration and Automation referred as Cloud Radio Access Network (CRAN) in which
O-RAN Open Radio Access Network
PUs of all the BSs are pooled to a standalone CP, formally
PDCP Packet Data Convergence Protocol
PU Processing Unit known as cloud processor [14]- [15].
QoS Quality of Service
RAN Radio Access Network A. Key Advancements in RAN
RIC Radio Intelligent Controller
RLC Radio Link Control Some key advancements that have been occurred over the
RRC Radio Resource Controller time in the previous/existing RAN are:
RT Real Time 1) BS centric to UE centric: Traditional RAN used to
RU Radio Unit
associate a BS to a UE on the basis of received signal
SDAP Service Data Adaptation Protocol
UE User Equipment strength from various BSs (the dominated one selected). This
sort of BS selection suffers from the fact that the interfering
power received by the cell edge users is usually comparable
to the power from serving BS. While, in the UE centric
approach, each UE is allowed to choose multiple BSs based
on the received signal strength and all these BSs works in
a coordinated manner to use/avoid the signal from adjacent
Fig. 2. An illustration of basic RAN BSs. In this way, UE centric approach provides interference
free scenario irrespective of user’s position [16].
2) mmWave and beamforming: As discussed, BS densifi-
• Processing Unit: Processing unit of RAN is responsible cation alone could not sufficient to fulfil the ever increasing
for radio management, resource utilization/sharing and data rate requirement sufficiently. High data rate also need
some other operations like (pre-coding, encryption ,etc.). an extra resources in terms of bandwidth. mmWave spectrum
Fig. 3 shows evolution of RAN over the time. Initially, the being relatively unoccupied, so it would be a best solution [17].
number of users as well as data rate requirement was very Though, the incorporation of mmWave has provided several
less. Due to the availability of some data restricted cellular fold increment in the available spectrum, it was challenging to
services (e.g., voice call, text messages, etc.), small number establish a reliable communication over mmWave due to high
of Base Stations(BSs) were sufficient to fulfil this demand. As attenuation and poor diffraction. As a result, multiple antennas
shown in Fig. 3(a), traditional RANs were equipped with the can be used at the RUs to form beams in the direction of the
integrated the RU and PU. Each BS was sufficient to cover the intended user [18].
significantly large area. Since the frequency reuse framework 3) Single point to multi point transmission: The traditional
was adopted, very less/no computation was required for inter- networks used BS centric approach to connect each user.
ference avoidance. Later, RU and Distributed Unit(DU) were Consequently, the edge user suffered from severe inter-cell
separated as shown in Fig. 3(b). The RUs were equipped at interference. CRAN has shifted towards UE centric approach
the height (usually at the top of tower to support large area) which uses several BSs to reduce interference from the other
and DUs used to be installed in the room underneath the BS. cells [19]. As a result, each UE is served by a cluster of BSs
Fiber optical cable were utilized to connect both the units. or radio heads which are governed by the CP.
Moreover, the introduction of data hungry applications and 4) Coordinated Transmission: Earlier, several frequency
increase in the number of UEs raised the demand of further bands used to orthogonalize the adjacent cell users. Due to this,
densification. However, the densification alone was unable to the resource utilization technique was very poor . CRAN uti-
support such huge data rate demand. Thus, the framework lizes multiple transmitting points to serve each user. This sort
has shifted towards the frequency reuse-1 scenario. Also, the of transmission is known as Coordinated Multipoint(CoMP)

Authorized licensed use limited to: Indian Institute of Technology (Ropar). Downloaded on August 02,2020 at 05:47:10 UTC from IEEE Xplore. Restrictions apply.
 D. Packet Switching Functions
Like radio control function, this layer plays a key role
in virtualization. Specifically, it performs packet processing
operations which involve multi-path handling, data scheduling,
dual connectivity management and encryption [23]- [26].
Due to dis-aggregation, these all units have a capability to
perform a specific task which provides agility to the network.
The ultimate goal of RAN is to connect RU (or UE) to the core
network. However, traditional network used to treat each appli-
cation indifferently. In contrast, the connecting path between
UE to the core network in O-RAN is decided by the nature of
Fig. 4. Expended Version of Future RAN service as depicted in Fig. 4. Some applications/services allow
to directly connect to the core network while some are diverted
through various stages of radio control function. Moreover,
Transmission. CoMP algorithms are run in the heart of CP as the depth of transmission increases, the transmission la-
which provides coordination among the intra-cluster BSs. tency also increases which specifically depends on the data
Coordinated beamforming, distributed transmission and joint carrying capacity of back-haul unit at each intermediate stage.
transmission are among some of the popular CoMP techniques Specifically, management units are responsible to define the
[20]- [22]. intermediate radio connectivity of each service. As a result, the
future RAN (i.e O-RAN) becomes highly complex and require
III. O PEN RAN: OVERVIEW AND I NFRASTRUCTURE ultra large computational capability. However, to overcome
such limitations, some modern learning methods and MEC
According to the latest version of RAN, RAN is basically capabilities are being indulged with O-RAN infrastructure.
divided into two units as shown in Fig. 3(c). However,a MEC will not only reduce computational complexity at the
number of future applications require ultra low latency, more service plane but also reduce the overall latency [27].
reliable network. Fig.4 shows the internal scenario of RAN,
Functions in the traditional RAN architecture that were
which contain dis-aggregated units like RU, baseband process-
aggregated into a single node are dis-aggregated in the O-
ing function and packet processing function [23]- [26]. This
RAN. This distribution increases the reliability by avoiding
disaggregation enables each unit to perform specific functions
any single point of failure. Moreover, allowing the separation
and thus, adds flexibility in the network. Basic working of
of control plane and user plane, the control plane function
these units can be explained as:
can be implemented on all server platform while specific real
time functions can be implemented on the highly specialized
A. Radio Functions at RU hardware. Furthermore, in the O-RAN, the control plane, user
RU contains transceiver antennas along with the special plane and transport plane are intended to work independently
radio hardware which perform physical layer operations (e.g., which increase the scalability and flexibility of the O-RAN. In
digital to analog conversion, filtering operation, modulation, a nutshell, it can be inferred from the above discussion that O-
etc). In addition, it is also responsible for signal amplification RAN is the dis-aggregated, virtualized, self driven, application
and regeneration [23]- [26]. specific and software oriented network which is able to support
IoT network as well as modern high speed applications in a
B. Baseband Processing Function stand alone flexible network. The ability to handle multiple
radio link protocol interoperability. In the advanced version,
This unit is responsible for upper layer functions (i.e., radio some function that are presently handled by radio network
link control, medium access control) which specifically per- layer are planned to move in the IP layer.
form carrier aggregation, soft combining, fast radio schedul-
The introduction of O-RAN will also have a great in-
ing, CoMP operations, etc. Moreover, it is also responsible
fluence on market and operators. The deployment of O-
for selection of MIMO scheme, beam formation and antenna
RAN could open the doors for public network operators to
selection [23]- [26].
achieve their core network independent of the existing access
network technology [29], which provides the operators the
C. Radio Control Functions
leverage of their core service based network, across variety of
This unit strives to control resource distribution and load technological support. As per economic statistics, the market
sharing among different set of applications and system areas. for RAN based equipment is worth at $30 billion a year
It is one of the most essential units of RAN and performs and significantly more when smart buildings and vehicular
virtualization and radio resource management. Basically, it network is included. O-RAN seems to disrupt the network
controls the overall performance on RAN based on radio by indulging several innovations and fast creation of flexible
control algorithms [23]- [26]. inter-operable network. Finally, ORAN design could attract the

Authorized licensed use limited to: Indian Institute of Technology (Ropar). Downloaded on August 02,2020 at 05:47:10 UTC from IEEE Xplore. Restrictions apply.
 Fig. 6. RIC near-RT

run time execution. RIC near-RT utilizes the database (known

as radio network information base) which tracks the state of
the underlying network by using E2 and A1. E2 strives to
provide a standard interface between RIC near-RT and CU/DU
which feeds data that include various RAN measurements for
radio resources management. Specifically, the near-RT RIC
provides radio management tracked by AI/ML. In addition,
this layer is also responsible for operations like handover,
QoS management, etc as shown in Fig. 6 . Moreover, the
Fig. 5. Reference Architecture of O-RAN interface AI is responsible for conveying the AI enable policy
and ML based training models to the RIC of non-real time.
Basically, non-RT control functions strive to support non-real
modern business models to incorporate latest wireless services time intelligence radio resource management and providing
and next generation communication frameworks. guidance to support the operations of RIC near-RT functions
E. Advantages of O-RAN that are supported by AI interface include [32]:
With the dis-aggregation of hardware and software, O- • Useful data from network to the RIC non-RT to support

RAN creates a unified architecture unified architecture through various requirements such as offline training, online on-
several advancements and brings several benefits (i.e., low line learning AI/ML model, etc.
latency and network slicing). In addition to facilitating network • Support for RIC near-RT functions such as deploying/

automation O-RAN provide several benefits given as [31]: updating ML/AI model into the RIC of near-RT and
1) Agility: The unification of the software enabled archi- sometime feedback to ensure that the operators meet the
tecture makes the network suitable for existing/past and future intended objectives.
generation. Fig. 7 shows that dis-aggregated Control Unit stack which is
2) Deployment Flexibility: Dis-aggregation and software responsible to support various protocols (including 4G, 5G and
association makes the network flexible for installation and other protocols). RIC near-RT issues command to implement
upgradtion/extension. basic functions (e.g., handovers) virtualization provides the
3) Real time responsiveness: O-RAN is the software driven ability to distribute capacity across multiple elements. DU and
service specific network which behaves on the basis of in- radio resources unit(RRU) are responsible to support radio
tended service and thus prefers the real time services which functions, radio processing, baseband processing etc [28]-
require very low latency over the less critical services. [30].
4) Operating Cost Reduction: It is estimated that the plug 1) Future Challenges: Some challenges regarding the im-
and play feature of O-RAN and modern learning methods may plementation of model given in Fig. 5 include:
reduce the maintenance cost upto 80%. Putting the software • It is challenging to deploy policies for the RIC near-RT
at the heart of the network, the operators can unify the con- and non-real time control loop meeting the economical
nectivity gains of all the generations under the same umbrella. and ecological aspects.
Doing this, the operators can save millions of dollars. • Coordination, updation and training is difficult with the

F. O-RAN Architecture modern learning techniques.(i.e., ML and AI).

• It is challenging to handle data (specifically cross layer
Fig. 5 shows a reference O-RAN architecture which is based
data) to support the intended operation while protecting
on the principle of openness [28]- [30]. As discussed in the
other internal operations.
previous sections that O-RAN is flexible, service oriented and
software defined network. Another add on this network is
IV. T ECHNICAL A DVANCEMENTS AND O PPORTUNITIES
the affiliation of artificial intelligence. Basically the reference
architecture of O-RAN includes various sub units. As shown With the acceleration of 5G evolution and O-RAN im-
in Fig. 5 , non-real time functionalities are decoupled from the plementation, several advancements would take place in the
real time function include service and model training for non- modern technologies. Applications associated to IOT devices
real time functionality [28]- [30]. While trained models and machine learning, mobile edge computing are expected to
real time control functions (produced in real time) are included reach at the peak in the coming years. Some Key technological
in the RAN intelligent controller of the near-real time for advancements and opportunities are discussed below:

Authorized licensed use limited to: Indian Institute of Technology (Ropar). Downloaded on August 02,2020 at 05:47:10 UTC from IEEE Xplore. Restrictions apply.
 Fig. 7. Control Unit Stack

A. An IoT enabled Network

The launch of O-RAN will remove the restriction in the Fig. 8. Applications of ML & IoT
IoT connectivity. It would provide a flexible architecture,
highly suitable for modern IoT connectivity. With this, more
things should be connected with various application including networks, such as Q-learning for resource allocation and
health care, retail, security and many more [34].Some new interference coordination. In addition, Bayesian learning is
applications may now add on (e.g., digital locks, e-health, etc). used or channel estimation in MIMO network.
Fig. 8 shows some existing and possible applications [35].
Under these applications, IoT devices would be connected V. C ONCLUSION AND F UTURE A SPECT
over O-RAN framework. Furthermore, disaggregation of RIC The co-existence of variety of applications require flexible,
non-RT enables to support massive IoT connectivity to the application oriented and adaptive network which are difficult
devices which require low throughput but large coverage and to support on the existing infrastructure. As a result, ser-
low power consumption. vice providers and mobile operators are moving towards dis-
B. Enabling MEC computing aggregation of existing RAN. Modern applications require a
flexible network which leads to the emergence of building
MEC is a form of architecture that enables an edge device to up a standard open interface enabled by AI based network
perform computing tasks. Due to rapid increase in the number function virtualization. This article explained the evolution of
of connected devices, the next generation RAN should be able RAN along with background of O-RAN and its reference
to manage the traffic far intelligently. MEC is supposed to a architecture. The architecture given in this work is a step
key technique for the same [36] . Currently, most of the ap- towards software oriented network. Further we discussed var-
plications handle their content storage and online computation ious challenges associated with the O-RAN implementation.
on the remote sensors which usually lie far away from the end Furthermore, opportunities created with the advent of O-RAN
user. MEC will bring those processes closer to the end user. have been discussed.
This shift will help to reduce the congestion on the mobile
Current version of O-RAN is focused on identifying the
network and cloud computer. In addition to reducing the
radio functions of RAN that can be grouped into functional
congestion, MEC will play a major role in reducing the latency
entities which can be embedded in the distributed system.
of 5G network [37]. Bringing the data closer to the end unit
Specifically, researchers are working on pushing several lower
and streaming it more directly at the end device, extremely low
layer functions (i.e., QoS, mobility, management and security)
latency can be achieved which enables to support applications
into the upper layer. However, there are some questions and
that require high speed data and computing.
challenges; one issue in O-RAN is to standardize the opera-
C. Inclusion of Modern Learning Methods tion, administration and management because it is difficult to
Machine Learning enables a computer to learn without an achieve inter-operability without standardization. Although O-
explicit program. ML is featured by learning useful informa- RAN seems to provide the required degree of inter-operability,
tion from the input data set, which makes it suitable for the the details are to be work out.
applications in which processing environment is dynamic in
R EFERENCES
nature. Specifically, ML can enhance wireless framework in
the following ways: [1] V. H. M. Donald, ”Advanced mobile phone service: The cellular concept,”
in The Bell System Technical Journal, vol. 58, no. 1, pp. 15-41, Jan. 1979.
• ML based resource management, mobility and network-
[2] T. S. Rappaport, ”The wireless revolution,” in IEEE Communications
ing algorithms can significantly adapt the dynamic envi- Magazine, vol. 29, no. 11, pp. 52-71, Nov. 1991.
ronment. [3] M. Zeng, A. Annamalai and V. K. Bhargava, ”Recent advances in cellular
wireless communications,” in IEEE Communications Magazine, vol. 37,
• ML is considered as the key to realize the goals of self
no. 9, pp. 128-138, Sept. 1999.
realizable network. [4] Qi Bi, G. L. Zysman and H. Menkes, ”Wireless mobile communications
Fig. 8 shows various ML enabled applications driven by at the start of the 21st century,” in IEEE Communications Magazine, vol.
39, no. 1, pp. 110-116, Jan. 2001.
modern learning methods [38]. As per a survey, some machine [5] U. Varshney, ”4G Wireless Networks,” in IT Professional, vol. 14, no. 5,
learning algorithms are already introduced for future wireless pp. 34-39, Sept.-Oct. 2012.

Authorized licensed use limited to: Indian Institute of Technology (Ropar). Downloaded on August 02,2020 at 05:47:10 UTC from IEEE Xplore. Restrictions apply.
[6] A. Ghosh, A. Maeder, M. Baker and D. Chandramouli, ”5G Evolution: [28] Kawahara, Sadayuki Abeta Toshiro, and Anil Umesh Ryusuke Mat-
A View on 5G Cellular Technology Beyond 3GPP Release 15,” in IEEE sukawa. ”O-RAN Alliance Standardization Trends.” (2019).
Access, vol. 7, pp. 127639-127651, 2019. [29] ”Open RAN - The Open Road to 5G.” SAMSUNG White Paper,(2019-
[7] B. Zong, C. Fan, X. Wang, X. Duan, B. Wang and J. Wang, ”6G 07-08).
Technologies: Key Drivers, Core Requirements, System Architectures, [30] Alliance, Open RAN. ”O-RAN: Towards an Open and Smart RAN.”
and Enabling Technologies,” in IEEE Vehicular Technology Magazine, White Paper, October (2018).
vol. 14, no. 3, pp. 18-27, Sept. 2019. [31] ”https://www.parallelwireless.com/wp-content/uploads/OpenRAN-
[8] J. C. Cano, V. Berrios, B. Garcia and C. K. Toh, ”Evolution of IoT: An Network-SW-Suite.pdf”
Industry Perspective,” in IEEE Internet of Things Magazine, vol. 1, no. [32] ”O-RAN Software Community PROJECT PROPOSAL For inclusion
2, pp. 12-17, December 2018. from Release A:Non Real-Time RAN Intelligent Controller (RIC non-
[9] E. Ohn-Bar and M. M. Trivedi, ”Looking at Humans in the Age of RT).”
Self-Driving and Highly Automated Vehicles,” in IEEE Transactions on [33] ”https://www.lightreading.com/webinar.asp?webinar id=1322.”
Intelligent Vehicles, vol. 1, no. 1, pp. 90-104, March 2016. [34] S. Chen, H. Xu, D. Liu, B. Hu and H. Wang, ”A Vision of IoT:
[10] A. Marotta, D. Cassioli, K. Kondepu, C. Antonelli and L. Valcarenghi, Applications, Challenges, and Opportunities With China Perspective,” in
”Exploiting flexible functional split in converged software defined access IEEE Internet of Things Journal, vol. 1, no. 4, pp. 349-359, Aug. 2014.
networks,” in IEEE/OSA Journal of Optical Communications and Net- [35] https://www.analyticsvidhya.com/blog/2016/08/10-youtube-videos-
working, vol. 11, no. 11, pp. 536-546, November 2019. explaining-the-real-world-applications-of-internet-of-things-iot/
[11] P. Arnold, N. Bayer, J. Belschner and G. Zimmermann, ”5G radio access [36] Y. Mao, C. You, J. Zhang, K. Huang and K. B. Letaief, ”A Survey
network architecture based on flexible functional control / user plane on Mobile Edge Computing: The Communication Perspective,” in IEEE
splits,” 2017 European Conference on Networks and Communications Communications Surveys & Tutorials, vol. 19, no. 4, pp. 2322-2358,
(EuCNC), Oulu, 2017, pp. 1-5. Fourth quarter 2017.
[12] Search Networking ”https://searchnetworking.techtarget.com/definition/radio-[37] N. Abbas, Y. Zhang, A. Taherkordi and T. Skeie, ”Mobile Edge
access-network-RAN”. Computing: A Survey,” in IEEE Internet of Things Journal, vol. 5, no. 1,
pp. 450-465, Feb. 2018
[13] Techopedia ”https://www.techopedia.com/definition/9089/radio-access-
[38] Application of ML ”https://medium.com/app-affairs/9-applications-of-
network-ran.”
machine-learning-from-day-to-day-life-112a47a429d0”.
[14] M. Peng, Y. Li, Z. Zhao and C. Wang, ”System architecture and
keytechnologies for 5G heterogeneous cloud radio access networks,” in
IEEENetwork, vol. 29, no. 2, pp. 6-14, March-April 2015.
[15] M. Peng, Y. Sun, X. Li, Z. Mao and C. Wang, ”Recent Advances
in Cloud Radio Access Networks: System Architectures, Key Tech-
niques,and Open Issues,” in IEEE Communications Surveys & Tutorials,
vol.18, no. 3, pp. 2282-2308, third quarter 2016.
[16] C. Pan, M. Elkashlan, J. Wang, J. Yuan and L. Hanzo, ”User-Centric
C-RAN Architecture for Ultra-Dense 5G Networks: Challenges and
Methodologies,” in IEEE Communications Magazine, vol. 56, no. 6,
pp.14-20, June 2018.
[17] M. Shafi et al., ”Microwave vs. Millimeter-Wave Propagation Chan-
nels:Key Differences and Impact on 5G Cellular Systems,” in IEEE
Communications Magazine, vol. 56, no. 12, pp. 14-20, December 2018.
[18] Z. Cao et al., ”Advanced Integration Techniques on Broadband
Millimeter-Wave Beam Steering for 5G Wireless Networks and Beyond,”
in IEEE Journal of Quantum Electronics, vol. 52, no. 1, pp. 1-20, Jan.
2016, Art no. 0600620.
[19] D. Gomez-Barquero, D. Navratil, S. Apple by and M. Stagg, ”Point-to-
Multipoint Communication Enablers for the Fifth Generation of Wireless
/Systems,” in IEEE Communications Standards Magazine, vol. 2, no.
1,pp. 53-59, MARCH 2018.
[20] J. Wu, Z. Zhang, Y. Hong and Y. Wen, ”Cloud radio access network (C-
RAN): a primer,” in IEEE Network, vol. 29, no. 1, pp. 35-41, Jan.-Feb.
2015.
[21] J. Lee et al., ”Coordinated multipoint transmission and reception in LTE-
advanced systems,” in IEEE Communications Magazine, vol. 50, no. 11,
pp. 44-50, November 2012.
[22] D. Lee et al., ”Coordinated multipoint transmission and reception in
LTE-advanced: deployment scenarios and operational challenges,” in
IEEE Communications Magazine, vol. 50, no. 2, pp. 148-155, Feb. 2012.
[23] Westerberg, Erik. ”4G/5G RAN architecture: how a split can make the
difference.” Ericsson Technology Review 93.6 (2016).
[24] L. M. P. Larsen, A. Checko and H. L. Christiansen, ”A Survey of
the Functional Splits Proposed for 5G Mobile Crosshaul Networks,” in
IEEE Communications Surveys & Tutorials, vol. 21, no. 1, pp. 146-172,
Firstquarter 2019.
[25] M. A. Habibi, M. Nasimi, B. Han and H. D. Schotten, ”A Comprehen-
sive Survey of RAN Architectures Toward 5G Mobile Communication
System,” in IEEE Access, vol. 7, pp. 70371-70421, 2019.
[26] H. Niu, C. Li, A. Papathanassiou and G. Wu, ”RAN architecture
options and performance for 5G network evolution,” 2014 IEEE Wireless
Communications and Networking Conference Workshops (WCNCW),
Istanbul, 2014, pp. 294-298.
[27] T. Taleb, K. Samdanis, B. Mada, H. Flinck, S. Dutta and D. Sabella, ”On
Multi-Access Edge Computing: A Survey of the Emerging 5G Network
Edge Cloud Architecture and Orchestration,” in IEEE Communications
Surveys & Tutorials, vol. 19, no. 3, pp. 1657-1681, third quarter 2017.

Authorized licensed use limited to: Indian Institute of Technology (Ropar). Downloaded on August 02,2020 at 05:47:10 UTC from IEEE Xplore. Restrictions apply.
View publication stats