Device-to-Device Communications over 5G

Systems: Standardization, Challenges

and Open Issues

G. Araniti, A. Raschellà, A. Orsino, L. Militano, and M. Condoluci

Abstract This chapter deals with a comprehensive analysis of the challenges

behind the introduction of Device-to-Device (D2D) communications in the context
of 5th Generation (5G) communications. In fact, although D2D communications
have emerged as an efficient solution for local traffic between mobile User
Equipments (UEs) in proximity in cellular environments, has recently also attracted
attention as a key enabling technology for 5G wireless networks. In details, this
chapter addresses an overview of the current standardization, its integration in the
cellular system architecture, future challenges and open issues. This chapter also
provides a performance analysis, which illustrates the important achievements in
terms of data rate in a scenario where the proximity communications between
devices in an LTE-A system has been introduced for multicast downloading
services.

1 Introduction

Recently, telco operators have been facing an increasing demand for high data
rate as a consequence of a proliferation of innovative applications and services
emerging in daily routines of mobile users. In this context, Device-to-Device (D2D)
communications have attracted significant attention as a key enabling technology
of 5th generation (5G) wireless networks [1] as witnessed by the recent Long
Term Evolution (LTE) release 12 [2] from 3rd Generation Partnership Project
(3GPP). Direct D2D communications are indeed a promising technological solution
for local traffic between mobile User Equipments (UEs) in proximity in cellular

G. Araniti () • A. Orsino • L. Militano

University Mediterranea of Reggio Calabria, Reggio Calabria, Italy
e-mail: [email protected]; [email protected]; [email protected]
A. Raschellà
Liverpool John Moores University, Liverpool, UK
e-mail: [email protected]
M. Condoluci
King’s College London, London, UK
e-mail: [email protected]

© Springer International Publishing Switzerland 2017 337

W. Xiang et al. (eds.), 5G Mobile Communications,
DOI 10.1007/978-3-319-34208-5_12
338 G. Araniti et al.

environments. In particular, devices being close to each other can activate direct
links and bypass the base station (BS) or access point (AP) by either using cellular
communications resources or using alternative radio technologies such as Wi-Fi.
The conventional cellular communications paradigm involving the BS remains
still appropriate when users are not usually close enough to make possible direct
communications and for traditional low data rate mobile services such as voice calls
and text messages. However, this communication scheme can become inadequate
with the modern high data rate services (e.g., video conferencing, interactive
gaming, and mobile social networking). When the UEs are in range for D2D
communications, this innovative communication paradigm is undoubtedly an advan-
tageous solution to be considered to optimize the spectral efficiency of the network.
Noteworthy, the advantages of D2D communications are not only limited to
an improved spectral efficiency. In fact, also link coverage, throughput, energy
consumption can benefit from this communication technology. Moreover, robust
public safety communications can be supported where devices provide at least local
connectivity in case of damage to the radio infrastructure.
D2D communications can be enabled through either cellular network transmit-
ting on cellular spectrum, known as inband communication, or existing technologies
that consider unlicensed bands such as Wi-Fi access points, known as outband
communication. Concerning the inband approach, the cellular spectrum may be
shared between D2D and cellular communications, which is knows as the underlay
inband D2D mode [3]. In this case a main field of investigation is the mitigation
of the interference between D2D and cellular communication [4]. As alternative,
other works such as [5] propose to assign a portion of the cellular resources to
D2D communications to avoid interference problems, which is known as overlay
inband D2D mode. With this approach, resource allocation becomes the most
important aspect to be investigated in order to avoid wasting precious spectrum
resources [6]. The coordination between radio interfaces is either controlled by
the BS (i.e, controlled outband mode) or by the users (i.e., autonomous outband
mode). However, in both outband modalities baseline rules are needed to coordinate
the communication over two different bands where also a second radio interface is
involved (e.g., Wi-Fi Direct). Hence, the studies on outband D2D involve aspects
such as power consumption and inter-technology architectural design [7–9].
Despite of the promising features of the D2D communications, there are some
important challenges and issues still waiting for a solution before a widespread
use of D2D communications in next 5G systems is granted. These include:
(1) device discovery procedures with the aim to detect the presence of other UEs in
the neighbourhood; (2) link setup strategies in order to properly select the spectrum
to be used in the D2D radio links between interested UEs; (3) interference avoidance
mechanisms that make possible the coexistence among D2D UEs with cellular net-
work; (4) new devices design. Moreover, some research works has been conducted
on as mobility management [10], multicast and broadcast communications and radio
access procedures for D2D received little attention so far.
The interest for D2D communications as a key technology in the context of
5G mobile systems is witnessed by the initiatives of the mobile and wireless
Device-to-Device Communications over 5G Systems 339

communications enablers for the twenty-twenty information society (e.g., METIS

European project [11], 5G-PPP association [12], Networld2020 platform [13]). The
main objective of these new initiatives is to handle the foundation of 5G, the
next generation mobile and the wireless communications systems. In particular, for
what concerns the D2D communications, the METIS project is currently evaluating
its role in various scenarios such as vehicle-to-vehicle communications, national
security and public safety, cellular network offloading, or service advertisement.
In the remaining of the chapter a general overview of D2D communications
will be provided firstly introducing its possible role in the context of 5G systems.
Then, an overview of the standardization, the system architecture, and some exam-
ples of scenarios and use cases will be given. Particular attention will then devoted
to D2D communications over cellular networks and possible services exploiting
this novel technology. Then a particular scenario will be presented, showing the
potentialities offered by this technologies in enhancing the main performance
features for the UEs in a downlink multicast service. In the last part of the chapter
challenges and open issues, such as mobility and full-duplex communications will
be discussed.

2 The Role of D2D Communications in Emerging

5G Systems

The D2D communications technology has been addressed in 3GPP LTE release
12 system [14]; notwithstanding, it is expected to have a complete standardization
of proximity services in next 3GPP releases 13 and 14 [15]. As mentioned in the
introduction to this chapter, the exploitation of D2D communications between UEs
in proximity is expected to achieve improvements in terms of spectrum utilization,
overall throughput, energy consumption, and to guarantee better public safety
networks management. In what is presented next, a general overview of the current
D2D standardization process is provided together with the system architecture
proposed to integrate this new technology in the current cellular systems, and a
number of possible applications in different scenarios and use cases.

2.1 Standardization Overview

The standardization process is an aspect of utmost importance to be consid-

ered for the commercial feasibility and future deployment of new technologies.
In the particular case of D2D communications, although direct communications
are already provided by the use of unlicensed Industrial, Scientific and Medi-
cal (ISM) bands (e.g., Wi-Fi Direct), its standardization in the context of the
cellular system is currently still ongoing. A first example of the introduction of
D2D communications into the LTE-Advanced (LTE-A) network is provided by
Qualcomm Company, which developed a mobile communication system called
340 G. Araniti et al.

FlashLinq [16]. In particular, FlashLinq is a PHY/MAC network architecture, which

allows cellular devices automatically and continuously discovering thousands of
other FlashLinq enabled devices within 1 km and communicating peer-to-peer, at
broadband speeds and without the need of intermediary infrastructures. Moreover,
peer-to-peer communications enabled through Qualcomm’s FlashLinq can share
connectivity with a cellular network technology unlike Wi-Fi Direct’s-based peer-
to-peer. FlashLinq discovery procedure is carried out by broadcasting public/private
expressions mapped into tiny 128-bit packages of data, which represent basic
information of either devices or users.
From a standardization point of view, 3GPP is focusing its efforts on D2D
communications (recently begun in release 12 [14]) for public safety Proximity
Services (ProSe) [17]. This strategy has been initially targeted to allow LTE
becoming a competitive broadband communication technology for public safety
networks used by first responders. However, from a technical perspective point
of view, the exploitation of the proximity nature of the communicating devices
will provide the further performance benefits: (1) D2D UEs will be able to
exploit high data rate with a low delay due to the short range; (2) compared
to traditional downlink/uplink cellular communication, D2D will enable energy
savings and improve radio resource utilization; (3) cellular data traffic offloading
and, consequently, lower overload in the network. In detail, the 3GPP Radio
Access Network (RAN) working group has proposed in TR 36.843 Rel. 12 [17]
two basic functions for supporting ProSe discovery and ProSe communications
over the LTE radio interface. ProSe discovery allows an UE using the LTE air
interface to identify other UEs in proximity. Two kinds of ProSe discovery exist,
namely restricted and open; the difference consists in whether the permission is
necessary or not for the discovery for a UE. ProSe communication instead, is the
data communication between two UEs in proximity using the LTE air interface.
3GPP Services working group (SA1) has defined in specification TR 22.803 [18]
the use cases and scenarios for ProSe. In the document, conditions for service flows
and potential requirements for different use cases are analyzed in order to provide
a support for D2D systems design. Some examples of use cases and scenarios
identified for general commercial/social use and network offloading are summarized
below.
The following terms are defined by 3GPP in the description of D2D use cases:
• ProSe Discovery: it is a process that identifies a UE in proximity of another, using
EvolvedUMTS Terrestrial Radio Access Network (E-UTRAN).
• ProSe Communication: it is a communication between two UEs in proximity
through an E-UTRAN communication path established between the UEs. The
communication path can for example be established directly between the UEs or
routed via local evolved-NodeB (eNB).
• ProSe-enabled UE: it is a UE that supports ProSe Discovery and/or ProSe
Communication.
• LTE D2D: it is a series of technologies characterized by ProSe capability.
Device-to-Device Communications over 5G Systems 341

2.2 Uses Cases and Scenarios Presented in 3GPP Rel. 12

Some examples of use cases for ProSe Discovery and ProSe Communication
scenarios defined by 3GPP SA1 in specification TR 22.803 [17] are given below.
Restricted/Open ProSe Discovery: these are use cases for a basic ProSe
Discovery scenario that can be exploited for any kind of application. In case of
restricted ProSe Discovery, a ProSe-enabled UE discovers another UE in proximity
only if it has previously achieved the permission; while, in case of open ProSe
Discovery, a ProSe-enabled UE is able to discover neighbor devices without the
necessity of a permission. An example of restricted use case is the friend discovery
in a social network where the discovery is constrained by the UE’s privacy settings.
While a shop/restaurant advertisement is an example of open use case because shops
and restaurants are open to be discovered by all the possible ProSe-enabled UEs in
proximity, being free of privacy issues.
Network ProSe Discovery: it is a use case for ProSe Discovery scenarios where
the Mobile Network Operator (MNO) verifies if a UE has the permission to discover
another UE and the proximity. Therefore, in this case the network should be able to
determine and provide the ProSe-enabled UEs with their proximity.
Service Continuity Between Infrastructure and E-UTRA ProSe Communi-
cation Paths: this is a use case for a ProSe Communication scenario where the
operator is able to switch user traffic from the initial infrastructure communication
path to the ProSe communication one. Then, the traffic can be addressed again
towards an infrastructure path, without being perceived by the users. Hence, the
operator should be able to dynamically control the proximity criteria (e.g., range,
channel conditions, achievable QoS) for switching between the two communication
paths.
ProSe-Assisted WLAN Direct Communications: WLAN direct communica-
tion is a use case available between ProSe-enabled UEs with WLAN capability
when they are in Wi-Fi Direct communications range. It is based on the ProSe
Discovery and the WLAN configuration information from the 3GPP Evolved Packet
Core (EPC). In this case the operator is able to switch data session between
infrastructure path and WLAN ProSe communication path.
ProSe Application Provided by the Third-Party Application Developer: in
this case the operator can provide ProSe capability features in a series of APIs to
third-party application developers. Through this cooperation between the operator
and third-party application developers, the user can download and use a wide
variety of new ProSe applications created by third-party developers. In this case
the operator’s network and the ProSe enabled UE should provide a mechanism
that enables to identify, authenticate and authorize the third-party application to use
ProSe capability features.
In Table 1 the available specifications together with the corresponding main
topics provided by the 3GPP working groups is summarized. It can be noticed the
presence of the mentioned SA1 and RAN working group handling, respectively,
feasibility study for ProSe and LTE radio interface issues. Other examples of topics
342 G. Araniti et al.

Table 1 Available documents for D2D

SA1 TR 22.803 Feasibility study for Proximity Services (ProSe)
TS 22.278 Service requirements for the Evolved Packet System
(EPS)
TS 22.115 Service aspects; changing and billing
TS 21.905 Vocabulary for 3GPP specifications
SA2 TR 23.703 Study on architecture enhancements to support
Proximity-based Services (ProSe)
TS 23.303 Proximity-based Services (ProSe); Stage 2
SA3 TR 33.833 Study on security issues to support Proximity Ser-
vices
RAN 1 & RAN 2 TR 36.843 Study on LTE device to device proximity services—
Radio Aspects
TS 24.333 Proximity-services management object (MO)
CT1
TS 24.334 Proximity-services (ProSe) user equipment (UE) to
Proximity-services function aspects; Stage 3

under investigation supporting ProSe are the study of the architecture, security
issues and Management Objects (MOs) representing parameters that handle the
configuration of ProSe-enabled UEs.

2.3 System Architecture

In order to support the scenarios illustrated earlier in this chapter, the enhancements
in the LTE architecture illustrated in Fig. 1 have been proposed. In details, this
architecture aims at meeting the following requirements introduced by the 3GPP
specifications to:
• Allow the operator to control the ProSe discovery feature in its network and
authorizing the functionalities required for the ProSe discovery of each UE.
• Allow the ProSe communication or ProSe-assisted WLAN Direct communica-
tion and seamless service continuity when switching user traffic between an
infrastructure path and a ProSe communication of the ProSe-enabled UEs.
• Allow an authorized third party ProSe application interacting with the 3GPP
network in order to use the ProSe services provided by the network.
• Be able to control ProSe communication between ProSe-enabled UEs in case the
UEs are served by either the same or different eNBs.
• Handle the ProSe-related security functions that correspond to privacy, support
for regulatory functions including Lawful Interception, and authentication upon
ProSe discovery and ProSe communication.
• Allow the operator’s authorization and authentication of the third-party applica-
tions before making use of the ProSe features.
Device-to-Device Communications over 5G Systems 343

ProSe APP
LTE-Uu SGi
S1
UE E-UTRAN UPC

PC5 LTE-Uu
PC4
ProSe APP

PC3 PC2
UE ProSe Function ProSe APP
Server

PC6

PC1

Fig. 1 3GPP D2D proposed architecture

As depicted in Fig. 1, in addition to the entities of the conventional LTE

architecture (i.e., Evolved UMTS Terrestrial Radio Access Network, E-UTRAN and
Evolved Packet Core, EPC), several new entities are required. These new entities are
illustrated in the following.
• Application servers (ProSe App Server): include the ProSe capability in order
to build the application functionalities, such as specific agencies (PSAP) in case
of Public Safety or social media in the commercial cases. These applications
are proposed outside the 3GPP architecture; notwithstanding there may be
reference points toward 3GPP entities. Moreover, the Application server can
communicate with an application in the UE.
• Applications in the UE (ProSe UEs App): exploit the ProSe capability in order
to build the application functionality. An example may be the communication
between members of Public Safety groups as well as social media application
that requires finding other UEs in proximity.
• ProSe Functions: are a reference point toward the ProSe App Server, the EPC,
and the UE. The functionalities may comprise (but not limited to): (1) inter-
networking via a reference point toward the third-party applications; (2) autho-
rization and configuration of UEs for discovery and direct communication;
(3) allowing the functionality of the EPC-level ProSe discovery, and charging.
Notice that for the interconnection of the new entities and the connection with
the conventional LTE ones, seven new interfaces/reference points are illustrated
in the figure as PC1, PC2, PC3, PC4, PC5, PC6, and SGi (Fig. 1).
344 G. Araniti et al.

2.4 Application Scenarios

Applications of 5G D2D communications include local service, emergency commu-

nication, and the Internet of Things (IoT) enhancement. A brief description of these
applications is provided in the following.

2.4.1 Local Service

In this scenario, user data is directly transmitted between terminals without being
routed through the network side. Local service is usually utilized for social apps
that are a basic D2D application based on the proximity feature. Through the D2D
discovery and communication functions, a user can find other close users in order
to share data or play games with them.
Another basic application of local service is the local data transmission, which
exploits the proximity and direct data transmission characteristics of D2D to extend
mobile applications while saving spectrum resources and then, making possible
a new source of revenue for operators. In fact, local advertising service based
on proximity can accurately target people in order to improve its benefits. Some
examples of local transmissions conceived to improve commercial benefits are: a
shopping mall where discounts and commercial promotions are sent to people
walking into or around the mall; a cinema where information about movies and
showtimes can be sent to people close by.
A third application of local service is the cellular traffic offloading that can reduce
network overloading problems. In fact, consider that nowadays media services
are becoming more and more popular; their massive traffic flows cause an extensive
pressure on core networks and spectrum resources. In this context, D2D-based
local media services allow operators to save spectrum resources in their core
networks. In hotspot areas, operators or content providers can exploit media servers
storing popular media services and sending them in D2D modality to the users.
Alternatively, users can utilize D2D communications to obtain the media content
from close terminals which have obtained media services. This enables to optimize
the downlink transmission pressure of operator cellular networks. Furthermore, the
cellular communication between short-distance users can be switched to the D2D
modality in order to offload cellular traffic.

2.4.2 Emergency Communications

Natural disasters such as earthquakes can damage traditional communication

network infrastructures making networks not available and causing enormous
rescue efforts. This problem could be overcome through the introduction of D2D
communications. In fact, although the communication network infrastructures may
be irremediably affected, a wireless network can still be created between terminals
Device-to-Device Communications over 5G Systems 345

based on the D2D connections. This means that an ad hoc network can be set up
based on multi-hop D2D to guarantee smooth wireless communication between
users. Moreover, a wireless network affected by terrain or buildings can have blind
spots. With single-hop or multi-hop D2D communication, users may be connected
in the blind spots to other users, which are in coverage areas and then, be connected
to the wireless network.

2.4.3 IoT Enhancement

One of the main aims of designing new mobile communication technologies is

to create an extensive interconnection among different networks involving various
types of terminals. This is the motivation, which has pushed forward the devel-
opment of the Internet of Things (IoT) in the cellular communication framework.
The industry forecast says that by 2020 there will be 50 billion cellular access
terminals on a global scale and most of them will be devices with the IoT feature.
In this context, the connection between D2D with IoT will drive towards a truly
interconnected wireless network.
A common application of D2D-based IoT challenge is vehicle-to-vehicle (V2V)
communication in the Internet of Vehicles (IoV). For instance, when a vehicle
runs at high speeds, it can warn close vehicles in D2D mode before it changes
lanes or slows down. According to the received messages, close vehicles warn
drivers or even automatically handle the driving in an emergency situation; hence,
thanks to this application drivers can react more quickly to diminish the number
of traffic accidents. D2D communications provide inherent advantages when they
are considered in the context of IoV security issues also thanks to their favorable
features in terms of communication delay and neighbor discovery.
As there exist many IoT devices in a 5G network, access load is becoming a
serious issue to be taken into account. Nevertheless, D2D-based network access is
expected to improve this problem. In a scenario characterized by many terminals,
low-cost terminals can access close special terminals in D2D modality instead of
direct connections with BSs. Moreover, if multiple special terminals are isolated,
the wireless resources for accessing low-cost terminals may be reutilized by these
special terminals. Notice that this not only improves access pressure on BSs, but also
optimizes the spectrum efficiency. Furthermore, the D2D-based access modality is
more flexible and costs less than the small cell structure of the existing 4G networks.
In a smart home application, a smart terminal may be considered as a special
terminal. Wireless appliances in the smart home access the smart terminal in D2D
modality; while, the smart terminal may access the BS in a traditional cellular mode.
The cellular-based D2D communication can represent a real breakthrough for the
development of the smart home industry.
346 G. Araniti et al.

2.4.4 Other Applications

D2D communications may also be considered in other potential scenarios, such

as multiuser MIMO enhancement, cooperative relaying, and virtual MIMO. In the
context of the traditional multiuser MIMO, BSs find precoding weights based on
the feedback received by the terminals in the respective channel in order to create
nulls and delete interference between users. Through the introduction of D2D
communications, paired users may directly exchange information about channel
status. Hence, terminals can put together channel status information to be sent to
the BSs improving the performance of multi-user MIMO.
D2D communications may also contribute to solve problems in new wireless
communication scenarios. For instance, in the indoor positioning terminals may not
achieve satellite signals if they are indoors. In this case, the traditional satellite-based
positioning cannot work efficiently. In case of D2D-based indoor positioning, either
pre-deployed terminals with given location information, or usual outdoor terminals
with given position can detect the location of terminals to be localized, and support
indoor positioning at a low cost in 5G networks.

3 State-of-the-art on D2D Communications Over

Cellular Networks

D2D communications are expected to play a key role in the ecosystem of future 5G
cellular networks. This is motivated by two aspects: (1) the amount of data traffic
exchanged over radio mobile systems is exponentially increasing and this dictates
novel communications paradigms for radio mobile networks; (2) use cases for D2D
communications presented above represent key 5G services. As a consequence, the
natively support of D2D communications becomes crucial in 5G systems.
D2D communication was initially proposed in cellular networks as a new
paradigm to enhance network performance. Several studies in the literature have
already discussed the improvements in terms of spectral efficiency and reduced
communication delay that D2D communication can provide in cellular networks
[19–24]. On the other hand, this new paradigm presents several aspects to be
investigated in terms for instance of interference control overhead and network
protocols. Therefore, the feasibility of D2D communications in the context of
LTE-A is currently a fascinating topic under investigation by academia, industry,
and the standardization bodies. A general overview of state-of-the-art applications
based on D2D communications for future 5G wireless systems is given next in
both, uplink and downlink scenarios. Then, some examples of services where
D2D communications have been efficiently exploited in LTE-A networks will be
illustrated and assessed through exhaustive performance evaluation.
Several studies addressing D2D communications for downlink services can be
found in the literature, covering several aspects and applications as for instance
Device-to-Device Communications over 5G Systems 347

mobile data offloading [25], cell coverage extension [26] or content sharing [27, 28].
Recently, D2D communications have been taken into account also for downloading
multicast services with focus on direct device communications over short links of
a different technology than the cellular one. To cite some of them, in [29] a subset
of mobile devices are considered as anchor points in a cell to forward the multicast
data received from the BS to other devices in proximity through multihop ad-hoc
Wi-Fi links. In [30] cellular users directly communicate to carry out cooperative
retransmissions using generic short-range communication capabilities. However,
the use of heterogeneous wireless interfaces introduces several issues in terms of
content synchronization that become essential in case of multicast video streaming
applications. Moreover, as also stated in [31], the use of cellular D2D links provides
several benefits compared to outband D2D links, like Wi-Fi, in terms of improved
user throughput. Although, the focus of the literature has been mainly on technical
issues for downlink services, uplink direction scenarios are of undoubted interest as
also witnessed by recent publications, such as [32] where relaying on smartphones
is proposed to transmit emergency messages from disconnected areas. Multihop
D2D communications have been also investigated in a very few recent works. In
[33, 34] network-assisted D2D communication is addressed with an analysis on
power control and mode selection on the direct links. However, the analysis refers
to a more traditional two-hop scenario, with a UE or the eNodeB as the last hop
node. Similarly, multihop D2D communication is considered in [35, 36] for end-to-
end Machine-to-Machine and human-traffic connectivity.
As an example of D2D communications over cellular LTE-A links a downlink
scenario for multicast transmission is considered in order to efficiently overcome
the limitations identified in [29, 30]. In details, in the scenario proposed in this study
a portion of multicast users, which sense poor channel qualities is split into clusters.
The members of these clusters are served through cellular D2D transmissions, while
the remaining users (i.e., those with better channel quality) are served over cellular
transmission from the BS.
All the solutions illustrated in this section exploit D2D communications relying
on LTE-A network infrastructure. In LTE-A, Orthogonal Frequency Division
Multiple Access (OFDMA) and Single Carrier Frequency Division Multiple Access
(SC-FDMA) are considered, respectively, in case of downlink and uplink. The
eNodeB handles the spectrum resources by providing the appropriate number of
RBs to each scheduled user and by selecting the Modulation and Coding Scheme
(MCS) for each RB. Scheduling solutions are based on the Channel Quality
Indicator (CQI) feedback, which is sent by a UE to the eNodeB over dedicated
control channels. Each CQI value correspond to a given maximum supported MCS
as specified in [37]. The MCS parameters can be adapted at every CQI Feedback
Cycle (CFC), which can last one or several Transmission Time Intervals (TTIs)
where one TTI is 1 ms.
348 G. Araniti et al.

3.1 A Case Study: D2D-Enhanced Conventional Multicast

Scheme for Video Delivery

Nowadays, several research and industrial organizations are considering solutions

to best manage the increased traffic and the dissimilar channel quality sensed by
users in the same multicast group and efficiently handle the available resources [38].
A basic approach to this issues is the Conventional Multicast Scheme (CMS) [38],
which serves all multicast users in a cell at every TTI, by constraining the data rate
to the user with the worst channel conditions. However, this approach translates
into poor performance in terms of data rate and low satisfaction levels for users
with good channel situations. An alternative solution is the Opportunistic Multicast
Scheme (OMS) [39], which only serves users with the best channel conditions in
each TTI. This enables to enhance the network data rate, but short-term fairness is
no more assured to users and this could affect the delivery of time sensitive services
such as video to some users. In this study it is demonstrated how multicast delivery
strategies in LTE-A systems can benefit from D2D communications. For instance,
neighboring UEs that belong to the same multicast group can exploit direct links by
utilizing cellular radio resources [21, 40], to cope with adverse cell-edge effects.
This study exploits the D2D communications in the multicast communication
framework to complement CMS in order to address some of the limitations of the
latter. In details, it is proposed a D2D-enhanced Conventional Multicast Scheme
(CMS) with Single Frequency (D2D-SF) by combining in an innovative way CMS-
based and D2D content delivery in order to improve the aggregate data rate of the
cell, while also keeping the short-term fairness between devices. D2D-SF solution
makes use of one or more mobile subscriber devices as forwarding devices (FD) to
retransmit the data received from the BS over direct local links to other members of
the multicast group. It is considered that the D2D links exploit uplink frequencies,
as suggested in [22] and all the FDs in the same cell simultaneously use the same
frequency to transmit multicast data over the D2D links, as illustrated in [41, 42].
The receivers take into account these retransmissions as multipath components of
the same signal. D2D-SF is defined in conjunction with a policy for selecting FDs
based on the devices clustering. Simulation results demonstrate how significant
enhancements of the overall performance when delivering video content can be
achieved compared to both state-of-the-art multicast strategies and a D2D novel
solution that does not consider the single-frequency paradigm.

3.1.1 System Model and Problem Formulation

The LTE/MBMS architecture [42] is composed as follows (i.e., please refers to

Fig. 2 to have a general overview). The access network is made of the eNodeB
and the MultiCell/Multicast Coordination Entity (MCE), in charge of transmission
parameters configuration in single- and multi-cell mode, respectively. The core
network involves: Mobility Management Entity (MME) that is in charge for
Device-to-Device Communications over 5G Systems 349

Fig. 2 LTE/MBMS architecture

authentication, security, and mobility management procedures; MBMS Gateway

(MBMS-GW), a logical entity whose principal function is data packets forwarding
to eNodeBs; Broadcast Multicast-Service Center (BM-SC), which is the MBMS
traffic source accomplishing also service announcement and group membership
functions. MBMS entities rely on the LTE-A system illustrated in Sect. 3 consid-
ering TDD, by referring to the frame structure type 2 foreseen by 3GPP [41] and
configuration 1 which guarantees an equal number of downlink and uplink slots
over the frame.
Let us consider a group of UEs that is interested in the same multicast content
provided by a single LTE-A cell. Under this condition a multicast video delivery can
be provided exploiting the enhancements offered by D2D communications among
the involved devices. In the considered LTE-A cell, a set of users, denoted by
K , is associated to the same multicast group (MG). The eNodeB performs link
adaptation procedures on both cellular and D2D links by managing N available RBs
according to the CQI feedbacks collected from each user. Let C be the number of
available CQI levels and let ck 2 f1; 2; : : : ; Cg be the CQI reported by multicast
member k, with k 2 K . Furthermore, let ck;j be the CQI value for each D2D link
between nodes k; j 2 K ; k ¤ j. For a given MCS value m, the available data rate
depends on the number of assigned RBs and on the spectral efficiency for the given
MCS, bm expressed in bit/s/Hz as explained in [31]. Hence, the spectral efficiency
m and bm (where m D 1; : : : ; C ) respectively in downlink and
is denoted with bdl ul

uplink transmissions. Moreover, f dl .m; nm / and f ul .m; nm / represent the data rate
respectively in downlink and uplink transmissions adopting the MCS associated
to the CQI m, as a function of m and the assigned RBs. The proposed radio
resource management (RRM) scheme is responsible of deciding which multicast
configuration to allow, by this meaning: (1) the set of UEs directly served by the
eNodeB in downlink, (2) the MCS for the downlink transmission, (3) the cluster
350 G. Araniti et al.

configuration for D2D relaying, and (4) the resource allocation and the MCS
assignment for the transmissions of each activated FD. The eNodeB carries out
the following steps when the service delivery starts. In details, a single execution
of the listed steps is carried out. However, when significant variations in the
channel conditions are sensed, these steps should be repeated to update the service
configuration.
1. Service registration: The eNodeB advertises the multicast service and all inter-
ested UEs within the cell join this service to create a single MG.
2. CQI collection: The eNodeB collects the CQI feedbacks from all UEs belonging
to the MG, i.e., ck 8k 2 K .
3. D2D CQI collection: The eNodeB collects the ck;j values from all UEs k; j 2
K ; k ¤ j belonging to the MG. This information will be utilized to discover the
UEs that can be reachable through D2D links by selected FDs in the MG.
4. FD selection and cluster formation: Being CQ  f1; 2; : : : ; Cg the set of CQI
levels in downlink for the UEs in the MG, for each m 2 CQ the eNodeB computes:
(1) the set of UEs that can correctly decode data if served by the BS, i.e., Kmdl D
fk 2 K jck  mg; (ii) the subset of served UEs Rm  Kmdl , that can act as
FDs; (3) the remaining UEs that are not served by the eNodeB, but can be served
by a FD through D2D connections. Hence, the eNodeB determines a D2D CQI
matrix (DCM) based on the ck;j values (where k 2 Kmdl and j 2 K n Kmdl ) for all
the links between the potential FDs (the matrix rows) and the remaining nodes
(the DCM columns). A ck;j D 0 value in the DCM represents that a D2D link
cannot be activated between nodes k and j. Following the values in the DCM, the
eNodeB will select the subset of UEs Dm;r  K n Kmdl to be associated to each
allowed FD r 2 Rm .
5. D2D link configuration: For each CQI level m 2 CQ evaluated for downlink
ul
transmissions, the eNodeB determines the resource Nm;r and the MCS level lm;r ,
to be utilized on the D2D link for each FD r 2 Rm . D2D links can be either
unicast or multicast. A conservative approach is considered in the multicast
case; hence, the FD serves all UEs in the D2D cluster in a single transmission
by using the MCS corresponding to the worst CQI value in the DCM, i.e.,
lm;r D min fcr;k g for FD r. This study takes into account two alternative
k2Dm;r
policies according to which the FDs manage the uplink frequencies to send data
in their own D2D cluster. The first policy associates different resources to the
different FDs; the second one implements the novel single-frequency-based D2D
paradigm, i.e., all the FDs considers the same RBs. In the former case, disjoint
sets of RBs are assigned to the D2D links. In the latter case the amount of
resources assigned to the D2D links are constrained by the cluster with the lowest
activated MCS. In general, devices connected on a D2D link are expected to be at
a short distance and with good channel conditions, thus they need a lower amount
of resources compared to those needed for a direct cellular communication.
6. Multicast service activation and resource allocation: Finally, the eNodeB
chooses the solution to activate, which is the one that optimizes the system
data rate under the constraint that all the UEs in a MG are served, either through
Device-to-Device Communications over 5G Systems 351

direct cellular links or through D2D links. In details, after the selection of the
MCS level m to activate in downlink and of the corresponding Kmdl , Rm ,
Dm ;r , Nmul ;r , and lm ;r values, the eNodeB assigns the available resources.

All values m 2 CQ are potential CQI levels to activate in downlink. For each of the
CQI levels a cluster formation algorithm is implemented to provide a configuration
of FDs and corresponding D2D clusters. If a given tested level is eligible, then the
corresponding data rate m is determined. A cluster configuration is taken into
account eligible if the FDs are able to forward the total amount of bits received
from the eNodeB over the D2D links to all users not served by the cellular link.
This needs the following conditions: (1) the enabled FDs can successfully serve all
the nodes belonging to K n Kmdl via D2D links, and (2) the N available resources
are enough to relay all data to the D2D receivers. If instead, no cluster configuration
for the tested CQI level m can be found, thus the iteration on the m 2 CQ value is
stopped and the final selection is carried out. In details, the iteration can be stopped
as the tested CQI levels follow an order from the minimum to the maximum CQI
value, and with higher values for the CQI level in downlink the probability of getting
an eligible configuration is reduced. Fundamental steps in the implementation of the
proposed RRM previously discussed, are the FD selection and cluster formation and
the D2D link configuration with the radio resource allocation. Concerning the first
strategy, let us consider the generic iteration where the m-th CQI level is tested for
downlink transmission. Given Kmdl , the set of UEs that can correctly decode the data
according to the considered CQI, and based on the DCM, the eNodeB determines
which nodes can potentially act as FDs for the remaining K n Kmdl nodes. Based
on this information the eNodeB can assign the resources to each D2D link. Let us
focus now on the BSC policy, which is based on the idea that the eNodeB chooses
“the best” FD for each UE not served in downlink. In particular, the best FD for
each node j belonging to K n Kmdl is considered as the node r 2 Kmdl which assures
the best D2D link conditions. In those cases where more than one FD can assure
the same CQIs, the eNodeB selects the FD serving more users in order to limit the
number of FDs.

3.1.2 Performance Results

Main simulation parameters are listed in Table 2. R D 100 RBs are available in
the LTE system on a 20 MHz channel bandwidth. Channel conditions for the UEs
are evaluated in terms of signal to interference and noise ratio (SINR) experienced
over each sub-carrier [43] when path loss and fading phenomena affect the signal
reception. The effective SINR is mapped onto the CQI level that guarantees a block
error rate (BLER) smaller than 1 % [43, 44].
The following metrics have been considered to assess the performance of the
proposed solutions with respect to CMS [38] and OMS [39]:
• mean data rate is the mean data rate value experienced by the multicast members;
352 G. Araniti et al.

Table 2 Main simulation parameters

Parameter Value
Cell radius 500 m
Frame Structure Type 2 (TDD)
TTI 1 ms (11 OFDM data symbols plus 3 control symbols)
Cyclic prefix/Useful signal frame length 16.67 s / 66.67 s
TDD configuration 1
Carrier Frequency 2.5 GHz
eNodeB Tx power 46 dBm
D2D node Tx power 23 dBm
Noise power 174 dBm/Hz
Path loss (cell link) 128.1 + 37.6 log(d), d[km]
Path loss (D2D link, NLOS) 40 log(d) + 30 log(f) + 49, d[km], f[Hz]
Path loss (D2D link, LOS) 16.9 log(d) + 20 log (f/5) + 46.8, d[m], f[GHz]
Shadowing standard deviation 10 dB (cell mode); 12 dB (D2D mode)
RB size 12 sub-carriers, 0.5 ms
Sub-carrier spacing 15 kHz
BLER target 1 % [44]

• aggregate data rate (ADR) is the sum of the data rates experienced by the
multicast users;
The performance analysis in terms of the metrics indicated focuses on the video
streaming towards multicast users lasting 1 s; within this interval the BS adapts the
transmission parameters every scheduling frame (i.e., 10 ms) [45]. In particular, for
this analysis, two different study cases are considered:
• Case A: This case analyses the impact that the channel bandwidth has on the
considered strategies. In this case the multicast group size jK j is set to 200,
whereas a variable number of resources R (ranging from 10 to 100 RBs) is
dedicated to the service. UEs are distributed within an area of 100 m  100 m
located near the cell-edge;
• Case B: This case analyses the impact a varying multicast group size has on
the considered strategies. The number of available resources R is set to 100 RBs,
whereas the number of UEs jK j ranges from 20 to 200. The same cell-edge
distribution of UEs as in case A is considered.
Video parameters are set in accordance to [46], where adaptive video coding [47]
is considered to be performed at the BS. Video parameters are tuned such that the
video stream has an average bit rate between 256 kbps and a maximum value, which
depends on the channel quality experienced by multicast users.
Case A The results achieved are plotted in Fig. 3. As expected, both mean data rate
and ADR improve with the number of available RBs for all solutions. The proposed
D2D-based strategies outperform CMS, with a better performance for D2D-SF . The
Device-to-Device Communications over 5G Systems 353

a 10 b1800
CMS CMS
OMS OMS
D2D 1500 D2D
8 D2D−SF

Aggregated Data Rate [Mbps]

D2D−SF
Mean Data Rate [Mbps]

1200
6

900

4
600

2
300

0 0
10 20 30 40 50 60 70 80 90100 10 20 30 40 50 60 70 80 90100
Total number of available RBs Total number of available RBs

Fig. 3 Performance analysis for video streaming analysis: study case A. (a) Mean UE data rate,
(b) aggregate data rate

single-frequency solution also outperforms OMS. The benefit compared to CMS is

constant with the number of available RBs and equal to 144 % for the D2D and
D2D-SF solutions. The mean data rate for CMS is lower than D2D (left-hand in
Fig. 3). The data rate of the OMS solution gets higher values than the D2D policy,
but lower values compared to the solution based on the single-frequency paradigm.
The price to pay when adopting OMS is the decrease in the number of served users
and in short-term fairness. A similar behavior is achieved for ADR (right-hand in
Fig. 3).
Case B The performance obtained by varying the multicast group size is shown
in Fig. 4. Also in this case, the novel scheme considering the single-frequency
paradigm, i.e., D2D-SF, outperforms the others. It is worth noting that for all
the solutions, the mean data rate slightly diminishes when the number of users
in the cell increases, as illustrated in left-hand plot in Fig. 4. In details, the CMS
shows a performance reduction of about 29 % when changing from 20 to 200 UEs.
When considering D2D there is a 25 % of reduction. This is an expected result as
greater the number of users in the group, higher the risk of having users with very
low channel conditions, which limit the overall performance. Also OMS senses
a reduction in its offered mean data rate with the increase in number of UEs.
Nevertheless, this phenomenon is less evident, as the mean data rate diminishes
from 7.9 Mbps (with 20 UEs) to 7.5 Mbps (with 200 UEs). As expected, the right-
hand in Fig. 4 illustrates how the ADR value increases with the number of multicast
members in the cell. Furthermore, the gain introduced by D2D with respect to the
CMS policy is larger when the number of multicast users increases.
354 G. Araniti et al.

a 10 b 1800
CMS
9 OMS
1500 D2D
8 D2D−SF

Aggregated Data Rate [Mbps]

7
Mean Data Rate [Mbps]

1200
6

5 900

4
600
3

2 CMS
OMS 300
1 D2D
D2D−SF
0 0
20 40 60 80 100 120 140 160 180 200 20 40 60 80 100 120 140 160180200
Number of multicast users Number of multicast users

Fig. 4 Performance analysis for video streaming analysis: study case B. (a) Mean UE data rate,
(b) aggregate data rate

4 Challenges and Ongoing Research

Future 5G cellular networks are attracting the interest of the research community
and engineers from the entire world. In particular, 5G technologies are expected to
attain 1000 times higher mobile data volume per unit area, 10–100 times higher
number of connecting devices and user data rate, 10 times longer battery life,
and five times reduced latency [48]. However, 5G is currently without an official
definition. Notwithstanding, the research community has already highlighted how
the above objectives can be potentially reached through a multi-tier heterogeneous
network architecture together with several technologies such as spatial modulation,
millimeter wave (mmWave), visible light communication (VLC), and massive
MIMO. In the following new challenges in the context of D2D communications
as innovative technology supporting 5G will be illustrated.

4.1 Mobility

The development of wireless networking technologies has brought enormous

changes to people’s everyday life. In particular, the vision of 5G networks empowers
users to communicate through different wireless access technologies enabling
new applications in modern communication paradigms, such as vehicular ad
hoc networks (VANETs) and mobile cloud computing (MCC). These scenarios,
however, introduce to one of the major challenging features related to D2D
communications, namely user mobility. In particular, it affects network connectivity
and the design of network protocols. On the other hand, another important aspect to
be considered is the complex traffic flow related to a dynamic network topology and
the double communication opportunities (i.e., infrastructure or infrastructureless
wireless networks). The interest for D2D communications in Vehicle-to-Vehicle
Device-to-Device Communications over 5G Systems 355

(V2V) communication is related to the strict delay requirements in traffic safety

use cases [49]. In particular, for collision avoidance systems it will be essential to
have very low latency. Hence, it can obtain more complete and reliable information,
which would be useful to avoid accidents. By using D2D communications rather
than relying on infrastructure, it is possible to both diminish the latency and to
develop a solution, which works without cellular network coverage so that a car
can get more complete and reliable information to avoid accidents.

4.2 Heterogeneous Networks

D2D communications may be also considered as an efficient solution for Multiuser

cooperative communication (MUCC). For instance, a benefited user (BU) is in an
area experiencing a poor cellular signal, but there is another user in the proximity
sensing a good cellular signal. This user may be considered for helping the BU to
enhance its experienced signal and then, acting as a so-called supporting user (SU).
There exist two radio links for the BU: BU towards eNB (i.e., link directly) and
SU directly towards picocells. Moreover, both BU and SU communicate with each
other through LTE-A D2D communication. This kind of communication enhances
the throughput because the system will always be able to schedule the best user
experiencing the most favorable channel quality. There exist different channel/pipes
and the probability that all channels are degraded at the same time is quite small,
and this will improve reliability. Moreover, any single pipe improvement may be
also considered at the same time (e.g., MIMO). Furthermore, MUCC improves
the performance of this scheme through the context information (e.g., location
information) of the users.

4.3 mmWave

Millimeter wave (mmWave) communication operating in the 10–300 GHz band,

appear to be an innovative candidate for next-generation cellular systems by which
multiple gigabit-per-second data rates can be supported. On the other hand, enabling
mmWave cellular systems in practice requires properly dealing with the channel
impairments and propagation characteristics of the high frequency bands. In fact, the
main propagation-related obstacles in realizing mmWave cellular are that free-space
path loss is much larger in mmWave due to the higher carrier frequency, scattering
is less significant which reduce the available diversity, and no line-of-sight (LOS)
links are weaker making blockage and coverage holes more pronounced. Moreover,
the noise power is larger due to the use of larger bandwidth channels. Furthermore,
it is difficult for mmWave signals to penetrate through solid materials that confines
outdoor mmWave signals to streets and other outdoor structures, although some
signal power may get inside the buildings through glass windows and wood doors.
356 G. Araniti et al.

The mentioned propagation features give opportunities and advantages in allowing

D2D communications over directional mmWave networks. In fact, mmWave can be
considered in D2D-enabled wireless devices for direct short range communications
amongst users or machines in near proximity. Moreover, D2D communication can
also provide a path between two wireless devices and between wireless devices
and mmWave BSs by relay if LOS paths are no available. Furthermore, mmWave
communication with highly directional antennas gives wireless connections charac-
terized by a high data rate for BS to BS (B2B) communications, in contrast with 4G
cellular networks where fiber links communications between BSs are considered.
Finally, interference management is one of the most important issue that may be
enhanced through the exploitation of D2D communications. Due to the directional
antennas and large propagation loss, mmWave communication has relatively low
multi-user interference (MUI) that may support simultaneous communications over
the same radio spectrum. Enabling multiple concurrent D2D links, the network
capacity and spectrum efficiency may be efficiently enhanced. Hence, new network
architectures and resource sharing strategies that will account for the directional
interference sources are needed in mmWave 5G cellular networks to fully exploit
the enhancements of both technologies.

4.4 Massive-MIMO

Massive MIMO is a technology whose features can be efficiently exploited in

the context of D2D communications in uplink scenarios of cellular networks. In
details, in a massive MIMO system, each BS utilizes a very large antenna array
to serve multiple users in each time-frequency resource block [50]. If the number
of antennas at a BS is considerably greater than the number of served users, the
channel of each user towards the BS is closely orthogonal to that of any other
user. This aspect enables to process techniques for very simple transmit or receive
such as matched filtering to be nearly optimal with enough antennas even in the
presence of interference. This means that, a large antenna array at a BS may
drive D2D communications towards a very low interference at the uplink massive
MIMO BS, making D2D very simple and appealing in massive MIMO systems.
However, although D2D-to-cellular interference can be effectively managed by the
large antenna array at a BS, cellular interference provoked in D2D communications
holds and can be worse in a massive MIMO system. In details, massive MIMO
is a multi-user transmission strategy developed in order to support multiple users
in each time-frequency block; the number of users simultaneously active in an
uplink scenario is scalable with the number of antennas at the BS. Considering
this increased number of uplink transmitters, the D2D connections reutilizing
uplink radio resources will sense increased interference. In order to protect D2D
connections against such increased interference, the number of simultaneously
active uplink users might have to be limited.
Device-to-Device Communications over 5G Systems 357

4.5 Full-Duplex

Duplex mode is a recent innovative research direction related to D2D communi-

cations development. In particular, when considering FDD, two receivers and two
antennas are needed. The Tx antenna and Rx one should have enough isolation
distance. It is always challenging implementing into UE with small physical
size. While TDD has the advantage that uplink downlink transmissions/receptions
are performed in different time slots and in the same frequency band. This is
basically why most of D2D researchers are currently focusing on TDD modality
[51]. It is predicted that in 5G D2D will be supported by both TDD and FDD.
Notwithstanding, D2D TDD mode will be the principal type considered thanks
to its lower cost, lower complexity, and the possible benefit of better frequency
availability.

5 Conclusion

In this chapter an overview of the D2D-communications has been presented as one

of the promising technological features for the definition of future 5G wireless net-
works. The details related to the standardization process, the possible applications
and technological issues, have been browsed together with the important benefits
such a technology can offer. In the second part of the chapter a practical application
of D2D communications has been discussed, with reference to a scenario where
the proximity communications between devices in an LTE-A system, introduces
benefits to a multicast downloading service. Important improvements in terms of
data rate increase for the involved users has been demonstrated through simulative
results. Finally, in the last part of this chapter open challenges and research field
directions have been introduced for the design of novel technological solutions
integrating also mobility issues, mmWave communications, massive MIMO and
Full-Duplex.

References

1. X. Lin, J.G. Andrews, A. Ghosh, R. Ratasuk, An overview on 3GPP device-to-device proximity

services. IEEE Commun. Mag. 52(4), 40–48 (2014)
2. 3GPP, 3rd Generation Partnership Project; Technical Specification Group SA; Feasibility Study
for Proximity Services (ProSe) (Release 12). TR 22.803 V1.0.0, August 2012
3. C.-H. Yu, K. Doppler, C. Ribeiro, O. Tirkkonen, Performance impact of fading interference
to device-to-device communication underlaying cellular networks, in Proceedings of IEEE
PIMRC (IEEE, Piscataway, 2009), pp. 858–862
4. L. Lei, Y. Zhang, X. Shen, C. Lin, Z. Zhong, Performance analysis of device-to-device
communications with dynamic interference using stochastic petri nets. IEEE Trans. Wirel.
Commun. 12(12), 6121–6141 (2013)
358 G. Araniti et al.

5. Y. Pei, Y.-C. Liang, Resource allocation for device-to-device communication overlaying two-
way cellular networks. IEEE Trans. Wirel. Commun. 12(7), 3611–3621 (2013)
6. L. Lei, Y. Kuang, X. Shen, C. Lin, Z. Zhong, Resource control in network assisted device-
to-device communications: solutions and challenges. IEEE Commun. Mag. 52(6), 108–117
(2014)
7. A. Asadi, V. Mancuso, Energy efficient opportunistic uplink packet forwarding in hybrid
wireless networks, in Proceedings of the Fourth International Conference on Future Energy
Systems (2013), pp. 261–262
8. A. Asadi, V. Mancuso, On the compound impact of opportunistic scheduling and D2D com-
munications in cellular networks, in Proceedings of the 16th ACM International Conference
on Modeling, Analysis and Simulation of Wireless and Mobile Systems (MSWiM ’13) (ACM,
New York, 2013), pp. 279–288
9. M. Condoluci, L. Militano, A. Orsino, J. Alonso-Zarate, G. Araniti, LTE-direct vs. WiFi-
direct for machine-type communications over LTE-A systems, in 2015 IEEE 26th Annual
International Symposium on Personal, Indoor, and Mobile Radio Communications (PIMRC),
30 August–2 September 2015, pp. 2298–2302
10. A. Orsino, M. Gapeyenko, L. Militano, D. Moltchanov, S. Andreev, Y. Koucheryavy, G.
Araniti, Assisted handover based on device-to-device communications in 3GPP LTE systems,
in IEEE Globecom Workshop on Emerging Technologies for 5G Wireless Cellular Networks,
December 2015
11. EU Project METIS, https://www.metis2020.com
12. The 5G Infrastructure Public Private Partnership (5GPPP), https://5g-ppp.eu
13. NetWorld2020, http://networld2020.eu
14. D. Astely, E. Dahlman, G. Fodor, S. Parkvall, J. Sachs, LTE release 12 and beyond [Accepted
From Open Call]. IEEE Commun. Mag. 51(7), 154–160 (2013)
15. Ericsson white paper: LTE Release 13, http://www.ericsson.com/res/docs/whitepapers/
150417-wp-lte-release-13.pdf
16. W. Xinzhou, S. Tavildar, S. Shakkottai, T. Richardson, L. Junyi, R. Laroia, A. Jovicic,
FlashLinQ: a synchronous distributed scheduler for peer-to-peer ad hoc networks. IEEE/ACM
Trans. Networking 21(4), 1215–1228 (2013)
17. 3GPP TR 36.843, Study on LTE device to device proximity services; Radio aspects, v12.0.1,
March 2014
18. 3GPP TR 22.803, Feasibility study for proximity services (ProSe), v12.2.0, June 2013
19. IEEE Standard 802.16e-2005, Air interface for fixed and mobile broadband wireless access
systems amendment for physical and medium access control layers for combined fixed and
mobile operation in licensed band (2005)
20. B. Kaufman, B. Aazhang, Cellular networks with an overlaid device to device network, in Pro-
ceedings of Asilomar Conference on Signals, Systems and Computers (2008), pp. 1537–1541
21. K. Doppler, M. Rinne, C. Wijting, C. Ribeiro, K. Hugl, Device-to-device communication as an
underlay to LTE-advanced networks, IEEE Commun. Mag. 47(12), 42–49 (2009)
22. K. Doppler, M.P. Rinne, P. Janis, C. Ribeiro, K. Hugl, Device-to-device communications;
functional prospects for LTE-Advanced networks, in Proceedings of IEEE ICC Workshops
(2009), pp. 1–6
23. A. Osseiran, K. Doppler, C. Ribeiro, M. Xiao, M. Skoglund, J. Manssour, Advances in device-
to-device communications and network coding for IMT-Advanced, in ICT Mobile Summit
(2009)
24. T. Peng, Q. Lu, H. Wang, S. Xu, W. Wang, Interference avoidance mechanisms in the hybrid
cellular and device-to-device systems, in Proceedings of IEEE PIMRC (2009), pp. 617–621
25. S. Andreev, O. Galinina, A. Pyattaev, K. Johnsson, Y. Koucheryavy, Analyzing assisted
offloading of cellular user sessions onto d2d links in unlicensed bands. IEEE J. Sel. Areas
Commun. 33, 67–80 (2015)
26. G. Fodor, E. Dahlman, G. Mildh, S. Parkvall, N. Reider, G. Miklos, Z. Turanyi, Design aspects
of network assisted device-to-device communications. IEEE Commun. Mag. 50, 170–177
(2012)
Device-to-Device Communications over 5G Systems 359

27. J. Seo, T. Kwon, V. Leung, Social groupcasting algorithm for wireless cellular multicast
services. IEEE Commun. Lett. 17, 47–50 (2013)
28. L. Militano, M. Condoluci, G. Araniti, A. Molinaro, A. Iera, G.-M. Muntean, Single
frequency-based device-to-device enhanced video delivery for evolved multimedia broadcast
and multicast services. IEEE Trans. Broadcast. 61(2), 263–278 (2015)
29. S.C. Spinella, G. Araniti, A. Iera, A. Molinaro, Integration of ad-hoc networks with infras-
tructured systems for multicast services provisioning, in International Conference on Ultra
Modern Telecommunications and Workshops, 2009. ICUMT ’09, 12–14 October 2009, pp. 1–6
30. Q. Zhang, F.H.P. Fitzek, V.B. Iversen, Design and performance evaluation of cooperative
retransmission scheme for reliable multicast services in cellular controlled P2P networks, in
IEEE 18th International Symposium on Personal, Indoor and Mobile Radio Communications,
(PIMRC), Athens, Greece, September 2007
31. L. Militano, M. Condoluci, G. Araniti, A. Molinaro, A. Iera, F.H.P. Fitzek, Wi-Fi cooperation
or D2D-based multicast content distribution in LTE-A: a comparative analysis, in IEEE
International Conference on Communications Workshops (ICC), June 2014, pp. 296–301
32. H. Nishiyama, M. Ito, N. Kato, Relay-by-smartphone: realizing multihop device-to-device
communications. IEEE Commun. Mag. 52(4), 56–65 (2014)
33. J. da Silva, G. Fodor, T. Maciel, Performance analysis of network-assisted two-hop d2d
communications, in Globecom Workshops (GC Wkshps), December 2014, pp. 1050–1056
34. L. Lei, X. Shen, M. Dohler, C. Lin, Z. Zhong, Queuing models with applications to mode
selection in device-to-device communications underlaying cellular networks. IEEE Trans.
Wirel. Commun. 13(12), 6697–6715 (2014)
35. G. Rigazzi, F. Chiti, R. Fantacci, C. Carlini, Multi-hop d2d networking and resource
management scheme for m2m communications over lte-a systems, in International Wireless
Communications and Mobile Computing Conference (IWCMC) (IEEE, 2014), pp. 973–978
36. L. Militano, A. Orsino, G. Araniti, A. Molinaro, A. Iera, A constrained coalition formation
game for multihop D2D content uploading. IEEE Trans. Wirel. Commun., 15(3), 2012–2024
(2015)
37. X. Lu, P. Wang, D. Niyato, A layered coalitional game framework of wireless relay network.
IEEE Trans. Veh. Technol. (63), 472–478 (2014)
38. A. Richard, A. Dadlani, K. Kim, Multicast scheduling and resource allocation algorithms for
OFDMA-based systems: a survey. IEEE Commun. Surv. Tutor. 15, 240–256 (2013)
39. T.P. Low, M.O. Pun, Y.W.P. Hong, C.C.J. Kuo, Optimized opportunistic multicast scheduling
(OMS) over wireless cellular networks. IEEE Trans. Wirel. Commun. 9, 791–801 (2009)
40. L. Lei, Z. Zhong, C. Lin, X. Shen, Operator controlled device-to-device communications in
LTE-advanced networks. IEEE Wirel. Commun. 19, 96–104 (2012)
41. 3GPP TS 36.300, Evolved Universal Terrestrial Radio Access (E-UTRA) and Evolved
Universal Terrestrial Radio Access Network (E-UTRAN), Rel. 11, September 2012
42. 3GPP TS 36.440, General aspects and principles for interfaces supporting Multimedia
Broadcast Multicast Service (MBMS) within E-UTRAN, Rel. 11, September 2012
43. C. Mehlfuhrer, M. Wrulich, J.C. Ikuno, D. Bosanska, M. Rupp, Simulating the long term evo-
lution physical layer, in 17th European Signal Processing Conference (EUSIPCO), Glasgow,
Scotland, August 2009, pp. 1471–1478
44. A. Urie, A. Rudrapatna, C. Raman, J.M. Hanriot, Evolved multimedia broadcast multicast
service in LTE: an assessment of system performance under realistic radio network engineering
conditions. Bell Labs Tech. J. 18, 57–76 (2013)
45. S. Deb, S. Jaiswal, K. Nagaraj, Real-time video multicast in WiMAX networks, in IEEE INFO-
COM 2008. The 27th Conference on Computer Communications, April 2008, pp. 1579–1587
46. S. Sharangi, R. Krishnamurti, M. Hefeeda, Energy-efficient multicasting of scalable video
streams over WiMAX networks. IEEE Trans. Multimedia 13, 102–115 (2011)
47. Y. Wang, L.P. Chau, K.H. Yap, Bit-rate allocation for broadcasting of scalable video over
wireless networks. IEEE Trans. Broadcast. 56, 288–295 (2010)
48. J.G. Andrews, S. Buzzi, C. Wan, S.V. Hanly, A. Lozano, A.C.K. Soong, J.C. Zhang, What will
5G be. IEEE J. Select. Areas Commun. 32(6), 1065–1082 (2014)
360 G. Araniti et al.

49. G. Piro, A. Orsino, C. Campolo, G. Araniti, G. Boggia, A. Molinaro, D2D in LTE vehicular
networking: system model and upper bound performance, in 2015 7th International Congress
on Ultra Modern Telecommunications and Control Systems and Workshops (ICUMT), 6–8
October 2015, pp. 281–286
50. X. Lin, R.W. Heath, J.G. Andrews, The interplay between massive MIMO and underlaid D2D
networking. IEEE Trans. Wirel. Commun. 14(6), 3337–3351 (2015)
51. L. Wang, T. Fei, T. Svensson, D. Feng, M. Song, S. Li, Exploiting full duplex for device-
to-device communications in heterogeneous networks. IEEE Commun. Mag. 53(5), 146–152
(2015)