2019 International Conference on Communication Technologies (ComTech 2019)

5G Security Artifacts (DoS / DDoS and

Authentication)
M Awais Javed Sohaib khan Niazi
Information Security Deptt Information Security Deptt
(Researcher) MCS NUST (Researcher) MCS NUST
Rawalpindi, Pakistan Rawalpindi, Pakistan
[email protected] [email protected]

requirements Software Defined Networking (SDN) and

Abstract. 5G will deliver a wider broadband (Network Virtualization Function) NFV will come into play
access, higher and faster mobility, unprecedented to upgrade this network management challenges to simplify
connectivity and coverage to massive matrix of personal the complex network controlling of these billions of devices /
devices, mobility based (vehicular/ commercial UAVs/ nodes [1]. As such networks will be transforming/ upgrading
Drones) and Internet of Things / Everything (IoT / IoE) from 4th generation of cellular standard of Long Term
nodes with ensured availability and optimum reliability. Evolution–Advanced (LTE-A) standard towards 5th
Availability is targeted through multipronged generation based LTE-A cellular standard, so 5G will be
Active/Passive attacks using DOS / DDOS (RF jamming / blending LTE-A protocols with advanced complex
Bandwidth jamming with malformed packets) and networking design to be accommodated within SDN
reliability is targeted through exploiting existing architecture. This blend will result in existing hardware infra
architectural weaknesses of radio and data structure of LTE-A combined with readily available
communication protocols (LTE-A / IP based Networking). networking devices / solutions with optional upgradations of
In prevailing threat landscape, concepts of adoption and SDN and NFV. On other hand interestingly it will also affect
integration of Software Defined Networking (SDN) and the vendor specific monopoly to shift their tilt to software
Network Function Virtualization (NFV) in 5G networks based solutions. Moreover 5G wireless cellular networks with
will open gateways of new networking paradigms with a Internet Service Providers (ISPs) will be switching all services
focus on centralized security dynamics (overall visibility to IP based core networks. In transition of this shift,
and substantial controls). These network security conventional threats are likely travelling in parallel with this
dynamics in diversified use cases (Internet of Everything transition with more sophistication and advanced tools of
/smart grids) will provide an optimum centralized control penetration. This will open gateways of multipronged threats
with a global overview, vigilant monitoring as well as of radio spectrum as well as networking protocol exploitation
prompt reaction (detection and prevention) to these in an IP based core network. This will blend the classical
prevailing threats. In this purview of the 5G cellular computer and network threats with the wireless channel of
security matrix with a centralized SDN and NFV, this User equipment (UEs) /devices and giving more freedom of
document has reviewed on 5G cellular secure operation to attackers due to limited processing power of these
communication channels to unearth the effectiveness of UEs / devices /nodes. However, introduction of SDN based
conventional DOS/DDoS and Authentication attacks 5G networks will centralize the network control through
within environment of SDN and NFV. This research will software programming and virtualize the deployment of the
also present proposed mitigation of DOS / DDOS and core networking policy. SDN networking slicing [2] is the
incorporation of advanced authentication mechanism for virtual deployment of logical / virtual networks as slices
more reliable cellular communication networks. created as and when required within predefined domain of
resources of core networking devices. The concept of Micro-
Keywords—5G Security; SDN; NFV; LTE-A; segmentation [3] has been driven from data centers, however
Communication Channels; DoS/ DDoS; Authentication; its implementation as a network slice in mobile / cellular
Dual-homed Switching Network (DSN) networks are considered viable. In this logical network
segmentation, one complete slice will provide minimalistic
functionality to make a 5G networking cell, while micro
Introduction – 5G Transformation Preamble segments may be limited to few functions within emblem of
5G communications and networks has emerged as promising and the specified programs. Micro segments may be deployed for
an advanced technology for future data communication. 5G fine-tuned isolation, specific access controls and security
networks and systems are ready to hit commercial, public and policies, and their centralized control for application specific
corporate sectors with highly advanced features to its trust models in 5G diversified use cases [4].
predecessors, which includes definitely the higher speeds (upto
Document Outline
10 Gbps), minimum latency (1 millisecond), broader Bandwidth,
massive matrix of interconnected IOT devices, almost 100% Section-I shall explains the multipronged threat landscape
coverage and availability, optimum usage of network energy with security challenges to cellular communications both in
and an extended battery life up to ten years. This complex LTE-A based RF channels and followed by conventional
network matrix of billions of connected devices and smart networking threat statement , possible threats in SDN and
nodes will definitely supersede the concept of conventional challenges in authentication mechanisms. Section-II further
networking. In this futuristic pressing networking elaborate these threats in a cohesive manner to assess their

987-1-5386-5106-3/19/$31.00 ©2019 IEEE

127
Authorized licensed use limited to: University of Sunderland. Downloaded on December 10,2020 at 14:15:38 UTC from IEEE Xplore. Restrictions apply.
effectiveness in a foreseeable future of cellular and IP based messages are in plaintext and has no encryption. Attacker can
networking convergence. Section-III will add possible easily access all type of network configuration of the desired
countermeasures and mitigation techniques. Document will be LTE-A system. This will definitely enhance the attacking
culminated with future research directions and conclusion. capability of an attacker to next level. LTE broadcast message
frame has mappings of physical channels to give an attacker
the initiative of selecting a specific physical channel
I. SECURITY CHALLENGES downlink. Table-I derived from [5] explains all signaling
channels of LTE-A functioning and their respective
vulnerabilities. Table-I give a brief description of various
A. Security Challenges in RF Channels attack vectors on LTE-A communication channels/signals.
As it is imminent that any wireless transmission channel is
prone to intentional or unintentional disclosure and disruption
B. Security Challenges in IP Based Networks
through radio frequency (RF) interference. Here focus is on
intentional or deliberate interference (disruption and sniffing Networks are generally considered quite susciptable to
/ eavesdropping) in LTE-A cellular networks as same unavailability due to easily manipulated attacks of Denial of
infrastructure is going to support upcoming 5G standards. service (DoS) with limited volume of traffic while Ditributed
Denial of Service (DDoS) becomes a higher volume of traffic
to choke the bandwidth either on wire or wireless. Anyway
TABLE-I both are considered an easy to implement and nasty attacks
LTE-A Signals Functions Vulnerability to adversely affect the networking operations. Present day
Cellular
Communication attackers are focusing more on cellular smart devices in
Channel
Synchronization Primary Downlink High power
shape of making part of their botnets with Advanced
Signals channel Synchronization Synchronization transmission can Persistent Threat (APT) due to limited defence capability of
Signals (PSS) signal to UE jam PSS /SSS as
Secondary Provide Physical Cell by default design the cellular device or nodes as well as lack of security
Synchronization identity (PCI) to UE these signals are
detectable at low
awareness in the user. Conversely these compromised UEs /
Signals(SSS)
SNRs devices are very handy in generating DDoS and DoS. In
Downlink Reference Cell –specific OFDM transmission Jamming a carrier
signals channel Reference Signal carries pilot or reference signal addition to such attacks, such UEs and devices are potentially
(CRS) reference signals will yield a higher beneficial as Theft of Service (ToS) and networking
called CRS for synch data error rate by
misaligning the protocol’s misuse / abuse results in degraded Quality of
reference
Downlink Broadcast Physical Broadcast Carries Message PBCH requires low
Service (QoS) (which closely related to from poor
signals channel Channel (PBCH) information Block power jamming to availability of service reaching again to some small scale of
(MIB) and use Cyclic target only 10% of
redundancy Check downlink sub DoS). One such example is malformed text messages to
(CRC) for error carriers and may
detection leads to a very
freeze or crash the targeted device. More likely threats which
efficient are spreading through non legitmate mobile apps updates or
synchronization
attack asking the extention of unprecedented previliges. Once such
Physical Downlink Carries system Ideal for attacker to
Control Channel information Block sniff and extract
updates executed or previlliges granted, these updates
(PDCCH) (SIB) which possess information of the compromises the targeted device and make it a part of a
complete cell cell and network
information and configurations mobile botnet / potential threat to existing cellular and
critical information of
network (e.g, eNodeB
networking operations. All these devices part of a botnet are
idle timer, ever ready to participate in an unaware DOS/DDOS attacks
PRACH,PCLI)
Downlink control Physical Control Contains Uplink and Being a sparse from command and control of the attacker or their services
channel Format Indicator
channel
Downlink resource
allocation information
channel, highly
prone to an
are stolen resulting degradtion of an assured QoS. Either way
(PCIFCH) efficient jamming service provider (LTE-A or ISPs) and normal user/consumer
attack
Hybrid-ARQ Physical Hybrid- Gives Positive and Being a sparse will suffer. Conventional networks attacks has undergone in
Indicator channel ARQ Indicator
channel (PHICH)
negative
acknowledgements of
channel, highly
prone to an
transformation with passage of time. Latest trends of such
uplink packets on a efficient jamming threats are presented in Table-II.
downlink channel attack
Downlink and Physical Downlink Use to transmit date Least important
Generally these attacks are broadly in two classes of flooding
Uplink Use Data shared channel from eNodeB to User threats as requires and logical attacks. Flooding attacks are the first one
(PDSCH) and and from User to a complex attack
Physical Uplink eNodeB combing jamming described in Table-II which are volume based attacks.
shared channel and network details
(PUSCH)
Logical attacks are explained as protocol based attacks or
Uplink Control Physical Uplink Sends uplink control Jamming requires application based attacks which simply exploit the
Channel control channel information (UCI) only 25-30% of
(PUCCH) bandwidth of weaknesses of the programming of the network based
PUCCH
services while the first attack exploits the inherent trust
factor.
LTE-A standard communication is generally divided in two
channels i.e, LTE-A Downlink (from base station (eNodeB) TABLE-II
to User device or equipment) and LTE-A Uplink (from User
device or equipment to base station (eNodeB)). Targeting Network Brief Description
specific DOS /
either way communication channel (Uplink and downlink DDOS Attacks
channels) for jamming and spoofing can be undertaken by an Volume Based
Attacks
These attacks target network bandwidth and consume it in for
undesired resources. Examples of such attacks are UDP Flood attack,
attacker even with limited available capabilities. Moreover it ICMP flood attack etc.
Protocol Based Purpose is to exhaust the network device functioning. Network devices
has also been learnt that LTE-A broadcast transmission Attacks may include servers, firewalls, load balancer and routers etc.

128
Authorized licensed use limited to: University of Sunderland. Downloaded on December 10,2020 at 14:15:38 UTC from IEEE Xplore. Restrictions apply.
 Application Based These are considered to be more sophisticated and serious attacks. switching between authenticating piers at all ends. As the key
Attacks They target the legitimate applications running on machine OS and
exploit the code vulnerabilities of specific applications. Such attacks management protocols in such multi-tier or heterogeneous
include Session Initiation Protocol (SIP) Flood Attack , Browser
application through Hyper Text Transfer Protocol (HTTP) Flood networks with the requirement of a fast and a rapid
Attack authentication modes will be difficult to manage, a secure
context information using SDN platform may be adopted as
C. Security Challenges in SDN mentioned in [9]. The scheme of secure-context Information
(SCI) will use attributes and signatures of physical layer
SDN has built-in monitoring tools in which network flow
information is fed to a centralized controller for analysis to including the MAC address of a device.
detect threats in that particular data flows . Attacks vectors
presented in [6] which will be coming with SDN with their
implementation in mobile networks. The 5G framework will
support SDN and NFV to enhance scalability of network II. COHISIVE OVERVIEW OF THREATS
analysis with additional algorithms like machine learning and LTE-A cellular communications, Network Data
optimum management control. Prevailing security threats will communication and SDN networks require secure channels
be identified with correlation of 5G domains in comparison to for business continuity to impede the prevailing threats.
presently enforced security standards and 3GPP LTE-A Almost all communication channels today preferably deploy
framework [7]. Security critical information is related to IPSec protocol for securing the communication lines. To
network configuration, network condition, data / IP based impart secure tunneling in 5G communication channels,
traffic statistics, cyber-attacks, application specific data IPSec may be implemented with little alterations as described
reports, commercial / consumer data leakage and other
in [10]. Other security algorithms like authentication,
detected incidental reports. 5G security threats are also going
integrity and encryption may also be integrated in LTE
to surge with a greater concern for availability as more
bandwidth means more Theft of Services (ToS). The 5G cellular communications as discussed in [11] to enhance the
wireless channels will provide easy accessibility for security of communication channels. Primarily such security
exploitation by network based jamming attacks (DOS or algorithms will also produce higher overheads as well as
DDOS) and there are no concrete solutions which even exist resource exhaustion, so applying such security solutions in
till date. 5G framework are not considered workable in terms of both
higher throughput and energy efficient requirements
Moreover, multi domain authentications in 5G will be respectively. Thus a higher level of security for critical
implemented as diversified use cases of horizontal and vertical communication is achievable by utilizing novel security
technologies will exist between subscribers, operators and features such a physical layer security parameters (RF
other relevant services providing parties. However,
signatures/ RF fingerprints) [12], using asymmetric security
authentication in 5G will get much more complex with
massive devices connected within these services than it has schemes [13] and dynamic security parameters which can
been imagined. In addition, security automation is also needed alter with the environment [14].
to make the 5G system robust against various security attacks
[4],[8]. MITM attack exploited at different layers to A. SDN Communication Flow Monitoring
compromise data confidentiality, integrity, and availability To monitor data channels in SDN, different methods are
(CIA). In present cellular network, a false base station based
marked by researchers to supervise the flow of
MITM attack forces a legitimate user to create a connection communication e.g, OpenNetMon for monitoring data flows,
with a fake base transceiver station [8]. Mutual authentication
OpenTM to optimize load balancing and to detect anomalies
is introduced to neutralize the false based station attacks. in SDN, OpenSAFE for deeper analysis and altogether all
Unsecured pre-authentication traffic is instrumental in
these technologies are well characterized in [15]. All these
extending such attacks as MITM in cellular GSM services. SDN based appliances are focused on quality and performance
These attacks are successfully implemented on all past cellular
as well as monitoring security threats which are degrading
generations. The basic GSM design specifications lacks
quality and performance. SDN based monitoring framework
network authentication and has an inbuilt potential of MITM has been further proposed and tested by segregating the
like attacks leading straight to compromise of privacy and
functions of monitoring data flows, data distribution, data
confidentiality. inference, and data control. However, it is pertinent to mention
that results of all these appliances revealed that each appliance
performs better in singularity for its optimum output in any
D. Security Challenges in Authentication SDN platform. This is depicting a new picture for application
Authentication session is initiated to authenticate the end based SDNs to streamline the autonomous control in handling
user/service provider or network operator using AKA and decreasing the false positives and negatives. Definitely
(Authentication and Key Management) protocol. After opening new avenues for the demoralized vendor based
authentication session is successfully completed, a session monopolies.
key is generated for the communicating piers for rest of the
communication. Later generations of GSM like UMTS and
LTE-A have adopted the concept of mutual authentication B. SDN Network centralization
rather than user only authentication. AKA (Authentication As SDN network centralization will transform to a virtual
and Key Management) protocol used in 4G LTE cellular control and programming with opening the doorways for
networks are symmetric key based. It is interesting that 5G potential hackers/attacker’s community. For example, the
networks with multi-tier architecture due to massive nodes centralized control is ideal for DoS or DDoS and exposing
connections matrix will be demanding a very fast and rapid whole network to a compromise as mentioned in [16] [17].

129
Authorized licensed use limited to: University of Sunderland. Downloaded on December 10,2020 at 14:15:38 UTC from IEEE Xplore. Restrictions apply.
As the SDN has the ability to control data flowing paths with embedded SIM can easily be exploited using side channel
a centralized controller in various communication channels, attacks and once secret key is revealed to an attacker, he can
this controller will visible to network for its operations and either eavesdrop passively and decrypt the communication or
ultimately lead to an ideal choice for DoS / DDoS attacks. undertake MITM by a simple impersonation of the victim
The centralization of network control can also make the inside the network. Moreover SIM cloning cannot be ruled
controller a congestion in case of saturation attacks on out in such embedded systems.
networks as mentioned in [17]. Moreover any malware once Normally it is customary that a symmetric-key based
entered SDN networking application, will bring down the authentication is used in the mobile networks. The mutual
whole network to a state of disruption and disparity until authentication is initiated and utilized in 3G cellular networks
detected and rectified [18]. In addition, flawed to upgrade the one way authentication procedure in legacy
implementations expand surface of attack vectors with cellular networks. After the authentication, to maintain data
potential RF capabilities or have direct access to the desired confidentiality and data integrity between cellular devices and
networks. Sustainable protection is required to be focused in base stations a cipher key and an integrity key are generated
forthcoming research work to upgrade the classical security to provide security of communication channel on both ends.
paradigm to new generation of data communication In contrary to the cryptographic mutual authentication,
protocols. Even NFV being an integral part of SDN networks, concept of a non-cryptographic fast authentication scheme is
still requires the prime security parameters like introduced in SDN [4], in which weighed secure-context-
confidentiality, integrity and non-repudiation [19]. For information (SCI) transfer to reduce latency with an efficient
mobile networks, these security challenges are presented in authentication during rapid handovers in a HetNet. Physical
layer characteristics are primers of SCI used to enhance the
[20] to show that NFV lacks fundamental security in an
authentication reliability with provision of unique attributes
isolated environment for NFV services. Basic NFV challenge (fingerprints) of the end user. The first user-specific physical
is configuration flaws leading to security vulnerability in a layer attribute is making a statistical model of arriving traffic
cellular mobile network [21]. Further SDN network is as a Pareto distribution in which authentication delay is
completely compromised if the network centralization analyzed with various network scenarios. Scheme of fast
hypervisor is hacked [19]. authentication is proposed which will address authentication
procedure using weighted secured context information (SCI)
C. SDN Based Network Authentication [4]. Once the first full authentication in carried out in one cell
of network, that is easily implemented and propagated in
As 5G networks will have a complex networking ecosystem surrounding / neighboring cells with verification of cellular
evolving to FANETS (Flying ad hoc networks), MANETS device MAC address including physical layer attributes, with
(Mobile Adhoc networks) and VANETS (Vehicular Adhoc local processing.
Networks), HETNETS (Heterogeneous networks) and D2D
(Device to Device) including smart grids, smart infrastructure
and smart cities concepts all converging in one better place of
5G. Thus, such convergence of technologies will be III. CULMINATION STRATEGY
welcoming all black hats to one destinations and contrariwise A. Mitigation Techniques For DoS / DDoS
direly demanding safe and secure networking ecosystem.
Likewise this security of end user will be initiated and It is pertinent to mention that UE chipset is a firmware with
supported with a rapid and instant authentication between end specific cellular technology (GSM/ UMTS/ CDMA etc), so any
piers regardless of who all are switching in which technology upgradation in existing technology of Base stations (eNodeB)
from any other technology with an extra out efforts for will also require the firmware of that cellar technology to be
throughput and latency issues. After a foolproof updated . These changes or upgradations may also require
authentication, these 5G actors will develop a factor of trust to replacement of both ends of service provider and user
further the exchange of sensitive and critical data. As of today equipment. Such changes may be foreseen as reallocation of RF
IPsec tunneling is predominant for secure communication spectrum or LTE-A technology itself. Somehow this greater shift
interfaces used in legacy cellular networks and reason of must consider all the prevailing vulnerabilities and needed to be
predominance is self-explanatory as attacking these interfaces addressed once for all. Cellular Technology vulnerabilities as
require exceptional skills to intrude these interfaces. IPsec role described in Table-I is countered in Table-III, however no
in 5G will not be the same especially once SDN based channel is safer in LTE-A once attacker is determent to disrupt
networking will be enforced, this will open gates for potential the LTE-A network.
threat surface to intrude in SDN. SDN based cellular networks
can be divided in data communication channel, data control Network based attacks are used to be protected through using
channel and internetworking channel as discussed in [22]. anti malware systems , Firewalls , Intrusion Detection /
Presently SDN based networks are deploying TLS (Transport Protection systems (IDS / IPS) followed by Unified threat
Layer Security)/ SSL (Secure Socket Layer) for securing these management products (UTMs). However all these defensive
channels [6]. It is however, important to highlight that systems are conventional techniques failed to dynamically
TLS/SSL are vulnerable to IP based attacks, SDN Scanner changing threat landscape in Next Generation networks
attacks [23] and absence of substantial authentication (NGNs). The main reason of failure was not the effective
procedures to withstand networking attacks [11]. Legacy SIM protection rather diversified nature of multi facet threats
cards will be transforming into embedded SIM card in which attacking the network in any single point of time. However,
users’ credentials are configured through Internet [24]. It is luckily later part of this decade has introduced the concept
important to undertake security penetration testing for this Security Operations Centre (SOC) with a centralized security
embedded SIM technology before deployment in the 5G control platform with long term logging, real-time monitoring
infrastructure. Flawed appliance of such innovation like and analysis of reported threats by using all conventional

130
Authorized licensed use limited to: University of Sunderland. Downloaded on December 10,2020 at 14:15:38 UTC from IEEE Xplore. Restrictions apply.
security systems more effectively and appropriate prevention essentially to meet the increasing demand of isolation
techniques. SOC amalgamation with SDN will further the functionalities SDN and NFV. As SDN will sub divide networks
concept of a robust and proactive security mechanism for both into slices and each slice security may be attributed with SOC
cellular and IP based networking. 5G standards will further for better monitoring and prevention. As discussed earlier these
improve its security architecture and will be able to provide the networking defensive or protective systems are not beneficial if
promising services as discussed in preamble of this document. working in an isolation.

Table-III C. Dual-Homed Switching of Networks (DSN)

LTE-A
Cellular Mitigation Likewise it is proposed that Wi-Fi is now available almost in
Communication
Channel every building and corner of a street and paving its way in all
Downlink Broadcast
signals channel
Spread spectrum techniques like using Dynamic Spread Spectrum
sequence (DSSS) will spread the signals on the all available
future smart infrastructures (smart cities / smart grids). This
Bandwidth (from 6-10 M Hz) all around ubiquitous availability of a wireless medium may
Primary PSS spoofing can be counter by staggering receiver timing for SSS. It
Synchronization may be time dependent and in case whenever time expires, the UE be used more judiciously in parallel of LTE-A
Signals (PSS) should not accept the PSS and select any other stronger cell within the communication channel. As a one of function of SDN is multi
same frequency.
Uplink Control Scrambling the RF spectrum allocated for PUCCH with an encryption domain authentications in 5G which may be utilized to
Channel which will hide the shared sequence of PUCCH and will not be
available on band edges of uplink band. Only an authentic user will
integrate LTE-A and Wi-Fi in a separate network slices for
be allowed to decrypt the scrambling of frame continents. one single UE. LTE-A based communication devices may be
Network Based As networking attacks will degrade services by only option of
Attacks absorbing it. However such attacks mitigation include dropping of IP provided with a Dual-Homed Switching of Networks (DSN)
packets from attacker by using Firewalls / IPS / NGN FW, deflect the
attack to Honeypots / Honey farms , Load balancing services and the
facility. DSN facility will switch to Wi-Fi channel preferably
last option is shutting down the services. when in range of friendly / authenticated Wi-Fi network and
Moreover SDN will be already equipped all such mitigation will substantially decrease the load on LTE-A base station,
technologies as SDN have a better capabilities of analysis for network
traffic flow and behavior. Same may be considered quite effective for
however core cellular network will be aware of UE presence
Network based attacks. Same is explained in more depth in next through attached IP-based network. Service provider will
paragraph of this article.
maintain handover session of all UE devices once entering or
leaving within predefined / friendly identified Wi-Fi
B. Mitigation Techniques For Authentication Misuse networks to and from a base stations (eNodeB). This concept
of DSN if implemented in coming LTE-A networks and Wi-
SDN is characterized with three basic attributes logical Fi networks will provide two manifolds to 5G cellular
intelligent centralization , centralized virtual programming and a standard.
high level of abstraction of complex networking functions [25]. To understand this concept let us suppose whenever UE
All these attributes contributes in improvement in 5G enters in friendly identified Wi-Fi networks, it will prefer
networking key factors in terms of scalability and flexibility and switching friendly/ identified Wi-Fi to connect over LTE-A
ultimately the cost of deployment such networks will Transmission channel once it is available. Now service
significantly reduce. Software Defined mobile Networking provider is in picture that a particular UE has switched to Wi-
(SDMN) architecture comprise in three distinct planes (an Fi IP based network. First manifold is security against
application plane, control plane, and data plane), to integrate compromise of any wireless channel (DoS / DDoS / RF
SDN, NFV and cloud computing all together. Though the key Jamming) that it will greatly enhance the redundancy in case
management is exhaustive with user rapidly switching in a second channel is available for back up. Second manifold is
network cells in 5G, a controller of SDN is optimized to map the rather more cost beneficial by reducing the ever increasing
user locations for predicting his future locations. There are
LTE-A nodes overload in near future and thus will greatly be
various physical layer attributes which constantly being sampled
improving QoS as well as Energy Efficiency (EE) for both
by SDN controller in order to analyze the multiple SCI
combination for prediction. These physical layer attributes are the base station and user devices. This proposal is simply an
fetched in terms of predetermined weighted SCI design of a amalgamation of legacy LTE-A and IP based networking
connecting node with predefined decision rules for joining and with Dual homed Switching Networking (DSN) capability to
leaving the SDN networks. The SDN utilizes Pareto distribution address the already stated multipronged threats in a robust
model for arriving and queuing traffic. Moreover, SDN as a and greener manner as depicted in the Figure-I.
centralized system with global visibility of the users’ locations
and activities as well as the network traffic behavior, desired
policies can be formulated for these roaming security scenarios
and broader network security. Presently NGN / advanced
network security systems include Firewalls systems and
Intrusion Protection systems (IPS) / Intrusion Detection systems
(IDS) can be incorporated and implemented in the flow tables of
SDN for a specific network traffic. In addition, such security
policies needed to be incorporated in the network with a global
view of the network. However, with these SDN and NFV service
delivery model implementation, network vulnerabilities may
also exist [26]. The separation of hardware platform specified
security attributes will definitely increase the threat surface by
decoupling software from hardware [2]. The concept of Network
slicing is produced for isolation of a specific network in [27] and

131
Authorized licensed use limited to: University of Sunderland. Downloaded on December 10,2020 at 14:15:38 UTC from IEEE Xplore. Restrictions apply.
 and ubiquitous wireless channel which is needed to be
overcome by some fundamental changes to provide security
against network based jamming attacks. Such novel security
aspects will also appear in 5G applications wrt HetNet, D2D,
M-MIMO, SDN and IOT, but we touched only SDN side for
an autonomous security and centralized control of 5G ultra
dense networks (UDN). We focused only communication
channel, DOS and basic authentication scenarios in 5G. For
this 5G may implement SDN, NFV and cloud computing to
overcome problems of massive connectivity with flexibility,
network security and definitely the cost factors. While
highlighting inherent security challenges in 5G, some of
security mechanisms and solutions for those specified
challenges are also presented like effectiveness of IPSec
protocol can be enhanced with a little improvement. However,
limited deployment scenarios of these security mechanisms in
5G networks, the potential threat surface cannot be
summarized. Similarly with practical implementation of new
Figure-I 5G single Cell with dense UE while traffic load is application services offered in 5G in line with massive IOT
balanced with concept of DNS (Dual homed switching of devices, the real facets of communication security and privacy
networks) using potential Wi-Fi WPA-3 standards challenges will be highlighted. 5G technologies and services
will bring a new wave of security challenges with on scene
comforts of technology and researchers must prepare to face
Further this proposal may require engagement of both the and address it in a befitting manner.
vendor specific UEs and the service providers. It is also At last not the least, the proposal of dual homed switching
added that Wi-Fi Alliance has announced the next generation network (DSN) proposal is highly recommended for research
of Wi-Fi standard in late 2018 and will ensuring its community for consideration.This paper is to be considered a
availability. In this purview, WPA-3 launching is in progress generalized reviewed effort to these challenges and prevailing
by 2019 which is also adding an additional layers of security threats against security of cellular and networking technology.
and considered it to be much suitable for dual homed
switching of cellular devices. DSN implementation with REFERENCES
SDN may be experimented for real-time analysis with launch
of 5G in production system. [1] N. Alliance, “5G White Paper,” By NGMN Alliance
1.0, p. 124, 2015.
D. Future Directions [2] NGMN Alliance, “5G security recommendations
As discussed in beginning, LTE-A broadcast messages are in Package #2: Network Slicing,” Ngmn, pp. 1–12,
plaintext with their cellular network information available 2016.
further for exploitation and to be considered as open check [3] J. Kindervag and A. Kindness, “Three Technical
for a willing attacker. Same situation prevails for DoS and Innovations Will Ignite Zero Trust,” 2015.
DDoS attacks, however SDN and NFV will be addressing all [4] D. Fang, Y. Qian, and R. Q. Hu, “Security for 5G
available threats in landscape of networking domain. It means Mobile Wireless Networks,” IEEE Access, vol. 6, pp.
that DoS / DDoS and authentication attacks will remain 4850–4874, 2017.
dominant even in coming ages of 5G. Future research shall [5] R. P. Jover, “Security Attacks Against the
consider this paradigm shift of all technologies at one Availability of LTE Mobility Networks : Overview
platform of 5G as an opportunity to come out of this and Research Directions,” no. January 2013, 2017.
persistent threatening environment prevailing both in cellular [6] M. Liyanage, A. B. Abro, M. Ylianttila, and A.
and networking technologies. Innovation together with Gurtov, “Opportunities and Challenges of Software-
intuitions suggest that cellular technology architecture Defined Mobile Networks in Network Security,”
redesigning which is already in progress may be cashed IEEE Secur. Priv., 2016.
optimistically, as in such prevailing threats how to trust the [7] U. M. Tel, “ETSI TS 13 Digital cellular teleco
concepts of evolution as smart grids smart infrastructures and communications system ( Pha 3GPP System A
smart cities? These concepts to be designed in keeping all Architecture Evolution ( SAE,” vol. 0.
such threats at bay to ensure widespread availability without [8] M. Conti, N. Dragoni, and V. Lesyk, “A Survey of
any deniability as well as a guaranteed authentication without Man In The Middle Attacks,” IEEE Commun. Surv.
the worries of a compromise by any stack holder (consumer / Tutorials, vol. 18, no. 3, pp. 2027–2051, 2016.
UE or administrator or a service provider). [9] X. Duan and X. Wang, “Fast authentication in 5G
HetNet through SDN enabled weighted secure-
Conclusion context-information transfer,” 2016 IEEE Int. Conf.
5G networks are anticipated be implemented highly Commun. ICC 2016, 2016.
advanced and innovative features of network connectivity
with automation, virtualization and centralization of [10] M. Liyanage, A. Braeken, A. D. Jurcut, M. Ylianttila,
resources. However, exceptions of security vulnerabilities are and A. Gurtov, “Secure communication channel
also highlighted caused by unsecured re-authenticated traffic architecture for Software Defined Mobile Networks,”

132
Authorized licensed use limited to: University of Sunderland. Downloaded on December 10,2020 at 14:15:38 UTC from IEEE Xplore. Restrictions apply.
 Comput. Networks, vol. 114, pp. 32–50, Feb. 2017. Proc. - 12th IEEE Int. Conf. Comput. Sci. Eng. CSE
[11] M. Y. Ijaz Ahmad∗, Tanesh Kumary, Madhusanka 2009, vol. 3, pp. 353–358, 2009.
Liyanagez, Jude Okwuibex, 5G Security: Analysis of [20] M. M. ; V. K. ; A. Gurtov, “NFV security
Threats and Solutions. 2018. considerations for cloud-based mobile virtual
[12] G. Baldini, R. Giuliani, and E. C. Pons, “An Analysis network operators,” IEEE, 2016.
of the Privacy Threat in Vehicular Ad Hoc Networks [21] B. Yi, X. Wang, K. Li, S. k. Das, and M. Huang, “A
due to Radio An Analysis of the Privacy Threat in comprehensive survey of Network Function
Vehicular Ad Hoc Networks due to Radio Frequency Virtualization,” Comput. Networks, vol. 133, pp.
Fingerprinting,” no. May, 2017. 212–262, 2018.
[13] Caidan Zhao ; Lianfen Huang ; Yifeng Zhao ; [22] M. Liyanage, A. Gurtov, and M. Ylianttila, Software
Xiaojiang Du, “Secure Machine-Type Defined Mobile Networks (SDMN), no. August 2017.
Communications toward LTE Heterogeneous 2015.
Networks,” Publ. IEEE Wirel. Commun. , vol. [23] S. Shin and G. Gu, “Attacking Software-Defined
Volume: 24, no. 1, 2017. Networks: A First Feasibility Study,” Second ACM
[14] M. Liyanage, A. B. Abro, M. Ylianttila, and A. SIGCOMM Work. Hot Top. Softw. Defin. Netw.
Gurtov, “Opportunities and Challenges of Software- (HotSDN 2013), pp. 165–166, 2013.
Defined Mobile Networks in Network Security,” [24] C. Notice and A. Notice, “Remote Provisioning
IEEE Secur. Priv., vol. 14, no. 4, pp. 34–44, Jul. Architecture for Embedded UICC Technical
2016. Specification,” 2014.
[15] F. Z. Y. ; M. G. ; V. F. ; B. G. ; D. von H. ; Bessem, [25] O. N. F. Solution and B. September, “深入研究---
“Network slicing with flexible mobility and Sb-Wireless-Mobile.Pdf,” 2013.
QoS/QoE support for 5G Networks,” IEEE, 2017. [26] M. Dabbagh, B. Hamdaoui, M. Guizani, and A.
[16] I. Ahmad, S. Namal, M. Ylianttila, S. Member, A. Rayes, “Software-defined networking security: Pros
Gurtov, and S. Member, “Security in Software and cons,” IEEE Commun. Mag., vol. 53, no. 6, pp.
Defined Networks : A Survey,” no. January, 2015. 73–79, 2015.
[17] S. Shin, “AVANT-GUARD : Scalable and Vigilant [27] “5G security - enabling a trustworthy 5G system -
Switch Flow Management in Software-Defined Ericsson.” [Online]. Available:
Networks,” 2013. https://www.ericsson.com/en/white-papers/5g-
[18] and P. V. D. Kreutz, F. M. Ramos, “Towards Secure security---enabling-a-trustworthy-5g-system.
and Dependable Software-defined Networks,” ACM [Accessed: 15-Dec-2018].
SIGCOM, 2013.
[19] A. Cleeff, W. Pieters, and R. Wieringa, “Security
implications of virtualization: A literature study,”

133
Authorized licensed use limited to: University of Sunderland. Downloaded on December 10,2020 at 14:15:38 UTC from IEEE Xplore. Restrictions apply.