Page 19 - ITU Journal Future and evolving technologies Volume 2 (2021), Issue 6 – Wireless communication systems in beyond 5G era

ITU Journal on Future and Evolving Technologies, Volume 2 (2021), Issue 6

and the core network, with 5G, the latency switches to • “[...] The appropriate psychological measures will be
an end‐to‐end perspective. As previously mentioned, the dependent on the communication context. Objective
5G network has a softwarized architecture thus, the net‐ psychological measures do not rely on the opinion of
work functions can run on servers that belong not only the user (e.g. task completion time measured in sec‐
to operators’ computing hardware but also to data cen‐ onds, task accuracy measured in number of errors).
tres in the Internet, which are networks themselves. So, Subjective psychological measures are based on the
as a consequence, the evaluation of the KPI should be on opinion of the user (e.g. perceived quality of medium,
an end‐to‐end basis. That is also the case of energy ef i‐ satisfaction with a service). [...]” [23].
ciency. Energy iciency “[...] is ined as the number
Next, the second de inition states that the QoE is the “[...]
of bits that can be transmitted per Joule of energy, where
overall acceptability of an application or service, as per‐
the energy is computed over the whole network, including
ceived subjectively by the end‐user [...]” [23]. It is im‐
potentially legacy cellular technologies, Radio access and
portant to notice that the QoE is an end‐to‐end metric,
Core networks, and data centres. [...]” [17]. Given this, it
which can also be affected by users’ expectations and con‐
appears clear that the virtualisation of the RAN can maybe
text. According to the above initions and considera‐
improve its energy ef iciency but, on the other hand, it can
also reduce it within the core network or the data centres. tions, 5G clearly arises more dif iculties in the evaluation
of communication performances. This is not only due to
its end‐to‐end communication perspective and its hetero‐
Next to KPI, another important metric of communication
geneity (as an ecosystem of heterogeneous network ar‐
networks is quality. In particular, there can be three
possible metrics to measure ’quality’: QoS, Quality‐of‐ chitectures, communications, and services), but it is also
Perception (QoP) (or user‐perceived QoS), and Quality‐of‐ due to the prominent and wide role of subjective metrics,
Experience (QoE). Network QoS is the “[...] degree of con‐ which had very narrow and limited impact in previous
generation networks and in the Internet.
formance of the service delivered to a user by a provider
with an agreement between them [...]” [23]. Next, the QoS
is the “[...] totality of characteristics of a telecommunica‐ 3.2 5G architectural characteristics
tions service that bear on its ability to satisfy stated and
5G has been the irst communication network trying
implied needs of the user of the service [...]” [23]. The
to inherently incorporate the network virtualisation
QoS relies on technical metrics so it is technology‐centred. paradigms, together with cloud computing. Moreover, 5G
The QoP “[...] is primarily concerned with the detectabil‐
started combining SDN and NFV, following the guidelines
ity of a change in quality or the acceptability of a quality
of the ETSI MANO SDN‐NFV architecture (see Fig. 3). In
level. [...]” [23]. As an example, the use of Mean Opinion parallel, the concept of SDR was also included within 5G in
Score (MOS) measures the ’perception of quality’ accord‐
order to have an end‐to‐end softwarized/virtual network,
ing to a subjective rating.
making the radio access technologies more lexible and
r igurable. In this end‐to‐end virtual network con‐
The target of 5G has also been the support for verticals, in
text, the virtual network functions have become service
which users are immersed as ’protagonists’ (e.g. Tactile
Internet, Augmented/Virtual Reality, etc.) [24]. This ’im‐ functions, which process ingoing information at the vari‐
mersion’ of human end users in the communication ser‐ ous layers of the protocol stack. Next, service functions
can be dynamically ordered into ic logical chains,
vices has made the QoS an incomplete metric. Because of
according to the network tasks that have to be performed
that, the concept of QoE appeared, which can be de ined
on communication messages. More precisely, a Service
as QoE, or user‐level or user‐assessed QoS. The QoE relies
on end‐user behaviour/perception so, it is user‐centred. Function Chaining (SFC) is ined as “[...] an ordered
However, the QoE is a harder metric to quantify since it set of abstract service functions and ordering constraints
involves subjective evaluation. that must be applied to packets and/or frames and/or
lows selected as a result of classi ication. [...]” [25].
In the literature, two de initions of QoE can be found. The
irst states that QoE is a “[...] measure of user perfor‐ The combination of all these technologies, that have been
mance based on both objective and subjective psychologi‐ mentioned, has enabled the adaptive‐optimal‐ lexible al‐
cal measures of using an ICT service or product. [...]” [23]. location of resources, but also the possibility for their
For this de inition, ETSI particularly mentions two notes: logical isolation. This has opened the way for another
communication network paradigm called end‐to‐end net‐
• “[...] It takes into account technical parameters (e.g. work slicing [1], [26]–[29]. Network slices are logical
QoS) and usage context variables (e.g. communica‐ and ’isolated’ virtual networks that can be instantiated
tion task) and measures both the process and out‐ on a unique physical network infrastructure. These vir‐
comes of communication (e.g. user effectiveness, ef‐ tual networks are generated by the allocation of subsets
iciency, satisfaction and enjoyment). [...]” [23]; of available physical resources and by the set of policies
to identify the speci ic traf ic lows, that are going to be
hosted. In fact, slices can be created to manage and oper‐
ate multi‐tenant communication scenarios or to address

© International Telecommunication Union, 2021 7