Content uploaded by Dimitris Giannopoulos
Author content
All content in this area was uploaded by Dimitris Giannopoulos on May 23, 2022
Content may be subject to copyright.
DRAFT
Providing Real-Time, High-Quality User and
Machine Generated Content over a 5G Network
Giannopoulos D., Papaioannou P.,
Tranoris C., Denazis S.
University of Patras, Patras, Greece
dimit.giannopoulos@upnet.gr,
papajohn@upatras.gr,
tranoris@ece.upatras.gr,
sdena@upatras.gr
Altman B., Rahav R.
LiveU
Baruch@liveu.tv
ran@liveu.tv
Margolin U.
Nokia
Kefar Sava, Israel
udi.margolin@nokia.com
Abstract—Fifth Generation (5G) technologies alongside
Network Function Virtualization (NFV) are revolutionizing
the way live events are covered. In this Demo we look at how
this can be done today.
Keywords—Network Function Virtualization, 5G, Media
Content Production
I. OVERVIEW
The present Demo focuses on the media-related
experience in generating professional and semi-
professional content for various purposes. These may
include live news coverage, live sports and other
entertainment coverage, telemedicine related live
transmission, such as video from rural medical centers to
medical experts or live from moving ambulances, real
time security cameras, automotive-generated content,
such as multiple cameras enabling teleoperation or
uploading huge amounts of sensory-generated data.
For the broadcast and news coverage transmissions, this
continues the application of replacing the traditional
satellite trucks (disruptive) while allowing many more
use cases (e.g. news coverage from indoors, from
underground, at no-time notice, on the move, from
drones, etc.). When either very reliable transmission is
needed or high video quality (in the past full HD,
nowadays 4K and in the coming years even 8K), and/or
during mobility, then multi-link bonding may be used.
Fifth Generation (5G) technology is required in order to
solve the issues of the current on-site production by
allowing to do the production itself remotely, in the
studio or in the cloud or distributed, while sending to the
field only the cameras and the cameramen. 5G is
expected to provide higher uplink (UL) bandwidth,
consistency, latency sustainability and overall UL
“guaranteed” Service Level Agreement/Quality of
Service (SLA/QoS) for multiple UL cameras/feeds. This
is expected to be done with SA with slices/services, PN
and NPN networks.
In this Demo, the professional or semi-professional
content generator minimizes the risk for transmission
failure from anywhere and regardless of potential
degradations in any single network, aggregates
bandwidth from one or more networks and also tries to
aggrege bandwidth from two 5G network slices of the
same Mobile Network Operator, to provide professional
grade transmission.
Network slicing is a method of creating multiple unique
logical and virtualized networks over a common multi-
domain infrastructure. Using Software-Defined
Networking (SDN), Network Functions Virtualization
(NFV), orchestration, analytics, and automation, Mobile
Network Operators (MNOs) can quickly create network
slices that can support a specific application, service, set
of users, or network. Network slices can span multiple
network domains, including access, core, and transport,
and be deployed across multiple operators [1].
Testing this multi-slice bonding depends on availability
of multiple network slices at the University of Patras
(UoP) 5G-VINNI. 5G Verticals INNovation
Infrastructure (5G-VINNI) is a large-scale, end-to-end
facility providing advanced 5G capabilities that can be
accessed and used by vertical industries for use case
trialing [2]. The 5G-VINNI facility in Patras is part of the
Patras Platform for Experimentation [3]. The Patras 5G-
VINNI facility site has its own management and
orchestration stack. This stack includes a Service
Orchestrator, taking care of the lifecycle management of
provided network slices at the application layer, i.e.,
network slice semantics, a Network Function
Virtualization (NFV) Orchestrator (NFVO), which
deploys and operates provided network slices at the
virtualized resource layer and a Virtualized Infrastructure
Manager (VIM), which manages the virtual deployment
units building up the cloud execution environment. The
Patras 5G-VINNI facility site utilizes Openslice as a
service orchestrator, the ETSI Open-Source MANO
(OSM) [4] as a NFVO and OpenStack [5] as a VIM.
Additionally, in this Demo, we utilize a Cross-Domain
Service Orchestrator (CDSO), developed by Nokia,
which is responsible for managing and orchestrating the
various services across multiple domains, in the case
where the service extends beyond the Patras 5G-VINNI
facility site to multiple domains.
Openslice [6] is an open-source, operations support
system (OSS) solution providing Service Orchestration
functionality, including both service fulfilment and
assurance lifecycle phase. From a customer-facing
viewpoint, Openslice defines a user-friendly web portal
that allows managing (e.g., authorization, authentication)
the interaction with vertical customers, capturing their
service orders and keeping them informed about the
status of the network slices, which their ordered services
are hosted on. From a resource-facing viewpoint, it
DRAFT
interacts with the NFVO, consuming SOL005 exposed
capabilities to deploy and operate the virtualized
components of the network slice.
II. INNOVATION
Cooperative Media Production, or At-Home Production
or Remote Integration (REMI) has become the new
standard for production companies, broadcasters, and
sports organizations of all sizes. Wireless At-
Home/Cloud Production solutions allow broadcasters to
reduce costs by producing live shows from a centralized
studio control room instead of on-site production and
satellite trucks. These on-site trucks involve huge costs
both in terms of duplicated equipment and personnel, in
very low efficiency due to the overhead involved such as
travel time of the teams and the equipment, quality
variance between events, production faults, lower
efficiency in multi-site simultaneous events management
etc. At-home production solves all of these inefficiencies
and reduced support of live events by allowing sending
only the camera and audio teams to the field, whereas the
production is done either in the professional studio
facility or anywhere else, as seen fit. Sports and event
producers can deliver multi-camera live events while
eliminating the need to spend a fortune on production
vehicles, satellite uplinks and travel expenses.
III. RELEVANCE
In this Demo, we showcase how 5G alongside
virtualization technologies are revolutionizing the
coverage of live events. In an era, where handheld smart
devices are ubiquitous, and unprofessional event
coverage has reached new levels of quality, it is only
logical that professionals aim to raise the standard. In
order to do this, modern technologies must be adopted,
and we believe there are many lessons to be learned for
all the parties involved.
IV. PROPOSAL SUMMARY
In this demo several video streams (e.g. from cameras or
other streaming devices) are each connected in the field
to cellular-based transmission devices, including
bonding devices to provide the utmost reliability and
bandwidth. As shown in Figure 1, these field devices then
transmit the video stream over the cellular (5G) and the
standard public internet (ISP) to a single
receiving/decoder server with several physical SDI
outputs on the 5G-VINNI Patras facility site. The
content, thus, reaches the LiveU server and is then
broadcast live on YouTube. The 5G network is
monitored and metrics are shown on a display in the lab,
side by side with the actual content. The entire network
service deployment is initiated by Nokia’s CDSO.
LiveU provides several bonding video encoders-
transmitters which use its Precision Timing feature to
allow synchronization of the video streams at the
receiving end – the remote production software. The
multiple video streams in the station or production
facility are used for the actual remote production. Video
quality is up to 4K from each camera.
ACKNOWLEDGEMENTS
This work is supported by the H2020 European Projects
5GVINNI (grant agreement No. 815279) and 5G-
SOLUTIONS (grant agreement No. 856691).
V. REFERENCES
[1]
"What Is Network Slicing?," [Online]. Available:
https://www.blueplanet.com/resources/what-is-network-
slicing.html.
[2]
K. e. a. Mahmood, "Design of 5G End-to-
End Facility for
Performance Evaluation and Use Case Trials," in
2019 IEEE 2nd
5G World Forum (5GWF)
, 2019.
[3]
C. Tranoris and S. Denazis, "Patras 5G: An open Source based
End-to-End facility for 5G trials," ERCIM
NEWS, Special Theme
5G, Number 117, p.10., April 2019.
[4]
"ETSI Open Source MANO (OSM)," [Online]. Available:
https://osm.etsi.org/.
[5]
"OpenStack," [Online]. Available: https://www.openstack.org/.
[6]
C. Tranoris, "Openslice: An opensource OSS fo
r Delivering
Network Slice as a Service," 2021.
Figure 1 5G Solutions LiveU Video streaming testbed at UoP
