TECNOLOGÍA DE RADIO 4G LTE

CAPÍTULO I: CONTEXTO DE LOS SERVICIOS Y APLICACIONES DE LTE

Service Evolution: from 3G to 4G
Services is often used to refer to inherent core and access network characteristics that support a number of actions
taking place in and around the network.

Applications refer to user-centric software programs that use the services inherent in the network to enable the user
to interact with the software.

Example: networks hold certain information about the user (such as location, operational status, credit available) that
can be provided as a service to be used by applications (such as mapping, real-time messaging, live video streaming and
payment applications).

As networks evolve, they are becoming capable of supporting many more advanced services, which in turn impact the
types of applications that can be designed. Furthermore, often the provider of the network services is different from
the provider of the applications, which has profound implications for the monetisation of networks and the
management of end-user relationships.
 Evolution from WCDMA to HSPA+
HSPA (High Speed Packet Access) and HSPA+ (HSPA Evolved)
HSPA (High-Speed Packet Access) is often referred to as 3.5G, and HSPA networks have been widely deployed since 2006.
Many new features were introduced with this technology to increase capacity and efficiency, and reduce latency.
Enhanced voice and
Broadband Broadband data capacity
Rel-5 Rel6 Rel-7 Rel8
Rel-99
(HSDPA) (HSUPA) (HSDPA)HSUPA)
WCDMA
HSPA HSPA+
DL: 384 Mbps DL: 14.4 Mbps DL: 14.4 Mbps DL: 42 Mbps
UL: 384 Kbps UL: 384 Kbps UL: 5.76 Mbps UL: 11 Mbps

HSPA comprises two technologies: the first phase, HSDPA (High-Speed Downlink Packet Access), and the second phase,
HSUPA (High-Speed Uplink Packet Access), which together are known as HSPA. All WCDMA (Wideband Code Division Multiple
Access) operators have now launched HSPA.
HSPA+ is a 3GPP (3rd Generation Partnership Project) Release 7 evolution of the HSDPA and HSUPA technologies that further
enhances both uplink and downlink speeds. It is sometimes referred to as 3.75G.
 HSPA (High Speed Packet Access) and HSPA+ (HSPA Evolved)
HSDPA
HSDPA focused on increasing the downlink capacity for asymmetric applications such as Internet access, TV, video-on-
demand, and rich media file downloads, for example MP3.
HSDPA was designed to be approximately five times faster than UMTS Rel. 99 (3G). The theoretical maximum downlink
speed for HSDPA is 14.4 Mbps, although the actual speed achieved depends on a number of other factors, such as device
capabilities and cell capacity. Operators are offering subscribers practical data rates in the range of 1–2 Mbps.

HSUPA
HSUPA extends the capacity of the uplink-dedicated transport channels to support symmetrical applications (sometimes
called ‘full duplex’ services). The main aim of HSUPA is to improve the performance of dedicated uplink transport channels
by increasing capacity and throughput and reducing delay. It offers uplink data speeds of up to 5.76 Mbps and achieves its
high performance through more efficient uplink scheduling in the base station and faster retransmission control. Six new
physical channels have been introduced to support HSUPA.
HSUPA enables a number of applications that require symmetrical fast data rates, low latency and real-time, interactive
participation of users, for example, person-to-person online gaming (sometimes called multiplayer gaming), personal mobile
web, mobile e-mail with attachments and high-quality video-conferencing.
 HSPA (High Speed Packet Access) and HSPA+ (HSPA Evolved)
HSPA+ : Features and Benefits of HSPA+
The theoretical maximum speeds of HSPA+ (168 Mbps downlink and 22 Mbps uplink) are rarely achievable in practice;
however, some operators are offering actual downlink speeds of 21 Mbps and higher.
The main improvements are gained through the use of high-level modulation such as QAM (Quadrature Amplitude
Modulation) and MIMO (Multiple Input Multiple Output) antenna systems. These systems enable even greater spectral
efficiency than HSxPA and a corresponding improvement in average rates per cell. This means that operators can support
greatly increased data traffic without having to deploy new cell sites.

Higher-Order Modulation Continuous Packet

Downlink 2x2 MIMO
(64QAM) Connectivity (CPC)
Double downlink capacity Increases uplink and Improves VoIP capacity,
and pealk rates downlink data rates extends talk time and give a
better experience

Single Frecuency
Multi-Carrier Mode Advanced Receivers
Netowork for Broadcast
Increases capacity and data More efficient, greater Increases broadcast capacity
rates capacity and higher data
rates.
 HSPA (High Speed Packet Access) and HSPA+ (HSPA Evolved)
In addition to improving the radio network
performance, Rel. 7 provides for a ‘flat’
network architecture that incorporates a
standalone Node B, which integrates certain
RAN (Radio Access Network) functions. The
Node B sends UP (User Plane) data (customer
traffic) directly to the PS (Packet Switched)
network nodes such as the SGSN (Serving GPRS
Support Node) and GGSN (Gateway GPRS
Support Node), bypassing the RNC (Radio
Network Controller). The key benefit of this
revised architecture is that it further reduces
the network latency for customer traffic.
Furthermore, the flat architecture inherent in
HSPA+ provides a natural migration path to
LTE/SAE.

Iu-PS(User Plane)
Iu-PS(Control Plane)
Limitations of 3G/3.5G Networks and Requirements of Mobile Broadband
Networks

Complexity

Lack of Support
Cost for Real-Time IP-
based
Applications
Limitations of
3G/3.5G Services

Congestion/
Data Rates
Contention

Latency
Complexity and Cost
Most mobile networks have been designed in a hierarchical way that makes them extremely complex
and, in some instances, over-engineered. This increases the costs of network design and manufacture, Cost
as well as generating ongoing network maintenance costs.

Data Rates
Although higher data rates achieved through 3G and 3.5G have undoubtedly improved the user
experience of so-called ‘high-bandwidth’ applications in both the uplink and downlink, they have
mainly been designed as capacity add-ons for existing networks (either CDMA or GSM). However, the
theoretical maximum data rates available on HSPA+ networks do not support a paradigm shift away Data Rates
from a network-centric approach to applications design towards a user-centric approach. This is
likely to continue to constrain the way that users are able to benefit from mobile broadband in terms
of the types of applications that are available.
Congestion/Contention
Although operators have invested billions in expanding the capacity of their mobile-data networks,
this has not necessarily resulted in an optimum user experience or in the best utilisation of network
resources. In some extreme cases, as expanded capacity has come on stream to cope with increased Congestion/
user demand through new device types, this has increased traffic to the extent that the networks Contention
have become highly congested, resulting in a sub-optimal user experience.
 Limitations of 3G/3.5G Networks and Requirements of Mobile Broadband
Networks

Latency
Even with HSPA+ networks, latency can be an issue, especially at cell edges. The problem becomes Latency
even worse when mobility speeds accelerate, for example with in-vehicle mobility.

Lack of Support for Real-Time IP-based Applications

The architectural complexity of 3G and 3.5G networks, which have been designed to support both CS
and PS protocols, means they have become, by their very nature, centralised and hierarchical. This is Lack of Support
at odds with the network service needs of users’ behaviour, where multiple classes of service may be for Real-Time IP-
required simultaneously depending on the types of applications the user is accessing at any one based
Applications
time. A native IP (Internet Protocol) environment is the only way that such applications can be
supported, and this is not feasible with existing 3G/3.5G networks.
 Requirements of Mobile Broadband Networks
External switch/router
is needed for co-
siting ?

Multipoint-to-
Adjacent eNBs need to Multipoint is more
GPS is always Subscriber
have direct links (fully suited to flat
needed ? security is
meshed network) ? architecture ?
sufficient ?
MME

Ethernet / IP
Serving Gateway

High-performance Transport capacity All transport

IPSec requires enodeB has to be nodes have
separate >500Mbit/s per to be IP
equipment ? eNB ? Transport delay routers ? O&M
has to be <5ms ?
 Requirements of Mobile Broadband Networks
True 3.9G/4G mobile broadband networks need:
 Vastly lowered latency in the network.
 Greater network capacity, with increased downlink and uplink peak data rates as well as improved customer
throughput.
 Bandwidth that is scalable in both the uplink and downlink, reducing overheads, improving network efficiency and
allocating network resources more effectively.
 Improved spectral efficiency, so that overheads are lowered and radio resources can be allocated more effectively
according to the type of application being used.
 Effective operation at different mobility speeds, depending on the user context.
 Interactivity with legacy networks.
 Provision of support to a wide range of devices and applications within an open ecosystem

These requirements cannot be met by incremental changes to existing 3G/3.5G networks, since the underlying
network hierarchy does not enable high enough data rates and low enough latency.
 LTE SERVICE ARCHITECTURE
The title Long Term Evolution points to the fact that 3GPP identified some time ago the requirement to develop a RAT
(Radio Access Technology) that extends well beyond the capabilities of HSPA. Under the title E-UTRA (Evolved Universal
Terrestrial Radio Access), 3GPP stated that ‘The objective of Evolved UTRA and UTRAN is to develop a framework for the
evolution of the 3GPP radio-access technology towards a high-data-rate, low-latency and packet-optimized radio-access
technology’.

LTE embraces the concept of a ‘flatter’ network architecture – that is, the strict hierarchy of network elements often
found in legacy networks is not required. A flat architecture, with simpler network nodes, means that there should be
fewer ‘hops’ in any connection, which aims to reduces latency and cost while improving overall network performance.
This flat design, as well as the enhanced functionality of LTE, enables the network to deliver services that support the
design and implementation of advanced mobile broadband applications.

LTE focuses on optimum support for packet-switched services. The main requirements for the design of an LTE system can
be summarised as follows:
 LTE SERVICE ARCHITECTURE
 Data Rate: Peak data rates target 100 Mbps (downlink) and 50 Mbps (uplink) for 20 MHz spectrum allocation, assuming
two receive antennas and one transmit antenna at the terminal.

 Throughput: The target for downlink average user throughput per MHz is three to four times better than Release 6
(HSPA). The target for uplink average user throughput per MHz is two to three times better than Release 6.

 Spectrum Efficiency: The downlink target is three to four times better than Release 6.

 Uplink target: The uplink target is two to three times better than Release 6.

 Latency: The one-way transit time between a packet being available at the IP layer in either the UE or RAN and the
availability of this packet at IP layer in the RAN/UE is less than 5 ms. Also, control plane latency is reduced.

 Bandwidth: Scalable bandwidths of 5, 10, 15, 20 MHz are supported. Also bandwidths smaller than 5 MHz are supported
for more flexibility, i.e. 1.4 MHz and 3 MHz for FDD (Frequency Division Duplex) mode.

 QoS: End-to-end QoS is supported.

 LTE SERVICE ARCHITECTURE
 Interworking: Interworking with existing UTRAN/GERAN systems and non-3GPP systems is ensured. Multimode
terminals support handover to and from UTRAN and GERAN as well as inter-RAT measurements. Interruption time for
handover between E-UTRAN (Evolved UTRAN) and UTRAN/GERAN is less than 300 ms for real-time services and less
than 500 ms for non-real-time services.

 MBMS (Multimedia Broadcast Multicast Service): MBMS is further enhanced and is referred to as E-MBMS.

 Mobility: The system is optimised for low mobile speed (0–15 km/h), but higher mobile speeds are supported, and
high-speed train environment speeds are included as a special case.

 Spectrum allocation: Operation is in paired spectrum (FDD mode) and unpaired spectrum (TDD [Time Division Duplex]
mode).

 Co-existence: Coexistence in the same geographical area and co-location with GERAN/UTRAN is provided for, as well as
coexistence between operators in adjacent bands and cross-border coexistence.

Other targets identified that the technology should be a cost-effective upgrade for network operators from existing 3GPP
networks.
 LTE APPLICATIONS AND ECOSYSTEMS
Defining End-User Requirements
The ITU (International Telecommunication Union) has established that applications offered via next-generation mobile
broadband networks should take account of various market-driven requirements. These include the following:

 Affordability – users will only buy mobile services that are affordable, given their individual economic circumstances.
 Speed – applications such as wireless video need networks that are fast enough to deliver acceptable performance in
terms of anytime, anywhere availability in a manner that is transparent to the end user.
 Convergence – applications need to be available across multiple networks and devices, including across voice/data
networks, fixed/mobile infrastructure and communications-centric/applications-centric devices or terminals.
 Personalisation – increasingly, users will value the ability to tailor or customise their mobile services to their own
requirements at any time and in any context, without having to subscribe to a standardised or packaged service.
 Quality – depending on the environment where the applications are used, different quality profiles may be required for
the same service/user, while always maintaining an acceptable level of quality according to the application being used.
 LTE APPLICATIONS AND ECOSYSTEMS - End-User Requirements

 Simultaneity – users may wish to use several applications at the same time (such as voice, photo upload and mapping
information), and even across several synchronised devices, and the networks must be able to facilitate those
requirements.
 Context – the network must be aware of the environment within which the user is performing the various tasks
undertaken, so that it automatically configures services in line with changing contexts rather than the user having to do
this manually.
 LTE as an Applications Enabler
To support user-determined, always-on,
on-demand, intelligent and limitless
access to mobile broadband applications,
the architecture of next-generation
networks must support a lower cost per
megabyte, as well as being highly scalable
and flexible to support a wide range of
user requirements. Furthermore, the
network must support access to multiple
types of content by many different types
of device simultaneously.
Key to achieving these objectives is
reducing network latency to support high-
data-rate symmetrical user applications,
as well as improving network efficiency so
that the right amount of bandwidth is
allocated to the right application at the
right time.
 LTE Applications - Voice Applications
Conversational Applications and Service Classes
The list of potential voice-based applications can be divided into six principal categories, which are described in the
table.

There is little doubt that voice will remain the most popular mobile application for many customers for several years to come.
How soon the more advanced rich conversational applications come to fruition will depend on user perceptions of the value
they receive from such applications, and on the business models that will emerge for the monetisation of advanced mobile
broadband applications.
 LTE Applications - Voice Applications
Voice in the LTE Environment

A further option, VoLGA (Voice over LTE Generic Access), was specified by the VoLGA Forum, a consortium of industry specialists, but has
not found favour with operators, primarily due to cost, and has been largely dropped.
 LTE Applications - Voice Applications
Voice in the LTE Environment
Dual-Mode Services
In the dual-mode scenario, the UE contains two modems: one for CS voice, the other for data, which is handled by LTE.
Verizon was the first operator to launch these dual-mode handsets. With this solution there is little disruption to the
network, but the size of the handset and the drain on battery life are major limitations. This solution is therefore expected
to be short term.

CSFB (Circuit Switched Fall Back)

CSFB enables the provision of voice, SMS, LCS (Location Services) and USSD (Unstructured Supplementary Service Data) by
legacy CS equipment. A UE that is attached to the E-UTRAN and that wishes to make a CS call will make use of the UTRAN or
GERAN to connect to the legacy CS domain. A speech call requires that the E-UTRAN coverage area will be overlapped by
UTRAN or GERAN coverage. (SMS does not require this).

CSFB functionality is realised by using the SGs interface between the MSC server and the MME in the EPC.
When the UE first registers with the LTE network it will indicate that a combined update is required. This instructs the MME
to pass the registration messages to the 2G/3G network. This method does not require major changes to the network
architecture.
 LTE Applications - Voice Applications
Voice in the LTE Environment

For incoming CS calls, the MSC will signal the MME via the SGs interface. The paging message can then be forwarded
directly to the UE if it is currently in an active state, otherwise an LTE paging procedure will be carried out and the CS
paging information can be forwarded to the UE once the radio connection is re-established.
The UE will then need to select a 2G/3G cell. The eNB may instruct the UE to take measurements from the 2G/3G network
in order to determine which cell to use, or it may blindly send instructions about a preconfigured cell, which may occur
when the LTE and 2G/3G cells are co-located. The paging response from the UE is carried out in the new 2G/3G cell.
 LTE Applications - Voice Applications
Voice in the LTE Environment

Iu-PS

Gs
Gb
Iu-CS
Uu A
S3
Um SGs
LTE-
Uu
S1-MME

Disadvantages to CSFB include (as well as the need to implement the SGs interface) a longer call set-up time, which
can be up to 1.5 seconds, and the need to modify the MSC nodes.
 LTE Applications - Voice Applications
VoLTE (Voice over LTE)
VoLTE provides a standardised means of transferring voice traffic over the LTE network by means of an IMS. It is 3GPP’s
preferred method of providing voice services, and has been specified as such since Release 8.

The IMS is ideally suited to delivering CS services over an IP network, as it uses peer-to-peer operation. Although the UE
remains connected to the EPS throughout the session, the IMS will establish dedicated EPS bearers to carry the traffic – SIP
signalling, voice and/or data – end to end across the E-UTRAN and the EPC. The IMS deploys a solution known as MMTel
(Multimedia Telephony) to provide the speech services (as well as other services such as video calling, chat, etc.) in the LTE
network.

The IMS is network-agnostic in that it is not concerned whether it is operating over an E-UTRAN and EPC or any other IP-
CAN or PDN. However, both the LTE RAN and the EPC need to support VoLTE, which is achievable through software
upgrades.

The aim is that full VoLTE provision should eventually meet the requirements for evolved mobile broadband networks: good
QoS, roaming, global interoperability and seamless mobility, over any access network and between any device. Handover to
non-LTE networks is supported by means of SRVCC (Single Radio Voice Call Continuity) technology.
 LTE Applications - Voice Applications
VoLTE (Voice over LTE)

The main problem is that an IMS is expensive to implement, so operators who do not already have an IMS in place are
facing considerable investment in order to roll out VoLTE.
 LTE Applications - Voice Applications
OTT (Over the Top) VoIP

An additional approach, and one that has not been initiated by operators, is the use of OTT services. In the OTT scenario,
network operators only provide the connectivity to end users: the voice service is delivered by a service provider, using
applications such as Skype, Viber and Google Talk. This trend has increased since 2011 and is causing operators some well-
grounded concern. Clearly, operators are keen to avoid this no-revenue business model and which represents a
cannibalisation of their own services. The problem becomes particularly prevalent for out-of-bundle minutes, which can be
very expensive to subscribers.

Some operators are attempting to combat the OTT threat by neutralising the effect, for example by offering unlimited
voice, unlimited messaging, and/or unlimited data packages. Others are emulating OTT players by offering OTT-like services,
for example O2 with their TU Go service, which enables cross-device number portability for voice, text and voicemail; and
Orange’s Libon (developed by France Telecom’s subsidiary, Orange Vallee), which allows users to call and send instant
messages for free. A few operators are partnering with OTT brands, for example KDDI are partnering with Skype.
OTT (Over the Top) VoIP

Example: Viber

 200 million registered users – May 2013

 140 million registered users – Dec 2012
 400,000 new users each day
 All operating system suuport :Android, iOS, BlackBerry, Windows Phone,
Windows 8, Bada, Nokia.
 Some operators block the use of Viber, ….
 Viber is doing to mobile what skype did to fixed-line telephony in 2013..

Services such as Viber are currently available only over 3G and Wi-Fi networks. But it is worth
noting the rapid growth of Viber subscriber numbers; as of May 2013, Viber had 200 million
users worldwide, up from 140 million at the end of 2012. The threat to operators from such
OTT players is real, especially as users progress from downloading such apps not just because
they are free, but – more dangerously – because they perceive that they are better than the
operator’s standard service.
OTT (Over the Top) VoIP

Example: Viber

 200 million registered users – May 2013

 140 million registered users – Dec 2012
 400,000 new users each day
 All operating system suuport :Android, iOS, BlackBerry, Windows Phone,
Windows 8, Bada, Nokia.
 Some operators block the use of Viber, ….
 Viber is doing to mobile what skype did to fixed-line telephony in 2013..

Services such as Viber are currently available only over 3G and Wi-Fi networks. But it is worth
noting the rapid growth of Viber subscriber numbers; as of May 2013, Viber had 200 million
users worldwide, up from 140 million at the end of 2012. The threat to operators from such
OTT players is real, especially as users progress from downloading such apps not just because
they are free, but – more dangerously – because they perceive that they are better than the
operator’s standard service.
 Evolved Packet Core
SAE (System Architecture Evolution) – 3GPP

Unlike preceding networks, the EPC has no circuit-switched element. It has been designed to support both
real- and non-real-time packet-based services, hence is IP from end to end.

Across the whole of the SAE, the architecture is said to have been ‘flattened’ – that is, fewer nodes are
defined in the standards, which as was noted above means fewer ‘hops’. In practice, however, operators
may find that this 3GPP vision of a simplified architecture may be difficult to achieve – at least initially.

Whereas in UMTS the Node B was supported by an RNC (Radio Network Controller), in LTE there is just one
RAN element: the eNB (evolved Node B). All the functionality that belonged to the RNC has been brought
into the eNB itself, which means it has a multitude of tasks to perform.
Evolved Packet Core

– EPC: Evolved Packet Core, also referred as SAE by 3GPP

– eUTRAN: Radio Access Network, also referred as LTE
Source: TS23.401,TS23.402,TS36.300 http://www.3gpp.org/
 MME: Mobility Management Entity
Evolved Packet Core
– Pure signalling entity inside the EPC:
• Signalling coordination for EPS bearer setup/release
– Subscriber attach/detach
– Tracking area updates
– Roaming Control
– Trigger and distribution of paging messages to UE
– Radio security control
S1-MME MME
– Authentication, HSS
S6a
integrity protection eNB
S11
S1-U
Serving
Gateway
Serving Gateway
– Manages the user data in the EPC
– Receives packet data from the eNodeB and sends packet data to it
 Evolved Packet Core
The EPC comprises four key elements:
 SGW (Signalling Gateway)
 PDN-GW (Packet Data Network Gateway)
 MME (Mobility Management Entity)
 HSS (Home Subscriber Server)

Other network elements may be required for specific service provision, and may vary depending on operator
implementation, but these elements are integral to any EPC configuration.

The desirability of separating signalling from data has been carried through into LTE, such that the user plane and control
plane are separated, enabling operators to adapt and dimension their networks more easily. This split is reflected in the
network architecture. The SGW and PDN-GW, which are logically connected together, are responsible for user plane
functionality, transporting data between the UEs and external networks. The SGW is the entry and exit point between the
EPC and the RAN; the PDN-GW is the entry and exit point between the EPC and other PDNs (Packet Data Networks) and is
responsible for IP address allocation to the UE.

The MME performs control-plane functions such as signalling and user mobility and is the termination point for the NAS
(Non-Access Stratum).
 Packet Data Network Gateway
Evolved Packet Core
– Connection between EPC and a number of external data networks (comparable to GGSN in
2G/3G networks)
– IP Address Allocation for UE MME
S7 Rx+
S6a PCRF
– Packet Routing/Forwarding between HSS
S11
Serving GW and external Data Network S5/S8 SGi
PDN
– Packet screening (firewall functionality) Serving PDN
Gateway Gateway
• Policy and Charging Rule Function
– Quality of Service (QoS) negotiation with the external PDN
– Charging Policy: How packets should be accounted

• HSS: Home Subscriber Server

– Permanent and central subscriber database
– Stores mobility and service data for every subscriber Contains AuC (authentication center)
functionality
 Evolved Packet Core
The HSS is modelled on the HLR and AuC in 2G/3G networks, and is effectively a database that contains user-related and
subscriber-related information. It also has some mobility management functions.

This network composition, and the functional split between the user and control planes, is illustrated in the simplified
figure below. These elements will all be considered in greater detail later in the module.
Evolved Packet Core
LTE Radio Interface, protocols and the X2 Interface

LTE-Uu interface
• Air interface of LTE
• Based on OFDMA in DL and SC-FDMA in UL
• FDD and TDD duplex methods
• Scalable bandwidth 1.4MHz to currently 20 MHz

X2 interface
• Inter eNB interface
• X2AP: special signalling protocol
• Functionalities:
– In inter- eNB HO to facilitate handover and
provide data forwarding.
– In RRM to provide e.g. load information to
neighbouring eNBs to facilitate interference
management
– Logical interface: It does not need direct site-
to-site connection
 Evolved Packet Core

Handover over X2 - Principle

X2 Handover - Capacity

30…50ms radio link interruption during HO

X2 latency should be less or equal than radio link interruption provisioning of extra capacity may only be justified for sites
time for optimum performance where high HO performance is required
 Evolved Packet Core
S1-MME & S1-U Interfaces and Protocols
S1 interface is divided into two parts:

S1-MME interface
• Control interface between eNB and MME
• S1AP:S1 Application Protocol
• MME and UE will exchange non-access
stratum signaling via eNB through this
interface (i.e. authentication, tracking area
updates)

S1-U interface
• User plane interface between eNB and serving
gateway
• Pure user data interface (U=User plane)
 IMS (IP Multimedia Subsystem)
The IMS is an all-IP architectural framework for delivering a range of multimedia services over IP networks. These include
services, such as speech, that traditionally transit CS networks.

The first iteration of IMS appeared in Release 5 of the 3GPP specifications, when it was represented as a possible means of
evolving core networks from circuit switching to packet switching, enabling the development of Internet-style packet-based
services to subscribers over UMTS networks. The IMS specifications were further developed in Releases 6 and 7. At Release 8,
3GPP integrated IMS into the LTE/SAE specifications, as a solution for offering voice and content-rich multimedia services over
the IP network.

The signalling protocol chosen for controlling real-time multimedia sessions via IMS in LTE networks is SIP (Session Initiation
Protocol), defined by the IETF (Internet Engineering Task Force). Voice data is tunnelled through the IP core network inside IP
packets. The advantage of such an architecture is that all services are handled through a common control architecture, by
means of SIP sessions.

The implementation of an IMS is extremely dependent on sufficient IP QoS mechanisms. It is also possible for operators to
deploy an IMS in parallel with retained MSC functionality. Indeed, it seems highly likely that, given the considerable investment
operators have made in their existing networks, operators’ legacy networks will stay in place for some time to come.
 IMS (IP Multimedia Subsystem)
The figure below shows the differences in service delivery over IMS when a UE is in LTE coverage and when it falls back to
UTRAN or GERAN (GSM/EDGE Radio Access Network) coverage.
Redes 5G
Mg./Msc. Henry A. Vasquez
Performance requirements: High data rates and traffic densities
Area traffic Area traffic
Experienced Experienced capacity capacity Overall user
Scenario Activity factor UE speed Coverage
data rate (DL) data rate (UL) density
(DL) (UL)
Pedes trians
2 2
100 Gbps /km 50 Gbps /km 2 and us ers in Full network
1 Urban macro 50 Mbps 25 Mbps 10 000/km 20%
vehicles (up to (note 1)
(note 4) (note 4) 120 km /h
Pedes trians
2 2
1 Gbps /km 500 Mbps /km and us ers in Full network
2 Rural macro 50 Mbps 25 Mbps 100/km 2 20%
vehicles (up to (note 1)
(note 4) (note 4) 120 km /h
Of f ice and
Indoor 2 2 2
3 1 Gbps 500 Mbps 15 Tbps /km 2 Tbps /km 250 000/km note 2 Pedes trians residential (note
hotspot 2) (note 3)
Broadband
[3,75]
4 access in a 25 Mbps 50 Mbps 2 [7,5] Tbps /km 2 [500 000]/km 2 30% Pedes trians Confined area
Tbps /km
crowd
Pedes trians
750 Gbps /km 2 125 Gbps /km 2 2 and us ers in Dow ntow n (note
5 Dense urban 300 Mbps 50 Mbps 25 000/km 10%
vehicles (up to 1)
(note 4) (note 4) 60 km /h)
Stationary
[15] TV us ers ,
Maxim um 200 N/A or m odes t
Broadcast- channels of [20 pedes trians Full network
6 Mbps (per TV (e.g., 500 kbps N/A N/A N/A
like services Mbps ] on one and us ers in (note 1)
channel) per us er)
carrier vehicles (up to
500 km /h)
Us ers in trains Along railways
High-speed
7 50 Mbps 25 Mbps 15 Gbps /train 7,5 Gbps /train 1 000/train 30% (up to 500
train
km /h) (note 1)
Us ers in Along roads
High-speed 2 2 2
8 50 Mbps 25 Mbps [100] Gbps /km [50] Gbps /km 4 000/km 50% vehicles (up to
vehicle (note 1)
250 km /h)

Airplanes 600
Us ers in TS 22.261
9 15 Mbps 7,5 Mbps 1,2 Gbps /plane 400/plane 20% airplanes (up (note 1)
connectivity Mbps /plane
to 1 000 km /h)
June 2018
 TS 22.261
Performance requirements: Low latency and high reliability June 2018
Communication
Traffic Connection Service area
End-to-end latency service Reliability User Payload
density density dimension
Scenario Jitter Survival time availability experienced
(note 3) (note 4) (note 4) data rate size (note 6) (note 7) (note 8)
(note 5)
Discrete automation – motion control 1 Mbps 2 2
1 ms 1 µs 0 ms 999999% 999999% Small 1 Tbps/km 100 000/km 100 x 100 x 30 m
(note 1) up to 10 Mbps
Discrete automation 10 ms 100 µs 0 ms 99,99% 99,99% 10 Mbps Small to big 1 Tbps/km 2 100 000/km 2 1000 x 1000 x 30 m
1 Mbps 100
Process automation – remote control 50 ms 20 ms 100 ms 999999% 999999% Small to big 1 000/km 2 300 x 300 x 50 m
up to 100 Mbps Gbps/km 2
Process automation ‒ monitoring 50 ms 20 ms 100 ms 99,9% 99,9% 1 Mbps Small 10 Gbps/km
2
10 000/km
2
300 x 300 x 50
100 km along power
Electricity distribution – medium voltage 25 ms 25 ms 25 ms 99,9% 99,9% 10 Mbps Small to big 10 Gbps/km 2 1 000/km
2
line
Electricity distribution – high voltage 100 1 000/km
2
200 km along power
5 ms 1 ms 10 ms 999999% 999999% 10 Mbps Small
(note 2) Gbps/km 2 (note 9) line
Intelligent transport systems –
10 ms 20 ms 100 ms 999999% 999999% 10 Mbps Small to big 10 Gbps/km 2 1 000/km
2
2 km along a road
infrastructure backhaul
Tactile interaction
0,5 ms TBC TBC [99,999%] [99,999%] [Low] [Small] [Low] [Low] TBC
(note 1)
[From low to
Remote control [5 ms] TBC TBC [99,999%] [99,999%] [Small to big] [Low] [Low] TBC
10 Mbps]
Performance requirements: Higher-accuracy positioning TS 22.261
June 2018
Dimension of service area
Service Position acquisition time Survival time Availability Position accuracy
(note 1)
Mobile objects
500 ms 1s 99,99% 500 x 500 x 30 m 0,5 m
on factory floor
SA1 completes its study into 5G requirements

…potential 5G requirements - resulting in TR22.891

SA1 completes its study into 5G requirements
 SA1 completes its study into 5G requirements

Massive Internet of Things focuses on use cases with massive number of devices (e.g., sensors and wearables). This group of
TR 22.861: FS_SMARTER – massive Internet of Things use cases is particularly relevant to the new vertical services, such as smart home and city, smart utilities, e-Health, and smart
wearables.

The main areas where improvements are needed for Critical Communications are latency, reliability, and availability to enable,
TR 22.862: FS_SMARTER – Critical Communications for example, industrial control applications and tactile Internet. These requirements can be met with an improved radio
interface, optimized architecture, and dedicated core and radio resources.

Enhanced Mobile Broadband includes a number of different use case families related to higher data rates, higher density,
TR 22.863: FS_SMARTER – enhanced Mobile Broadbanddeployment and coverage, higher user mobility, devices with highly variable user data rates, fixed mobile convergence, and
small-cell deployments.

The use case group Network Operation addresses the functional system requirements, including aspects such as: flexible
TR 22.864: FS_SMARTER – Network Operation
functions and capabilities, new value creation, migration and interworking, optimizations and enhancements, and security.

SMARTER : Services and Markets Technology Enablers

Comparison between 4G and 5G
 SA1 completes its study into 5G requirements
5G Diverse Scenarios Require Flexible Network

- The key capabilities are greatly different for eMBB,

mMTC & URLLC, meet all the requirements at the same
time is a waste of network resource

- Transport network should be flexible to:

•Meet all the application requirements
•Achieve efficient resource utilization
•Improve cost effectiveness
The 10 insights highlight the collective expectations, plans and concerns of
the industry for the 5G era
Insight 5G as a technology will evolve over time and leverage a variety of spectrum
2
ranges, plus robust security, to support new use cases
The evolution to 5G, both in terms of technology and network launches, will occur via the following use cases:
• Stage 1: FWA networks
• Stage 2: extensive trials at sporting events pre-2020
• Stage 3: wide commercial network rollout from around 2020.
Mobile operators maintain that 5G requires a new business model in order to drive the level of investment required to deliver
some of the early identified use cases. A concerted effort is required to frame that agenda and identify the sustainable network
operator business model(s) of the future. Otherwise, given 5G’s evolutionary nature, the industry risks ingraining current business
models across the 5G era. This in turn is likely to constrain network investment, particularly when it comes to the deployment of
small cells to support mmW 5G services.
Evolutionary technology and new business model 49%
Perceptions of 5G Revolutionary technology and new business model 21%
Evolutionary technology and traditional business model 15%
Revolutionary technology and traditional business model 8%
Other combination 8%
Source: GSMA - 2017 Mostly a marketing activity 0%
Mobile Broadband Networks of the Future

PRESENT FUTURE

• Insufficient Spectrum • Balanced Portfolio of Licensed and

• Capacity Constrained Unlicensed Spectrum
• Competitive Alternative to Wireline for Some • Significantly Greater Capacity
• Competitive Alternative to Wireline for Many
Examples of FWA deployment alternatives
Outline

Release 16 will be "5G phase 2" and should be completed in December 2019 (TSG SA#87).
Timeline
Timeline