Harry Holma and Antti Toskala: 1.1 Mobile Voice Subscriber Growth
Harry Holma and Antti Toskala: 1.1 Mobile Voice Subscriber Growth
Harry Holma and Antti Toskala: 1.1 Mobile Voice Subscriber Growth
Introduction
AL
Harry Holma and Antti Toskala
RI
TE
1.1 Mobile Voice Subscriber Growth
The number of mobile subscribers increased tremendously from 2000 to 2010. The first
MA
billion landmark was passed in 2002, the second billion in 2005, the third billion 2007,
the fourth billion by the end of 2008 and the fifth billion in the middle of 2010. More
than a million new subscribers per day have been added globally – that is more than ten
subscribers on average every second. This growth is illustrated in Figure 1.1. Worldwide
D
mobile phone penetration is 75%1 . Voice communication has become mobile in a massive
TE
way and the mobile is the preferred method of voice communication, with mobile networks
covering over 90% of the world’s population. This growth has been fueled by low-cost
mobile phones and efficient network coverage and capacity, which is enabled by standard-
GH
ized solutions, and by an open ecosystem leading to economies of scale. Mobile voice is
not the privilege of the rich; it has become affordable for users with a very low income.
RI
Second-generation mobile networks – like the Global System for Mobile Communications
(GSM) – were originally designed to carry voice traffic; data capability was added later.
Data use has increased but the traffic volume in second-generation networks is clearly
CO
dominated by voice traffic. The introduction of third-generation networks with High Speed
Downlink Packet Access (HSDPA) boosted data use considerably.
Data traffic volume has in many cases already exceeded voice traffic volume when
voice traffic is converted into terabytes by assuming a voice data rate of 12 kbps. As an
example, a European country with three operators (Finland) is illustrated in Figure 1.2.
The HSDPA service was launched during 2007; data volume exceeded voice volume
during 2008 and the data volume was already ten times that of voice by 2009. More
than 90% of the bits in the radio network are caused by HSDPA connections and less
than 10% by voice calls. High Speed Downlink Packet Access data growth is driven by
1The actual user penetration can be different since some users have multiple subscriptions and some subscriptions
are shared by multiple users.
LTE for UMTS: Evolution to LTE-Advanced, Second Edition. Edited by Harri Holma and Antti Toskala.
© 2011 John Wiley & Sons, Ltd. Published 2011 by John Wiley & Sons, Ltd.
2 LTE for UMTS: Evolution to LTE-Advanced
8000 100 %
World population
7000 90 %
Mobile subscribers
Penetration 80 %
6000
70 %
5000
Penetration
60 %
Million
4000 50 %
3000 40 %
30 %
2000
20 %
1000 10 %
0 0%
98
99
00
01
02
03
04
05
06
07
08
09
10
19
19
20
20
20
20
20
20
20
20
20
20
20
Figure 1.1 Growth of mobile subscribers
Figure 1.2 HSDPA data volume exceeds voice volume (voice traffic 2007 is scaled to one)
high-speed radio capability, flat-rate pricing schemes and simple device installation. In
short, the introduction of HSDPA has turned mobile networks from voice-dominated to
packet-data-dominated networks.
Data use is driven by a number of bandwidth-hungry laptop applications, including
internet and intranet access, file sharing, streaming services to distribute video content and
mobile TV, and interactive gaming. Service bundles of video, data and voice – known also
as triple play – are also entering the mobile market, causing traditional fixed-line voice and
broadband data services to be replaced by mobile services, both at home and in the office.
A typical voice subscriber uses 300 minutes per month, which is equal to approximately
30 megabytes of data with the voice data rate of 12.2 kbps. A broadband data user can
easily consume more than 1000 megabytes (1 gigabyte) of data. The heavy broadband
data use takes between ten and 100 times more capacity than voice usage, which sets
high requirements for the capacity and efficiency of data networks.
Introduction 3
It is expected that by 2015, five billion people will be connected to the internet. Broad-
band internet connections will be available practically anywhere in the world. Already,
existing wireline installations can reach approximately one billion households and mobile
networks connect more than three billion subscribers. These installations need to evolve
into broadband internet access. Further extensive use of wireless access, as well as new
wireline installations with enhanced capabilities, is required to offer true broadband con-
nectivity to the five billion customers.
1.000
Optics
Wireline VDSL 2 100 Mbps
100 25–50 Mbps LTE
User data rate (Mbps)
ADSL2+
ADSL 16–20 Mbps
10 HSPA+
6–8 Mbps
ADSL
1–3 Mbps HSDPA
1 3.6–7.2 Mbps
HSDPA
1.8 Mbps
WCDMA Wireless
0.1
EDGE 0.384 Mbps
0.236 Mbps
0.01
2000 2005 2010
Year of availability
Figure 1.3 Evolution of wireless and wireline user data rates GPON = Gigabit Passive Opti-
cal Network. VDSL = Very High Data Rate Subscriber Line. ADSL = Asymmetric Digital
Subscriber Line
4 LTE for UMTS: Evolution to LTE-Advanced
• spectral efficiency two to four times more than with HSPA Release 6;
• enables a round
peak rates exceed 100 Mbps in the downlink and 50 Mbps in the uplink;
• packet switched trip time of <10 ms;
• high level of mobility
optimized;
• optimized terminal powerand security;
• frequency flexibility with allocations
efficiency;
• from below 1.5 MHz up to 20 MHz.
LTE targets
Fa
cto
2-4
r
10
of
of
of
r
cto
-3
r
cto
Fa
Fa
Up to 20 MHz
Uplink … SC-FDMA
Downlink … OFDMA
Frequency
GGSN S-GW
Core network functionality split
SGSN MME • MME for control plane
• User plane by-pass MME
RNC
eNodeB functionalities
• All radio protocols
NodeB eNodeB • Mobility management
• All retransmissions
• Header compression
= Control plane
= User plane
are located in Radio Network Controller (RNC) in 3GPP Release 6 architecture. The core
network is streamlined by separating the user and the control planes. The Mobility Man-
agement Entity (MME) is just a control plane element and the user plane bypasses MME
directly to Serving Gateway (S-GW). The architecture evolution is illustrated in Figure 1.7.
02
03
04
05
06
07
08
09
10
20
20
20
20
20
20
20
20
20
20
3GPP schedule
LTE
WCDMA HSDPA HSUPA HSPA+ LTE-A
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
will soon be turning to, GSM/WCDMA for voice evolution and to HSPA/LTE for data
evolution to access the benefits of the large and open 3GPP ecosystem and for economies
of scale for low-cost mobile devices. The number of subscribers using 3GPP-based tech-
nologies is currently more than 4.5 billion. The 3GPP Long Term Evolution (LTE) will
be built on this large base of 3GPP technologies.
The time schedules of 3GPP specifications and the commercial deployments are illus-
trated in Figure 1.9. The 3GPP dates refer to the approval of the specifications. WCDMA
Release 99 specification work was completed at the end of 1999 and was followed by
the first commercial deployments during 2002. The HSDPA and HSUPA standards were
completed in March 2002 and December 2004 and the commercial deployments followed
in 2005 and 2007. The first phase of HSPA evolution, also known as HSPA+, was com-
pleted in June 2007 and the deployments started during 2009. The LTE standard was
approved at the end of 2007, backwards compatibility started in March 2009 and the first
commercial networks started during 2010. The next step is LTE-Advanced (LTE-A) and
the specification was approved in December 2010.
The new generations of technologies push the data rates higher. The evolution of the
peak user data rates is illustrated in Figure 1.10. The first WCDMA deployments 2002
offered 384 kbps, first HSDPA networks 3.6–14 Mbps, HSPA evolution 21–168 Mbps,
LTE 150–300 Mbps and LTE-Advanced 1 Gbps, which is a more than 2000 times higher
data rate over a period of ten years.
LTE-Advanced
LTE
1 Gbps
HSPA+
150 –300 Mbps
HSPA
21–168 Mbps
WCDMA
14 Mbps
EDGE
2 Mbps
a rate
472 kbps
l peak dat
retica
Theo
The 3GPP technologies are designed for smooth interworking and coexistence. The
LTE will support bi-directional handovers between LTE and GSM and between LTE
and UMTS. GSM, UMTS and LTE can share a number of network elements including
core network elements. It is also expected that some of the 3G network elements can
be upgraded to support LTE and there will be single network platforms supporting both
HSPA and LTE. The subscriber management and SIM (Subscriber Identity Module)-based
authentication will be used also in LTE.
networks will initially be built on 700 and 1700/2100 MHz frequencies. In Japan the LTE
deployments start using the 2100 band followed later by 800, 1500 and 1700 bands.
Flexible bandwidth is desirable to take advantage of the diverse spectrum assets:
refarming typically requires a narrowband option below 5 MHz while the new spectrum
allocations could take advantage of a wideband option of data rates of 20 MHz and higher.
It is also evident that both FDD and TDD modes are required to take full advantage of
the available paired and unpaired spectrum. These requirements are taken into account in
the LTE system specification.
2300–2400
2100 2200 2300 2400 2500 2600 2700 2800 2900 3000
Capacity bands
3400–3800
3100 3200 3300 3400 3500 3600 3700 3800 3900 4000
used for terrestrial TV broadcasting. The 790–862 MHz sub-band was identified in Europe
and Asia-Pacific. The availability of the band depends on the national time schedules of
the analogue to digital TV switchover. The first auction for that band was conducted in
Germany in May 2010 and the corresponding frequency variant is Band 20. The band
allows three operators, each running 10 MHz LTE FDD.
The 698–806 MHz sub-band was identified for IMT in Americas. In the US part of the
band has already been auctioned. In Asia, the band plan for 698–806 MHz is expected
to cover 2 × 45 MHz FDD operation.
The main capacity band will be 3.4–4.2 GHz (C-band). A total of 200 MHz was iden-
tified in the 3.4–3.8 GHz sub-band for IMT in Europe and in Asia-Pacific. This spectrum
can facilitate the deployment of larger bandwidth of IMT-Advanced to provide the highest
bit rates and capacity.
The 2.3–2.4 GHz band was also identified for IMT but this band is not expected to
be available in Europe or in the Americas. This band was identified for IMT-2000 in
China at the WRC-2000. The 450–470 MHz sub-band was identified for IMT globally,
but it is not expected to be widely available in Europe. This spectrum will be narrow
with maximum 2 × 5 MHz deployment. Further spectrums for IMT systems are expected
to be allocated in the WRC-2016 meeting.
1.9 LTE-Advanced
International Mobile Telecommunications – Advanced (IMT-Advanced) is a concept
for mobile systems with capabilities beyond IMT-2000. IMT-Advanced was previously
known as ‘Systems beyond IMT-2000’. The candidate proposals for IMT-Advanced
were submitted to ITU in 2009. Only two candidates were submitted: LTE-Advanced
from 3GPP and IEEE 802.16m.
It is envisaged that the new capabilities of these IMT-Advanced systems will support
a wide range of data rates in multi-user environments with target peak data rates of up to
approximately 100 Mbps for high mobility requirements and up to 1 Gbps for low mobility
requirements such as nomadic/local wireless access. IMT-Advanced work within 3GPP
is called LTE-Advanced (LTE-A) and it is part of Release 10. 3GPP submitted an LTE-
Advanced proposal to ITU in October 2009 and more detailed work was done during 2010.
The content was frozen in December 2010 and the backwards compatibility is expected
Mobility
High
IMT
-2000 IMT
-2000 IMT-
evolution Advanced
LTE-
Low WCDMA HSPA LTE Advanced
1 10 100 1000 Peak data
rate (Mbps)
40 –100 MHz
More bandwidth
8x MIMO 4x
More antennas
Relays
Heterogeneous
networks
in June 2011. The high-level evolution of 3GPP technologies to meet IMT requirements
is shown in Figure 1.14.
The main technology components in Release 10 LTE-Advanced include:
• carrier aggregation up to 40 MHz total band, and later potentially up to 100 MHz;
evolution up to 8 × 8 in downlink and 4 × 4 in uplink;
• relay nodes
MIMO
• heterogeneousfor networks
providing simple transmission solution;
• macro, micro, pico and femto
for optimized interworking between cell layers including
cells.