Report The economics of small cells and Wi-Fi offload
The economics of small cells and Wi-Fi offload
By Monica Paolini
SENZA
CONSULTING
Report The economics of small cells and Wi-Fi offload
© 2012 Senza Fili Consulting • www.senzafiliconsulting.com • Reproduction and redistribution prohibited |2|
Key results
Expected increase in data traffic requires a massive increase in capacity in mobile
networks.
Small cells and Wi-Fi are crucial to providing the increase in capacity density that
operators need and subscribers expect.
Operators need both small cells and Wi-Fi to cost-effectively meet their capacity
requirements.
The small-cell and Wi-Fi infrastructure can be colocated to increase efficiency and
reduce per-bit costs.
To maximize the benefits of small-cell and Wi-Fi colocations, integration of Wi-Fi into
the mobile network core is required.
The TCO analysis shows that adding more interfaces and sectors to small cells leads
to only modest marginal increases in the TCO.
Even at low densities, LTE small cells and Wi-Fi quickly take on a dominant role
relative to macro cells in transporting mobile traffic.
Small cells and Wi-Fi enable operators to slash per-bit TCO by at least half.
Per-bit TCO shows that Wi-Fi added to small cells greatly improves the small-cell
business case, especially for 3G small cells.
Source: Senza Fili Source: Senza Fili
Source: Senza Fili
Source: Senza Fili
Report The economics of small cells and Wi-Fi offload
© 2012 Senza Fili Consulting • www.senzafiliconsulting.com • Reproduction and redistribution prohibited |3|
Table of contents
1. Introduction:
Why small cells and Wi-Fi? Two complementary tools, both needed to address traffic
growth 5
2. Capacity and capacity density
Where will the growth in capacity come from? 7
3. Small cells versus Wi-Fi
A comparison of two complementary approaches to capacity increase 10
4. Small cells and Wi-Fi
Why it makes sense to deploy them alongside each other 11
5. Past and future of Wi-Fi
Integration into mobile networks is key to maximizing Wi-Fi’s contribution 13
6. Small cells, Wi-Fi access and Wi-Fi offload
Defining terms and scope used in the report 15
7. Cost assumptions
Building the TCO model 16
8. Comparing the costs for macro cells, small cells and Wi-Fi
The base TCO model 19
9. The capacity contribution of small cells and Wi-Fi
Incremental capacity increase with more small cells per macro cell 23
10. Per-bit TCO
Assessing the cost-effectiveness of small cells and Wi-Fi 25
11. Findings: two (or three) is better than one
Synergies among LTE, 3G and Wi-Fi strengthen small-cell business case 27
12. Acronyms 28
Report The economics of small cells and Wi-Fi offload
© 2012 Senza Fili Consulting • www.senzafiliconsulting.com • Reproduction and redistribution prohibited |4|
List of figures
Figure 1. Capacity density 5
Figure 2. Combining small cells and Wi-Fi 6
Figure 3. Mobile IP traffic 7
Figure 4. Europe data traffic and revenue growth 7
Figure 5. Fixed, mobile and Wi-Fi IP traffic 8
Figure 6. Capacity increase, past and future 9
Figure 7. Integrating Wi-Fi in the mobile network: looser and tighter coupling 14
Figure 8. Outdoor and indoor TCO 19
Figure 9. TCO for outdoor and indoor small cells, as a percentage of macro-cell TCO 20
Figure 10. Capex and equipment as a percentage of TCO 21
Figure 11. Backhaul capex and opex as a percentage of TCO for macro cells, indoor and outdoor
small cells, and Wi-Fi 22
Figure 12. Capacity contribution of small cells and Wi-Fi as the density of small cells increases
24
Figure 13. Per-mbps TCO for LTE and 3G small cells and macro cells, and Wi-Fi 25
Figure 14. Per-bit TCO for LTE and 3G small cells and Wi-Fi as a percentage of per-bit TCO for
macro cells 26
List of tables
Table 1. Sources of capacity increase in the RAN 9
Table 2. Comparison of Wi-Fi offload and small cells 10
Table 3. Small cells and Wi-Fi: Together or separate? 11
Table 4. What is a small cell? 15
Table 5. Cost, capacity and backhaul assumptions for outdoor sites 17
Table 6. Cost, capacity and backhaul assumptions for indoor sites 18
Report The economics of small cells and Wi-Fi offload
© 2012 Senza Fili Consulting • www.senzafiliconsulting.com • Reproduction and redistribution prohibited |5|
1. Introduction: Why small cells and Wi-Fi? Two complementary tools, both needed to address traffic growth
Under pressure from more users, more devices and more applications, mobile networks have
to transport higher traffic loads, which are straining the current infrastructure and frustrating
subscribers when they cannot use the services they have paid for (Figure 1). The increase in
traffic load will continue over the coming years and is profoundly changing how mobile
operators plan, deploy and operate mobile networks, and charge for access.
Until recently, mobile networks mostly expanded horizontally to reach new areas or
obstructed locations or to improve indoor coverage. The new wave of expansion is orthogonal
and aimed at providing depth of coverage – that is, more capacity in already covered areas.
The initial horizontal expansion of mobile networks was achieved mostly through the
deployment of additional macro cells. This is no longer sufficient to get depth of coverage.
This report is based on the premise that both small cells and Wi-Fi are necessary to provide
the additional capacity needed, where it is needed.
Small cells and Wi-Fi add capacity to mobile networks in ways that are complementary, and in
sufficiently distinct ways that it is not possible to substitute one for the other across the
mobile footprint. Subscribers will not give up Wi-Fi access in the foreseeable future, and only
cellular networks can provide the coverage and mobility support that subscribers have come
to expect.
But the report will go a few steps further and explore the implications of the coexistence of
small cells and Wi-Fi. Where should operators deploy small cells instead of Wi-Fi? Where is
Wi-Fi more cost effective? Should small cells and Wi-Fi be deployed in largely segregated
networks, or should they be integrated – both in the RAN and in the core network? We argue
in favor of a deep integration of the two.
In many locations where either Wi-Fi or small cells are installed or planned, it will often make
sense to add the other wireless interface as well. This is especially true if mobile operators
integrate Wi-Fi within their core network and can manage cellular and Wi-Fi traffic effectively
Higher traffic load
•Growing smartphone penetration
•Multi-device subscribers
•More applications
•Higher per-subscriber traffic
Higher capacity needed
• More base stations
• More spectrum
• New technologies
• Multiple interfaces
•Effective traffic management
Higher capacity density
• Traffic load unevenly distributed across footprint and time of day
• Two directions:
• Dense metro areas
•Indoor coverage (home, office)
Figure 1. Capacity density. Source: Senza Fili
Report The economics of small cells and Wi-Fi offload
© 2012 Senza Fili Consulting • www.senzafiliconsulting.com • Reproduction and redistribution prohibited |6|
using the same tools to support the same services on both interfaces. We envision a
deployment model in which locations with the higher traffic levels (e.g., dense metro areas,
stadiums, airports) benefit from small cells with Wi-Fi, but in which small cells with a lower
traffic load (e.g., in suburban areas) and a wider range may not warrant the addition of Wi-Fi.
Similarly, in residential and most enterprise locations, Wi-Fi meets the capacity requirements,
thus making the addition of small cells unnecessary (Figure 2).
The report explores how this model addresses some of the deployment and business model
challenges of small cells – and, to a lesser extent, Wi-Fi – mostly from an economic
perspective, by looking at the TCO for different small-cell and Wi-Fi configurations along
multiple dimensions:
Location: outdoor versus indoor
Equipment: single-sector versus multi-sector small cells
Interface: LTE versus 3G small cells
Capacity: number of small cells per macro cell.
Figure 2. Combining small cells and Wi-Fi. Source: Senza Fili
Report The economics of small cells and Wi-Fi offload
© 2012 Senza Fili Consulting • www.senzafiliconsulting.com • Reproduction and redistribution prohibited |7|
2. Capacity and capacity density Where will the growth in capacity come from?
According to the widely cited Cisco VNI, mobile traffic will increase by a multiple of 18
between 2011 and 2015. It will account for 10% of global IP traffic by 2016 (Figure 3) and
represent five times the volume that global internet traffic had in 2005. One fact less
frequently cited is that this figure assumes the growth rate is decreasing, a trend that
Vodafone has already started to observe (Figure 4). This decrease in mobile traffic growth rate
is to be expected – the same happened to fixed internet traffic – and likely has multiple
causes. One is that new recruits to the mobile internet do not use it as heavily as the early
adopters. Another is that traffic caps and better understanding of what drives traffic volume
among subscribers may discourage some traffic. Finally, easier Wi-Fi connectivity and devices,
such as tablets, that connect predominantly through Wi-Fi are moving a higher percentage of
what used to be cellular traffic over to Wi-Fi. But even when we take all these factors into
account, the strong growth in cellular traffic will continue in the coming years.
However, to put the role of mobile traffic in perspective, it is helpful to view its contribution to
overall global IP traffic and to Wi-Fi traffic. In 2011, mobile data traffic accounted for 2% of
global IP traffic, and this percentage is expected to grow to 10% by 2016, again according to
Cisco VNI. In comparison, Wi-Fi plays a much larger role, although most traffic is generated in
a fixed environment. Cisco expects that Wi-Fi traffic will surpass wired traffic by 2015 and
account for 51% of global IP traffic by 2016 (Figure 5). While mobile data CAGR at 92% over
the 2011–2016 period outstrips the Wi-Fi CAGR by 39%, mobile traffic will be only 17% of the
volume of Wi-Fi traffic by 2016.
When we look at the overall mobile device market and usage models, these data indicate that
Wi-Fi is and will continue to be a major driver and enabler of mobile connectivity, not an
interim, second-best access interface in relationship to cellular. This carries important
implications for mobile operators. It is not to their advantage to consider Wi-Fi solely as an
offload technology onto which they can conveniently funnel all the traffic that would overload
their networks and forget about it. Rather, they will be better served by taking Wi-Fi into the
Figure 3. Mobile IP traffic. Source: Cisco VNI
Figure 4. Europe data traffic and revenue growth. Source: Vodafone
Report The economics of small cells and Wi-Fi offload
© 2012 Senza Fili Consulting • www.senzafiliconsulting.com • Reproduction and redistribution prohibited |8|
fold and integrating it with their mobile networks – and, as we will argue in this report, in
particular with small-cell deployments.
Even if the Cisco VNI forecast turns out to be overly generous, new spectrum allocations and
technological innovation will not be sufficient to address the anticipated steep traffic increase.
Take spectrum, for instance. While new allocations will make capacity expansion possible by
providing new channels and enabling the adoption of more spectrally efficient technologies
like LTE, the new spectrum on its own may at best allow two or three times the current
capacity across the network footprint. A similar case can be made with technological
improvements in spectral efficiency brought by LTE and LTE Advanced – especially in light of
the fact that, for many years, most networks will still use 2G and 3G technologies.
Not only does this fall short of providing for the capacity needs implied in the Cisco VNI
forecast – and in even more conservative ones – but it does not take into account the fact that
the growth in traffic load is not uniformly distributed in space or time. If it were, new
spectrum and new technology would be more valuable at increasing capacity, because they
can be used across the entire network footprint. Instead, traffic load increase is concentrated
mostly in the areas where traffic loads are already high and where macro-cell density is, as a
result, equally high. To make things worse, traffic demand is not equally distributed
throughout the day, but has different time-of-day traffic curves that depend on the type of
location. In metro areas, peaks tend to be during work hours. In residential areas, traffic
surges at night.
Because required capacity is based on peak-hour loads and not on average traffic, the need
for additional capacity is most acute in a limited subset of locations and a few hours of the
day. This makes it much more challenging to meet the new capacity requirements, because
the capacity injection has to be carefully targeted.
The main way to achieve this is to increase cell density. This should not come as a surprise.
Most of the increase in RAN capacity over recent decades has been accomplished by adding
cell sites, or by adding base stations to existing ones (Figure 6). In the future, vendors and
operators agree, cell density will continue to be a crucial driver of increase in capacity density.
The tools to get more capacity, however, will be different. Many mobile operators have
exhausted, or nearly exhausted, their ability to increase capacity in high-density areas by
Figure 5. Fixed, mobile and Wi-Fi IP traffic. Source: Cisco VNI
Report The economics of small cells and Wi-Fi offload
© 2012 Senza Fili Consulting • www.senzafiliconsulting.com • Reproduction and redistribution prohibited |9|
simply adding macro cells, without increasing interference and costs – and hence without
seeing much lower marginal returns on their investments.
This is where small cells and Wi-Fi come in, bringing the RAN infrastructure closer to the user
– hence affording a better link budget, which translates into a more efficient use of spectrum
and network resources to provide a highly targeted capacity increase where needed and at a
much lower per-bit cost (Table 1).
Table 1. Sources of capacity increase in the RAN
Benefits Challenges
Macro cell Well-understood deployment model
Mature technology
Good support for mobility
Fewer sites to manage
High lease costs
Higher per-bit costs for mobile broadband
Greater interference, caused by higher density
Small cell High increase in capacity
density
Better spectrum and resource utilization
Lower cost per bit
Better link budget because cell is closer to subscribers
Impact of interference still not fully understood
Higher number of cell sites to identify, install and manage
Cells located on non-telecom assets such as lampposts, which are difficult to protect
Expensive or limited availability backhaul options
Femto cell
(residential
and
enterprise)
Efficient way to add coverage or capacity in areas where they are limited
Cost-effective equipment
Do not require deployment or direct management by operator
Interference triggered by high density or improper placement by subscribers (i.e., not under control of the operator)
Higher cost and less availability than Wi-Fi
Wi-Fi
Mature technology
Inexpensive access point equipment
Ubiquitous in mobile devices
Familiar to subscribers, who use it extensively (and want to continue to do so)
Interference due to high utilization of the technology
No support for mobility (i.e., handoffs)
Limited mobile voice support
Not yet integrated with the mobile network (or limited integration)
Most of Wi-Fi traffic is outside the control of the operator
Figure 6. Capacity increase, past and future. Source: Alcatel-Lucent, Arraycomm,
KDDI, Senza Fili.
Report The economics of small cells and Wi-Fi offload
© 2012 Senza Fili Consulting • www.senzafiliconsulting.com • Reproduction and redistribution prohibited |10|
3. Small cells versus Wi-Fi A comparison of two complementary approaches to capacity increase
Both small cells and Wi-Fi bring an important contribution to capacity increase, but they do so
in different ways, which are largely complementary (Table 2). Some of these differences stem
from the fact that Wi-Fi is already a widely adopted, mature technology, while small cells are
still in the planning stage, despite the strong commitment from operators.
Another difference that will continue to play an important role is the use of license-exempt
spectrum in Wi-Fi networks and of licensed spectrum in small cells. This imposes some
limitations on the effectiveness of traffic and interference management in Wi-Fi. At the same
time, small cells have to address interference from the macro layer (if the small cells are
deployed in the same channels, as is expected in most cases). This is a source of interference
that the operator controls but cannot eliminate.
In both cases, the main challenges are due to the integration with the mobile network, as well
as to the business case and the operational aspects of deploying and managing a high number
of cell sites (e.g., site acquisitions, lease agreements, and backhaul in outdoor locations).
Another important difference is that Wi-Fi is used mostly at indoor locations (both for home
and residential networks, and for public hotspots), while small cells are initially targeted
mostly at outdoor locations. This difference is largely handed down by prevailing practice,
rather than performance or cost benefits. Wi-Fi has been predominantly deployed indoors
because this is where most usage (both for Wi-Fi and mobile) comes from. Small cells are
often initially positioned to fill the gap between macro wide-area coverage and Wi-Fi indoor
coverage. As more small cells are installed and indoor usage grows, small cells will start to also
occupy indoor locations.
As a result, we expect that this difference will wane over time and that the availability of
equipment to support both cellular and Wi-Fi in the same enclosure will accelerate the
convergence of small cells and Wi-Fi in the underlay public networks.
Table 2. Comparison of Wi-Fi offload and small cells
Wi-Fi Small cells
Already deployed, but being expanded A few deployments launched, major
deployments in two to three years
License-exempt spectrum Licensed spectrum, but often shared
with the macro layer
Interference from other Wi-Fi networks Interference with macro layer
Mostly indoors Mostly outdoors
Residential offload is major benefit to
mobile operators Focus on high-traffic urban areas
Operators can benefit from third-party
networks even without building own
hotspot network
Infrastructure sharing or neutral-host
wholesale arrangements may be used to
contain costs and widen footprint
Best-efforts QoS tiered services can be implemented
Next steps:
Transparent access to subscribers
Introduction of Hotspot 2.0 / Passpoint with SIM-based authentication to improve user experience
Integration of Wi-Fi traffic management within the cellular network
Next steps:
Network coordination and management of interference with macro network
Selection of backhaul that is affordable and meets performance requirements
Site acquisition and lease agreements
Report The economics of small cells and Wi-Fi offload
© 2012 Senza Fili Consulting • www.senzafiliconsulting.com • Reproduction and redistribution prohibited |11|
4. Small cells and Wi-Fi Why it makes sense to deploy them alongside each other
Despite the somewhat divergent paths that Wi-Fi and small cells have taken, there are many
reasons to add Wi-Fi within the small-cell enclosure (Table 3), either by adding a Wi-Fi module
or by swapping out an already installed Wi-Fi access point in favor of a small cell with
embedded Wi-Fi.
Fundamentally, this is all made possible by the fact that equipment costs play a small role in
the overall TCO for small cells and Wi-Fi hotspots (and, in fact, for macro cells as well), and
that adding Wi-Fi to small cells does not substantially increase the equipment cost. Once a
network is planned, the marginal capex and opex associated with the addition of Wi-Fi is
minimal, but the increase in capacity is substantial.
This is especially true for 3G small cells, which gain most of their capacity from the Wi-Fi
component. As a result, the 3G small-cell business case is strengthened by the addition of
Wi-Fi, because Wi-Fi lowers the per-bit prices of the combined 3G and Wi-Fi small cell.
In the case of LTE cells, Wi-Fi has a positive impact on the per-bit costs and does not
substantially affect the opex and capex. It is a “Why not?” consideration more than an
element that is necessary to close a business case.
A good case for adding Wi-Fi to small cells is the fact that by joining the cellular and Wi-Fi
infrastructure, mobile operators will be able to deploy and operate fewer sites. However,
because Wi-Fi and small cells are not necessarily deployed in the same locations, the gains will
be determined by the distribution of small cells with Wi-Fi and the traffic distribution across
the covered area.
For instance, if an operator plans to deploy 50 small cells and 50 Wi-Fi access points within the
same area, it does not mean that adding Wi-Fi to each small cell will result in the operator
achieving the same capacity target with 50 cell sites. It may need more than 50 cells if the
initial plan called for Wi-Fi and small cells to cover different zones (e.g., Wi-Fi for indoor
Table 3. Small cells and Wi-Fi: Together or separate?
Reasons to combine
LTE small cells: Increase capacity per small cell at a low marginal cost
3G small cells: Improve an otherwise challenging business case
Operate a smaller number of cell / access point sites
Accelerate deployment of small cells by leveraging already-acquired Wi-Fi locations
Address limitations in number of sites for small cells and Wi-Fi access points
Facilitate indoor deployments of small cells for operators who already have a Wi-Fi network in indoor locations
Expand outdoor Wi-Fi coverage for operators deploying outdoor small cells
Reasons to keep separate
Can meet traffic demand with just one interface (LTE, 3G or Wi-Fi)
Leverage own spectrum, if operator has sufficient spectrum to get capacity needed from small cells, without relying on its own Wi-Fi network
Optimally place small cells and Wi-Fi where each performs best or where acquiring new sites is neither expensive nor difficult
Avoid Wi-Fi because high levels of Wi-Fi availability make an operator-run Wi-Fi network redundant
Keep indoor deployments limited to Wi-Fi, and deploy small cells outdoors
Avoid delaying Wi-Fi deployments, if operator is not yet ready for small cells
Adopt a Wi-Fi–first strategy that requires small-cell deployments only when and where Wi-Fi is no longer able to cost-effectively provide additional capacity
Report The economics of small cells and Wi-Fi offload
© 2012 Senza Fili Consulting • www.senzafiliconsulting.com • Reproduction and redistribution prohibited |12|
coverage and LTE for outdoor areas).
Conversely, it is also possible that the operator may need fewer than 50 sites if the reduced
site count enables it to pick more effective but also more expensive locations that have a
higher utilization. In this case, the 50 cells have the same capacity as the 100 cells, but they
are used more intensively and therefore carry more traffic. In general, though, operators
should see a decline in the number of sites while preserving the capacity, and this will have a
positive effect on their bottom line.
The swap of a Wi-Fi access point for a small cell with embedded Wi-Fi has a very different
value proposition. In this case, time-to-market considerations and the ability to gain a
foothold in indoor coverage are key advantages. When deploying small cells, the process of
identifying a site, getting a lease, bringing backhaul and power, and installing the equipment
can take a long time and require protracted efforts.
If swapping Wi-Fi access points with small cells, operators already have many of these
elements accounted for. The site, backhaul and power are already in place. In most cases, the
operator simply needs to replace the access point with a small cell. The downside of this
approach is that it may force the operator to pick existing Wi-Fi locations that serve
subscribers well yet fail, with the new small cell, to cover areas with a high density of
subscribers that the existing Wi-Fi infrastructure does not serve.
To be fair, there are also many reasons not to combine small cells and Wi-Fi (Table 3). In areas
where there is a need for Wi-Fi or small cells, there may not be sufficient demand to justify
the further capacity injection provided by either Wi-Fi or LTE. Even though the addition of
Wi-Fi has little impact on the TCO, there is no reason to add it if it would not provide any
benefit.
A mobile operator might also have good reasons not to deploy Wi-Fi if it does not need its
own Wi-Fi network (e.g., if the operator has sufficient spectrum to deploy all the small cells it
needs, or if the area targeted with small cells already has sufficient free Wi-Fi access and a
new Wi-Fi access point could suffer from high levels of interference). Alternatively, the
operator might want to use Wi-Fi for indoor coverage and small cells for outdoor coverage, or
use Wi-Fi to its full capabilities before starting to deploy small cells.
Report The economics of small cells and Wi-Fi offload
© 2012 Senza Fili Consulting • www.senzafiliconsulting.com • Reproduction and redistribution prohibited |13|
5. Past and future of Wi-Fi Integration into mobile networks is key to maximizing Wi-Fi’s contribution
Mobile operators have embraced Wi-Fi, initially, with considerable circumspection
and suspicion. From a network infrastructure, they ended up encouraging people to
use license-exempt spectrum, after paying princely sums for the spectrum
allocations and while lobbying for additional spectrum to be made available. How
could it be that freely available spectrum could be more efficient at providing
wireless broadband than operator-controlled spectrum? And if it is, why not expand
availability of license-exempt spectrum instead of licensed spectrum? Of course we
need both, but from the operators’ perspective, this can be a slippery slope.
Equally important, operators saw in Wi-Fi the potential for cannibalization or
reduction of perceived value of the service they provided to their subscribers. Why
should a subscriber pay high monthly fees for data connectivity, when Wi-Fi is widely
available for free – and it is even faster?
Even as they adopted Wi-Fi, most operators tried to keep their distance from it.
Subscribers could of course use Wi-Fi, but they were often not actively encouraged
to take full advantage of it, because operators saw a risk that Wi-Fi usage would
reduce their control over the devices and the subscriber. To some extent, this has
happened: usage behavior suggests that their subscribers perceive their devices to
have two distinct personalities – the paid-for and slower cellular one, and the fast-
and-free Wi-Fi one, usually with their heart going to the second one.
In this context, it is easy to understand the mobile operators’ initial attitude that
Wi-Fi was an interim fix until the arrival of LTE and small cells.
But it has become clear that Wi-Fi will continue to play a central role in mobile
networks, and many mobile operators acknowledge this. Some are moving further,
to go beyond don’t-ask, don’t-tell offload and start integrating Wi-Fi access into their
network. In addition to providing relief from the traffic crunch, Wi-Fi gives operators
The future of Wi-Fi
Wi-Fi is a mature technology that has greatly improved in performance and functionality through the years, with the introduction of IEEE 802.11n (throughput), WPA2 (security), WMM (QoS), Power Save (battery life), Miracast (video) and streamlined setup interfaces.
In 2013, IEEE 802.11ac will open Wi-Fi to additional 5 GHz spectrum bands and use wider channels. This will bring more capacity and throughput in the gbps range (i.e., doubling the current 802.11n throughput).
IEEE 802.11ad will extend Wi-Fi to the 60 GHz spectrum to provide very high throughput but within a shorter range than current Wi-Fi access points have. It will complement existing Wi-Fi and provide better performance for video and other high-throughput applications in environments with a high density of devices.
Passpoint, the Wi-Fi Alliance certification program based on the Hotspots 2.0 specification, enables seamless access, with SIM-based authentication, to Wi-Fi hotspots managed by the mobile operator and its partners. Because the device is authenticated with the same SIM-based credentials used for the cellular network, the operator can extend to the Wi-Fi network the security, policy and charging frameworks used for mobile services. Early commercial launches are expected for 2013.
To improve Wi-Fi roaming, the Wireless Broadband Alliance has launched the Next Generation Hotspot (NGH) initiative to establish roaming best practices for Wi-Fi and facilitate the creation of roaming partnerships among mobile operators.
3GPP efforts to provide a framework that integrates mobile and Wi-Fi networks have created ANDSF to enable mobile devices to discover and select non-3GPP networks on the basis of PCRF-defined rules. ANDSF enables real-time traffic management and policy enforcement, and it improves power-saving management on the mobile device.
Report The economics of small cells and Wi-Fi offload
© 2012 Senza Fili Consulting • www.senzafiliconsulting.com • Reproduction and redistribution prohibited |14|
a way to offer a wider range of personalized – and revenue-generating – services.
Most Wi-Fi offload models today keep all traffic (user plane and control plane) outside the
mobile network. As a result, the mobile operator loses all visibility into offloaded traffic, which
precludes the ability to enforce security, to use policy, to manage traffic and to use load
balancing for network selection.
Solutions that enable a deep integration of Wi-Fi into mobile networks are now available and
allow mobile operators to retain full control of Wi-Fi traffic in the control plane, while
offloading user-plane traffic to avoid congestion in the core network (Figure 7).
By integrating Wi-Fi within their core networks, operators can treat Wi-Fi traffic in the same
way they do cellular traffic and benefit from the same functionality for traffic management,
policy, charging and security.
This integration makes the combination of small cells and Wi-Fi even more powerful. Not only
does Wi-Fi add capacity to the small cells, it also enables the operator to allocate traffic to one
or the other interface based on real-time traffic load and available throughput (which in turn
may depend on network and interference conditions that change through time), subscriber
preference and plan-based policy, or traffic type (e.g., voice, video, browsing). Although this
adds complexity to the small-cell management, in areas where capacity comes at a premium
these tools may provide operators the ability to maximize network resource utilization. As a
result, they can either increase the transported traffic load (improving service to subscribers),
or delay the need to deploy more small cells (reducing costs, but maintaining a good service
level).
Figure 7. Integrating Wi-Fi in the mobile network: looser and tighter coupling. Source:
4G Americas
Report The economics of small cells and Wi-Fi offload
© 2012 Senza Fili Consulting • www.senzafiliconsulting.com • Reproduction and redistribution prohibited |15|
6. Small cells, Wi-Fi access and Wi-Fi offload Defining terms and scope used in the report
The terminology and definitions for small-cell and Wi-Fi are not yet firmly established; the
industry is still trying to find small cells’ and Wi-Fi’s role in mobile networks. For instance,
initially small cells had only one sector, but multi-sector small cells are now available. They can
be used in high-traffic areas to share a site location among operators or to combine multiple
wireless interfaces (e.g., 3G and LTE) in the same site.
Similarly, what should be included in Wi-Fi offload? Definitely the traffic from a laptop without
cellular connectivity is not included, but what about the traffic from a tablet with a cellular
connection that is only sporadically used?
With time the industry will settle on definitions that will emerge from established practices. In
the meantime we simply lay out the working definitions and scope used in this report. The
TCO financial analysis in this paper covers only traffic from small cells and Wi-Fi access points
that are managed by the mobile operator. Furthermore:
Macro-cell traffic is used as a reference to compare small-cell and Wi-Fi TCO.
Residential and enterprise femto cells are excluded from the analysis. The same
equipment can be used both for a small cell and a femto cell, so we distinguish small cells
from femto cells on a functional basis. Residential and enterprise femto cells are
managed by the homeowner or enterprise and share a backhaul connection that is not
owned or controlled by the mobile operator. Small cells are deployed, backhauled and
managed by mobile operators, and provide access to all subscribers within the coverage
area.
Only the Wi-Fi traffic that crosses the mobile operator’s Wi-Fi infrastructure is included in
the TCO analysis. Wi-Fi offload that takes place at home or in the enterprise, in networks
owned and managed by homeowners or enterprises, is excluded even when it involves
devices with mobile connectivity. Similarly, Wi-Fi offload traffic that uses free public
access points or third-party Wi-Fi operators is not included in the TCO model.
Table 4. What is a small cell?
Interfaces LTE, 3G, with the optional addition of Wi-Fi added as a module in
a single enclosure.
Sectorization Initially mostly single sector, with an omnidirectional antenna.
Two-sector and three-sector small cells are also available. We
expect multi-sector small cells to become common, because they
can support multiple interfaces (e.g., LTE and 3G) or facilitate
infrastructure sharing or colocation arrangements, or simply be
used in multi-sector arrangements with directional antennas.
Coverage Up to 200 meters radius, but typically deployed to cover a 50-
meter range.
Form factor Up to 10 kg compact enclosure, typically with a SoC architecture,
with integrated antennas. Small cells can be installed on a variety
of street-level or indoor assets, including utility poles, building
walls, and MSO cable strands. For some operators, a one-box
form factor (small cell and backhaul module, with antenna in the
same enclosure) is a desirable form factor.
Small-to-
macro ratio
The ratio of small cells to macro cells is still subject to heated
debate, but eventually it will depend on how rapidly traffic
grows. In the report we consider ratios up to 15 small cells to a
macro cell.
Capacity We assume that the small cell has the same capacity as a macro
cell that uses the same channel.
Report The economics of small cells and Wi-Fi offload
© 2012 Senza Fili Consulting • www.senzafiliconsulting.com • Reproduction and redistribution prohibited |16|
7. Cost assumptions Building the TCO model
To look at the economics of adding capacity to mobile networks through small cells, Wi-Fi
access points and small cells with Wi-Fi embedded modules, we built a five-year TCO
model. The model computes the TCO for a single small cell and, for reference, a three-
sector macro cell. Only RAN and backhaul costs are included in the model. While small
cells and Wi-Fi deployments require additional costs in the core network, these costs play
a minor role in the TCO and their impact is similar for all small-cell configurations and,
hence, does not have a substantial impact in the comparison among configurations.
The TCO model includes multiple small-cell configurations to allow comparisons among
them, which vary along three dimensions:
Wireless interface: LTE, Wi-Fi and/or 3G
Location: indoor and outdoor
Number of sectors: one, two or three (these options are needed for configurations
that include both LTE and 3G, and also to explore business models that require
multiple sectors).
Capex and opex inputs (Table 5 and Table 6) are used to generate the base TCO. They
derived from input from vendors, operators and independent research. We assume a mix
of wireless and wireline backhaul for each small-cell configuration that is listed in Table 5
and Table 6.
Capacity assumptions are used to compute per-bit costs for each small-cell configuration.
The estimates are intended to reflect values averaged across multiple locations, because
the capacity of cells will depend on a large number of environmental factors that the
model does not intend to capture.
To show the impact that small cells and Wi-Fi have on overall network capacity, we
looked at the overall (macro, small-cell and Wi-Fi) TCO for a varying number (from zero to
15) of small cells per macro cell. We chose this metric because it provides an easy-to-
understand framework for the comparison among configurations. The analysis can be
extended to networks of any size by using the macro cell as the basic unit, or by
calculating the contribution of each interface as a percentage.
Per-bit costs are computed as the TCO divided by the mbps. This provides a measure of
the marginal cost of adding capacity. Calculating this per-mbps cost for each small-cell
configuration is useful for comparing the cost-effectiveness of different configurations
and of small cells and Wi-Fi access points in reference to macro cells.
Cost assumptions vary greatly across countries, operators and individual markets within
the operator footprints, so we went one step further and computed the per-bit TCO as a
percentage of the macro-cell TCO. This approach increases the applicability of the results
across environments, because cost assumptions may change, but the relative differences
among small-cell configurations and between small cells and macro cells are less variable.
Report The economics of small cells and Wi-Fi offload
© 2012 Senza Fili Consulting • www.senzafiliconsulting.com • Reproduction and redistribution prohibited |17|
Table 5. Cost, capacity and backhaul assumptions for outdoor sites
Macro
Three LTE
sectors
LTE
One sector
LTE
Three sectors
3G
One sector
LTE, 3G
Two sectors
LTE
One sector,
Wi-Fi
3G
One sector,
Wi-Fi
LTE, 3G
Two sectors,
Wi-Fi
Wi-Fi
CAPEX
Equipment: base station $27,500 $3,500 $7,000 $3,500 $5,250 $4,200 $4,200 $5,950 $1,750
Equipment: wireless backhaul $7,500 $2,500 $3,500 $2,000 $3,000 $3,125 $2,500 $3,625 $1,250
Equipment: wireline backhaul $2,000 $1,000 $1,400 $800 $1,200 $1,250 $1,000 $1,450 $500
Planning, installation, commissioning $40,000 $6,500 $7,000 $6,500 $6,750 $6,600 $6,600 $6,850 $3,600
OPEX
Site lease: base station $15,000 $1,200 $1,440 $1,200 $1,320 $1,320 $1,320 $1,440 $900
Backhaul: wireless $6,000 $1,800 $1,980 $1,800 $1,980 $1,980 $1,980 $2,160 $540
Backhaul: wireline $24,000 $10,000 $12,000 $8,000 $11,000 $11,000 $8,800 $12,000 $6,000
Power, maintenance, etc. $10,000 $1,250 $1,640 $1,250 $1,445 $1,445 $1,445 $1,640 $765
Capacity
Mbps 162 54 162 13.5 67.5 102 61.5 115.5 48
Backhaul
% wireless 50% 70%
% wireline 50% 30%
Report The economics of small cells and Wi-Fi offload
© 2012 Senza Fili Consulting • www.senzafiliconsulting.com • Reproduction and redistribution prohibited |18|
Table 6. Cost, capacity and backhaul assumptions for indoor sites
LTE
One sector
LTE
Three sectors
3G
One sector
LTE, 3G
Two sectors
LTE
One sector,
Wi-Fi
3G
One sector,
Wi-Fi
LTE, 3G
Two sectors,
Wi-Fi
Wi-Fi
CAPEX
Equipment: base station $2,975 $5,950 $2,975 $4,463 $3,570 $3,570 $5,058 $1,488
Equipment: wireless backhaul $1,875 $2,250 $1,500 $2,250 $2,250 $1,800 $2,625 $2,000
Equipment: wireline backhaul $400 $480 $320 $480 $500 $400 $580 $400
Planning, installation,
commissioning $5,225 $5,650 $5,225 $5,438 $5,310 $5,310 $5,523 $3,093
OPEX
Site lease: base station $720 $864 $720 $792 $792 $792 $864 $540
Backhaul: wireless $800 $1,080 $990 $990 $990 $990 $1,080 $990
Backhaul: wireline $5,000 $6,000 $4,400 $5,500 $5,500 $4,400 $6,000 $4,000
Power, maintenance, etc. $970 $988 $970 $979 $979 $979 $988 $596
Capacity
Mbps 54 162 13.5 67.5 102 61.5 115.5 48
Backhaul
% wireless 10%
% wireline 90%
Report The economics of small cells and Wi-Fi offload
© 2012 Senza Fili Consulting • www.senzafiliconsulting.com • Reproduction and redistribution prohibited |19|
8. Comparing the costs for macro cells, small cells and Wi-Fi The base TCO model
Macro cells versus small cells. The TCO for a three-sector LTE macro cell over a five-
year period is $279,412, assuming a 50:50 mix of wireless and fiber backhaul (Figure
8). A macro cell can cost more than six times as much as a small cell (in this case the
3G small cell, which has the lowest TCO among the configurations considered). For
the three-sector LTE small cell with Wi-Fi, which is the most expensive small cell to
deploy in our lineup, the TCO as a percentage of macro-cell TCO ranges from 16% to
23% for outdoor small cells, and from 13% to 20% for indoor small cells, which are
less expensive to deploy and operate than outdoor ones (Figure 9). The TCO for
Wi-Fi is even lower, at 10% of macro-cell TCO for outdoor access points and 11% for
indoor ones.
The differences in cost stem from the larger size and more stringent requirements of
macro cells, which in turn drive higher backhaul, lease and equipment costs. The
percentage of the TCO that is due to capex and equipment is approximately the
same for macro and small cells, and slightly lower for Wi-Fi (Figure 10).
One major difference in the TCO comes from backhaul. In macro cells, backhaul
accounts for 36% of TCO, while for small cells it accounts for 60% to 48%, and for
Wi-Fi for 65% to 46% (Figure 11). The higher impact of backhaul is due to the
relatively higher costs associated with RAN equipment for macro cells. In small cells,
RAN and backhaul equipment are much closer in size and cost, so the RAN and
backhaul account for comparable portions of the overall TCO.
Small cells versus Wi-Fi. In our TCO model, Wi-Fi costs less than a small cell to
deploy and operate not primarily because of the equipment costs (which account for
10% to 14% of the TCO), but because Wi-Fi access points are operated under less
stringent performance and reliability requirements than small cells. The need to
coordinate transmission with the macro network to manage interference makes it Figure 8. Outdoor and indoor TCO. Source: Senza Fili
Report The economics of small cells and Wi-Fi offload
© 2012 Senza Fili Consulting • www.senzafiliconsulting.com • Reproduction and redistribution prohibited |20|
imperative that small cells have carrier grade or near-carrier-grade reliability and low
latency.
A mobile operator may choose to deploy a carrier-grade Wi-Fi hotspot network, and
in that case the TCO would be the same as for small cells, but typically this is not how
Wi-Fi access points are deployed today. They are managed independently from the
mobile network and used as complementary but not essential network resources, so
carrier grade is not required. Wi-Fi networks that are not carrier-grade are likely to
have a lower TCO, but also have lower reliability and performance. This has to be
kept in mind this when comparing the Wi-Fi TCO to the small-cells TCO as it accounts
for most of the cost differences between the two solutions.
Because of this, the model assumes that outdoor Wi-Fi access points may use lower-
cost sub-6 GHz license-exempt spectrum for backhaul, while small cells use only
licensed spectrum in the sub-6 GHz band. Small cells in the model can also use
license-exempt spectrum in the 60 GHz band, but in that band, interference is not an
issue yet because it is lightly used and uses PTP transmission.
Incidentally, the use of license-exempt sub-6 GHz backhaul at outdoor locations
accounts for the higher cost of indoor Wi-Fi access points, which do not take equal
advantage of the cheaper wireless backhaul. In the small-cell case, indoor TCO is
slightly lower because outdoor backhaul is more expensive (and in turn this is due to
the fact that outdoor small cells use only licensed spectrum for NLOS backhaul).
The TCO for an outdoor Wi-Fi access point is 57% of that of a single-sector LTE small
cell, and 61% of a 3G small cell. For an indoor Wi-Fi access point, which is relatively
more expensive than an outdoor one, the TCO is 65% of that for a single-sector LTE
small cell, and 82% of the TCO of a single-sector 3G small cell.
Adding Wi-Fi to a small cell. To compute the cost of a small-cell and Wi-Fi enclosure,
we assumed that a Wi-Fi module was added to the small cell. Because the
requirements for small cells are more stringent, we consider this to be more
appropriate than adding an LTE or 3G radio module to a Wi-Fi access point. The
marginal equipment cost of adding Wi-Fi to a small cell is $800 for an outdoor one
and $580 for an indoor one. The addition of Wi-Fi leaves most of the other cost
Figure 9. TCO for outdoor and indoor small cells, as a percentage of macro-cell TCO. Source:
Senza Fili
Report The economics of small cells and Wi-Fi offload
© 2012 Senza Fili Consulting • www.senzafiliconsulting.com • Reproduction and redistribution prohibited |21|
items unaffected, with the exception of the backhaul, whose cost increases due to
the higher backhaul requirements driven by Wi-Fi.
As a result, the TCO for the same small-cell configuration with and without Wi-Fi
increases by only a small percentage. The addition of Wi-Fi to an LTE or 3G outdoor
small cell drives an increase in TCO ranging from 9% (three sectors) to 11% (one
sector). For indoor small cells, the corresponding figures are 8% and 11%,
respectively.
LTE and 3G. While the future is LTE small cells, there is a role for 3G small cells today,
because network congestion is almost exclusively a 3G issue today. Many operators
still don’t have an LTE network (or even the spectrum to deploy it), but they need
additional capacity today. From a TCO viewpoint, however, 3G small cells are less
cost effective, because their cost is very similar to LTE but their capacity is lower. For
an outdoor location, the TCO for a single-sector 3G small cell is 92% of the LTE TCO,
and for an indoor small cell the percentage is 88%. The difference in TCO mostly
depends on the more limited backhaul requirements for 3G cells.
If they have an LTE network or plan to deploy one, mobile operators may deploy LTE
and 3G in a two-sector small cell (or deploy 3G initially and add LTE when the
network is available), or as doing so makes the deployment more cost effective. The
addition of an LTE module to a 3G small cell adds only 13% to the TCO of an outdoor
or indoor small cell. As in the previous cases, the additional module does not have a
major impact on the cost base.
How many sectors? Unless high traffic, business model (e.g., ability to share
infrastructure), or other considerations justify it, it does not make sense to deploy
multi-sector small cells. But where a multi-sector small cell provides a benefit to the
operator, the TCO comparison to a single-sector small cell is positive. With an
additional 27% of TCO (outdoor) or 26% (indoor), the operator can move from an
LTE single-sector small cell to a three-sector small cell with three times as much
capacity.
Indoor versus outdoor. We noted above that indoor small cells have a lower TCO.
The capex is slightly lower for indoor small cells, but the main difference comes from
the opex, which for all configurations and interfaces accounts, not surprisingly, for
Figure 10. Capex and equipment as a percentage of TCO. Source: Senza Fili
Report The economics of small cells and Wi-Fi offload
© 2012 Senza Fili Consulting • www.senzafiliconsulting.com • Reproduction and redistribution prohibited |22|
most of the TCO (72% to 79% in our analysis). Specifically, indoor small cells have
access to lower-cost wireline backhaul and lower leases.
Backhaul costs. As noted above, the impact of backhaul costs is substantially higher
for small cells because, in small cells, backhaul scales less effectively than RAN
equipment. Specifically, fiber and LOS wireless backhaul links are less expensive for a
small cell than for a macro cell, but the decrease in cost is, as a percentage, smaller
than for other items, such as RAN equipment, power or lease costs.
Backhaul costs do not increase linearly with the increase in capacity. Rather, across
small cells, backhaul costs vary as a percentage of TCO, with the percentage
decreasing as the number of sectors increases. An exception to this is 3G small cells,
where a substantially lower capacity allows operators to select less-expensive
backhaul technologies. The differences are small, though, with backhaul accounting
for 48% to 60% of the TCO across the indoor and outdoor small-cell configurations
considered. Wi-Fi backhaul is lower than for small cells in the outdoor case, because
we assume that Wi-Fi access points use license-exempt spectrum. In indoor
configurations, Wi-Fi backhaul costs are similar to those for small cells, but because
the non-backhaul costs are lower for Wi-Fi, the percentage of TCO accounted for by
the backhaul is higher.
Figure 11. Backhaul capex and opex as a percentage of TCO for macro cells, indoor and
outdoor small cells, and Wi-Fi. Source: Senza Fili
Report The economics of small cells and Wi-Fi offload
© 2012 Senza Fili Consulting • www.senzafiliconsulting.com • Reproduction and redistribution prohibited |23|
9. The capacity contribution of small cells and Wi-Fi
Incremental capacity increase with more small cells per macro cell
Small cells and Wi-Fi hotspots are deployed to increase capacity, so when examining the business case, capacity assumptions and per-bit costs take center stage. In this section we
look at the capacity assumptions, so that in the following sections we can move to examine the per-bit costs.
Our model relies on a capacity estimate for each cell configuration, as shown in Table 5 and Table 6. These estimates were derived from the standards-defined estimates of the
maximum and average throughput of the wireless interface considered (10 MHz channels for LTE, 5 MHz for HSPA; Wi-Fi n). But they include a further reduction in throughput due
to the impact of interference and other environmental variables that affect transmission, such as obstruction, to ensure that the TCO model did not overestimate the contribution
of small cells.
This defensive approach is necessary because of the complexity of estimating the capacity for macro cells, small cells and Wi-Fi. They are deployed in a wide range of environments
and subscriber distribution that create a large variability in the throughput for the same base station in different locations.
In the case of small cells, there are additional considerations that further complicate estimates of capacity. In the macro network, cells are placed in a way that minimizes the
overlap in their coverage area. On the other hand, small cells use the same frequency as macro cells, and their coverage area is located within the macro coverage area. This
creates interference, at levels that depend on the location of the small cell within the macro-cell coverage area, the transmission power and other environmental factors. 3GPP-
based interference management tools such as COMP and eICIC are being introduced to coordinate macro-cell and small-cell transmission, but their impact is still unclear, as is the
rate of adoption that they will have among operators.
At the same time, small cells being placed closer to subscribers can be more spectrally efficient than more-distant macro base stations, because the small cell uses a better
modulation scheme. If placed close to where the demand is, a small cell can be highly beneficial and can add more capacity than a macro cell with the same number of sectors and
using the same spectrum but located farther away. At the same time, the reverse may be true if the small cell is not in a well-suited location.
While the uncertainty of capacity estimates cannot be avoided in a model that tries to generalize across countries and operators, we focused on the relative contribution of
different interfaces in different small-cell configurations and assumed that the capacity of a single-sector base station is the same in a macro cell and in a small cell. As a result, the
capacity of a three-sector LTE macro cell is the same as that of a three-sector LTE small cell, installed either indoors or outdoors.
For simplicity we also assumed that, as the density of small cells increases – that is, as the number of small cells per macro cell grows – the capacity increases linearly for the range
of cell densities (from zero to 15) that we considered. So, for instance, five small cells add a fivefold increase in capacity over one cell. Commercial deployments of small cells are
Report The economics of small cells and Wi-Fi offload
© 2012 Senza Fili Consulting • www.senzafiliconsulting.com • Reproduction and redistribution prohibited |24|
still rare and limited in size and have low small-cell to macro-cell ratios, so it is not yet known what the capacity gains will be as the density of small cells increases. So the linear
assumption offers a straightforward estimate that allows us to do a preliminary exploration of the contribution that small cells – and Wi-Fi – bring to mobile networks.
The results are shown in Figure 12, for eight small-cell and Wi-Fi configurations and using small-cell to macro-cell ratios that range from zero to 15. Capacity increases as the
number of sectors increases with the addition of Wi-Fi, with the transition from 3G to LTE, and, of course with the small-cell density. Small-cell and Wi-Fi capacity quickly surpasses
macro-cell capacity. Four one-sector LTE small cells or two three-sector LTE small cells contribute more capacity than a macro cell. So do three Wi-Fi access points within the
macro-cell coverage area. We need 13 single-sector 3G small cells to reach the capacity of an LTE macro cell (but the number drops to three if we use a 3G macro cell for the
comparison). Wi-Fi adds 89% to the capacity of a single-sector LTE small cell and 30% to a three-sector LTE small cell. In a 3G small cell, the Wi-Fi contribution is larger and
represents 3.5 times the small-cell capacity.
Figure 12. Capacity contribution of small cells and Wi-Fi as the density of small cells increases. Source: Senza Fili
© 2012 Senza Fili Consulting • www.senzafiliconsulting.com • Reproduction and redistribution prohibited |25|
10. Per-bit TCO Assessing the cost-effectiveness of small cells and Wi-Fi
As the next step in our analysis, we used the TCO and capacity results from the previous
sections to compute the per-bit costs as:
Per-mbps TCO over the five-year period (Figure 13)
Per-bit TCO for small cells and Wi-Fi access points as a percentage of macro-cell per-bit
TCO. (Figure 14)
For both analyses, we compare the per-bit TCO to an LTE macro cell and to a 3G macro cell,
depending on the configuration of the small cell. Wi-Fi access points were included in both
cases. The split in the comparison is due to the fact that an operator with an LTE network is
likely to include LTE in its small-cell deployments. On the other hand, comparing the per-bit
TCO for a 3G small cell to an LTE macro cell is not very relevant, because only operators with
3G-only networks are likely to deploy 3G-only small cells.
Across all the configurations, the 3G per-mbps TCO is much higher than LTE. This reflects the
lower per-bit costs of LTE due to its greater spectrum efficiency and wider channels.
Also consistently across configurations, indoor small cells have a lower per-mbps TCO than
outdoor, which reflects a lower TCO but equal capacity. For Wi-Fi access points, the difference
is reversed, following the lower TCO for outdoor Wi-Fi access points due to the lower
backhaul costs.
As the number of sectors increases or when Wi-Fi is added, per-bit costs decrease, showing
that as the total capacity of the small cell increases, the per-bit costs go down. In other words,
multi-sector small cells and small cells with Wi-Fi are more expensive to deploy and operate,
but are more cost effective.
In the single-sector outdoor small cells, the per-bit costs are 50% of the macro cell costs in the
LTE case and 47% in the 3G case. Adding two more sectors brings the costs of an LTE small cell
down to 21% of the macro per-bit costs. The addition of Wi-Fi to the single-sector LTE lowers Figure 13. Per-mbps TCO for LTE and 3G small cells and macro cells, and Wi-Fi.
Source: Senza Fili
Report The economics of small cells and Wi-Fi offload
© 2012 Senza Fili Consulting • www.senzafiliconsulting.com • Reproduction and redistribution prohibited |26|
the per-bit TCO to 30% of the LTE macro per-bit cost. Hence, adding two extra LTE sectors to a
small cell results in greater cost efficiency than adding a Wi-Fi module. However, the efficiency
gains are almost the same if we add a second sector or a Wi-Fi module to the same LTE small
cell, because the increases in TCO and in capacity are similar in the two cases (additional LTE
sector and Wi-Fi module; data not shown in graphs).
In the comparison between an outdoor single-sector LTE small cell and an outdoor Wi-Fi
access point, Wi-Fi is the more cost-effective interface, with a per-bit cost of 32% of the LTE
macro cell, versus 50% for the LTE small cell.
Combining Wi-Fi and LTE in a single-sector small cell brings a modest increase in cost
efficiency (from 32% to 30% of LTE macro per-bit TCO for an outdoor small cell), but it is an
interesting comparison that shows that the combination of Wi-Fi and LTE makes small cells as
cost effective as Wi-Fi access points.
The cost dynamics for 3G small cells are similar. Wi-Fi is a more cost effective way to transport
traffic than 3G macro cells, because the per-bit Wi-Fi TCO is 8% to 9% of the per-bit 3G macro
TCO. Wi-Fi is also cheaper than 3G small cells on a per-bit basis. Wi-Fi per-bit TCO is 17% of a
single-sector outdoor 3G small cell, and 23% of an outdoor 3G small cell. However, the
addition of Wi-Fi to a 3G single-sector small cell greatly improves the cost-efficiency of the 3G
cell, lowering the per-bit TCO of the outdoor small cell from 47% (no Wi-Fi) to 11% (with
Wi-Fi) of the per-bit TCO of the 3G macro cell.
Figure 14. Per-bit TCO for LTE and 3G small cells and Wi-Fi as a percentage of per-
bit TCO for macro cells. Source: Senza Fili
Report The economics of small cells and Wi-Fi offload
© 2012 Senza Fili Consulting • www.senzafiliconsulting.com • Reproduction and redistribution prohibited |27|
11. Findings: two (or three) is better than one Synergies among LTE, 3G and Wi-Fi strengthen small-cell business case
Expected increase in data traffic requires a massive increase in capacity in mobile
networks. Additional spectrum allocations and technological advances are not
sufficient to address this need. An increase in cell density will be the main source of
capacity expansion, because it provides the best way to increase capacity density in
high-traffic areas such as metro hotzones, stadiums, airports and commuter hubs.
Small cells and Wi-Fi play a crucial role in providing the increase in capacity density
that operators need and subscribers expect. Most operators have exhausted the
possibility of increasing capacity by adding more macro cells, so they have to direct
their efforts toward a sublayer of small cells and Wi-Fi access points in high-traffic
areas. Building this underlay network and managing it effectively requires a long-
term effort, substantial funding, and tight coordination between network layers.
Operators need both small cells and Wi-Fi to cost-effectively meet their capacity
requirements. Cellular and Wi-Fi interfaces are to a large extent complementary,
and they will both have a key role to play. Subscribers love the flexibility of Wi-Fi and
will not abandon it even as faster cellular connectivity becomes widespread with
LTE.
The small-cell and Wi-Fi infrastructure can be colocated to increase efficiency and
reduce per-bit costs. Cellular and Wi-Fi modules can be added to the same
enclosure, thus packing more capacity into each small cell without greatly increasing
the marginal cost of equipment, installation and operation. More powerful small
cells that combine LTE, 3G and Wi-Fi reduce the number of sites operators have to
manage and allow effective traffic management across interfaces, which in turn is
conducive to higher RAN resource utilization.
To maximize the benefits of small cells and Wi-Fi colocation, integration of Wi-Fi
into the mobile network core is required. This will give operators visibility into the
Wi-Fi traffic and enable them to manage it as they manage mobile traffic.
The TCO analysis shows that adding more interfaces and sectors to small cells
leads to only modest marginal increases in the TCO. Operators have to face
additional equipment and backhaul costs associated with more interfaces, but the
cost of installing and operating these more powerful small cells is largely the same as
a single-sector, single-interface small cell.
Even at low densities, LTE small cells and Wi-Fi quickly take on a dominant role
relative to macro cells in transporting mobile traffic. The contribution of 3G small
cells is smaller in terms of capacity, yet they can play a crucial role in offloading
macro traffic, because most operators today face congestion on their 3G networks,
not on their newer LTE networks.
Small cells and Wi-Fi enable operators to slash per-bit TCO by at least half. This
does not mean that operators will or should reduce their use of macro cells. Macro
traffic is more expensive to transport, but it is also more valuable, because it
provides the ubiquitous coverage and mobility support that small cells and Wi-Fi are
not designed for or capable of providing.
Per-bit TCO shows that Wi-Fi added to small cells greatly improves the small-cell
business case, especially for 3G small cells. Combining LTE, 3G and Wi-Fi increases
the cost effectiveness of small cells. The Wi-Fi contribution is particularly strong for
3G cells, which face a more challenging business model due to their lower capacity.
The capacity injection from Wi-Fi into a small cell quickly drives down the per-bit
TCO and makes 3G small cells more attractive, whether as a stand-alone proposition
or as an interim solution to be rolled out ahead of LTE deployments.
© 2012 Senza Fili Consulting • www.senzafiliconsulting.com • Reproduction and redistribution prohibited |28|
12. Acronyms
ANDSF Access network discovery and selection function
CN Core network
E-UTRAN Evolved universal terrestrial radio access network
2G Second generation
3G Third generation
3GPP Third Generation Partnership Project
4G Fourth generation
ANDSF Access network discovery and selection function
CAGR Compound average growth rate
COMP Coordinated multi-point
eICIC Enhanced inter-cell interference coordination
HetNet Heterogeneous network
HSPA High Speed Packet Access
IEEE Institute of Electrical and Electronics Engineers
IP Internet protocol
IT Information technology
LOS Line of sight
LTE Long Term Evolution
MSO Multiple system operator
NGH Next Generation Hotspot
NLOS Non line of sight
PB Petabyte
PCRF Policy and charging rules function
PTP Point to point
QoS Quality of service
RAN Radio access network
SIM Subscriber identity module
SoC System on a chip
TCO Total cost of ownership
VNI Visual Networking Index
WLAN Wireless local area network
WMM Wi-Fi Multimedia
WPA2 Wi-Fi Protected Access II
About Senza Fili Senza Fili provides advisory support on wireless data technologies and services. At Senza
Fili we have in-depth expertise in financial modeling, market forecasts and research,
white paper preparation, business plan support, RFP preparation and management, due
diligence, and training. Our client base is international and spans the entire value chain:
clients include wireline, fixed wireless, and mobile operators, enterprises and other
vertical players, vendors, system integrators, investors, regulators, and industry
associations.
We provide a bridge between technologies and services, helping our clients assess
established and emerging technologies, leverage these technologies to support new or
existing services, and build solid, profitable business models. Independent advice, a
strong quantitative orientation, and an international perspective are the hallmarks of our
work. For additional information, visit www.senzafiliconsulting.com or contact us at
[email protected] or +1 425 657 4991.
About the author Monica Paolini, PhD, is the founder and president of Senza Fili. She is
an expert in wireless technologies and has helped clients worldwide to
understand technology and customer requirements, evaluate
business plan opportunities, market their services and products, and
estimate the market size and revenue opportunity of new and
established wireless technologies. She has frequently been invited to
give presentations at conferences and has written several reports and
articles on wireless broadband technologies. She has a PhD in
cognitive science from the University of California, San Diego (US), an
MBA from the University of Oxford (UK), and a BA/MA in philosophy
from the University of Bologna (Italy). She can be contacted at
© 2012 Senza Fili Consulting, LLC. All rights reserved. No selection of this material can be copied, photocopied, duplicated in any form or by any means, or redistributed without express written permission from Senza Fili Consulting. While the report is
based upon information that we consider accurate and reliable, Senza Fili Consulting makes no warranty, express or implied,
as to the accuracy of the information in this document. Senza Fili Consulting assumes no liability for any damage or loss
arising from reliance on this information. Names of companies and products here mentioned may be the trademarks of their
respective owners.
SENZA
CONSULTING