008 Interference MGMT Topic Brief
008 Interference MGMT Topic Brief
DOCUMENT
008.04.02
Topic brief: Interference Management
in UMTS Femtocells
February 2013
www.scf.io/ www.smallcellforum.org
SMALL CELL FORUM
RELEASE Four
Small Cell Forum supports the wide-scale deployment of small cells. Its mission
is to accelerate small cell adoption to change the shape of mobile networks and
maximise the potential of mobile services.
‘Small cells’ is an umbrella term for operator-controlled, low-powered radio access nodes,
including those that operate in licensed spectrum and unlicensed carrier-grade Wi-Fi. Small
cells typically have a range from 10 metres to several hundred metres. These contrast with
a typical mobile macrocell that might have a range of up to several tens of kilometres. The
term ‘small cells’ covers residential femtocells, picocells, microcells and metrocells.
This document forms part of Small Cell Forum’s Release Four: Urban. Urban small cells
are at an earlier stage in their commercial development than their more mature residential
and enterprise counterparts. As such, the present Release focuses on establishing the need,
evaluating the business case and identifying key barriers to commercial deployment. It
offers shared deployment learnings from leading operators and vendors, further refinement
of our technical works and reporting progress on our activities to strengthen the ecosystem
through improved multivendor interoperability.
Release Four also contains works clarifying market needs and addressing barriers to
deployment of residential, enterprise and rural small cells.
Small Cell Forum Release website can be found here: www.scf.io and an overview of all the
material in Release Four: Urban can be found here: www.scf.io/doc/104
All content in this document including links and references are for informational purposes only and
is provided “as is” with no warranties whatsoever including any warranty of merchantability, fitness
for any particular purpose, or any warranty otherwise arising out of any proposal, specification, or
sample.
No license, express or implied, to any intellectual property rights is granted or intended hereby.
©2007-2014 All rights reserved in respect of articles, drawings, photographs etc published in
hardcopy form or made available in electronic form by Small Cell Forum Ltd anywhere in the world.
If you would like more information about Small Cell Forum or would
like to be included on our mailing list, please contact:
Email [email protected]
One of the concerns for mobile operators considering large-scale femtocell deployment is to minimise and
risk of harmful interference to their existing network. Management of interference is a normal part of the
planning and operation of today’s macrocell-oriented networks, permitting essentially unlimited volumes of
traffic to be accommodated in limited spectrum.
The interference management techniques which have been developed by the femtocell community enable
femtocells to be distributed within a macrocell network, thereby allowing much more efficient reuse of
spectrum. They automate real-time interference management, while ensuring that operators are always in
control of the behaviour of every individual femtocells.The resulting network can supply much higher data
and voice capacity and higher data throughputs than can be supported in a traditional macrocell-only
network.
This report reviews the findings of research and simulations performed for 3G UMTS technology, including
HSPA (High Speed Packet Access) at both 2100 MHz and 850 MHz bands, which are also applicable to other
similar bands such as 900 MHz and 1900 MHz. The initial studies by 3GPP were expanded by Small Cell
Forum with further analysis and simulations on the most challenges scenarios in these frequency bands.
The report also indicates how large-scale deployments have verified the practical applicability of these
techniques with positive performance results.
• Femtocells enable dense frequency re-use to significantly increase network capacity as well
enhancing coverage exactly where consumers want it.
• Femtocells deployed on a dedicated carrier or configured as ‘open access’ can provide these
benefits with little or no negative impact to the existing macrocell service.
• Where a dedicated carrier is not available and the femtocell is configured in closed access mode,
a number of corner case interference scenarios may occur at the extremities of the coverage.
Femtocells prove that automated interference management works in large scale commercial deployments
• Automated techniques have been developed and proven in large scale commercial deployments
which mitigate interference to ensure the benefits of femtocells can be realised without any
compromise to user experience.
• Simulation of a typical co-channel, closed access deployment scenario shows adding femtocells
to an existing macrocell network brings 100x improvement in network capacity and mean user
throughput.
• This topic brief provides an overview of the work of Small Cell Forum in analysing the
interference scenarios and proposing solutions.
Tables
Table 3-1 Summary of Interference challenges and mitigation techniques ............ 8
Table 4-1 Performance benefits of HSDPA Femtocells ........................................ 9
Figures
Figure 2-1 Contribution of different factors to increases in capacity of wireless
networks. Source: Agilent ................................................................ 2
Figure 2-2 Channel Sharing Options for Femtocells and Macrocells ....................... 3
Figure 2-3 Access Mode Options for Small Cells.................................................. 4
Figure 3-1 Summary of corner case interference scenarios for femtocells .............. 5
Figure 3-2 Shared carrier femtocell may create ‘deadzone’ in macrocell layer ........ 6
Figure 3-3 Additional Noise rise at macrocell from femtocell attached UEs ............. 6
Figure 3-4 Strong macrocell signals may reduce coverage area of femtocell .......... 7
Figure 3-5 High power macrocell or femtocell attached devices may ‘deafen’
femtocell receiver ........................................................................... 7
Figure 4-1 Simulated user throughput for macrocell-only and
macrocell+femtocell HSDPA networks ............................................... 9
Femtocells are a recognised solution for enhancing coverage and capacity in the home or office, or wherever
consumers desire.
A key concern from operators is whether deployment of femtocells by end users may have a negative
impact on the existing macrocell service – especially when the same spectrum channel is reused for both.
In the macrocell world, deployments are tightly planned and managed and networks regularly tested and
tuned to resolve any issues. Such effort is not practical with femtocells, due to their large numbers and
deployment by customers, which points to a need for automated management in a ‘Self Organising Network’
(SON).
The Small Cell Forum and industry as a whole have performed detailed studies into the different
interference scenarios for femtocells, including both common use-cases and the more challenging corner
cases and identified for each case techniques to mitigate them.
This paper outlines the findings of those studies, the recommended interference management techniques,
and proof that they work in the real world in large scale commercial deployments.
• Section 2 Why Interference Management is Needed explains how the different femtocell
configurations may lead to interference scenarios.
• Section 3 Corner case interference scenarios and mitigation then describes in detail the
interference scenarios themselves and the potential impact if left unremedied. It goes on to
describe the automated interference management techniques needed to mitigate the worst case
interference.
• Section 4 Implementation, adoption and benefits provides simulation results on the benefits
of femtocell networks and consideration of their applicability to high (~2GHz) and low
(~800MHz) frequency bands. We also provide a testimonial from an operator who has made a
large scale commercial deployment of femtocells.
• Section 5 Experience from large scale deployments sets out the practical experience of
operators who have deployed femtocells in large numbers since the research was conducted.
This chapter outlines the various configurations for deploying femtocells and which of these are most likely
to cause interference issues.
Demand for mobile network capacity has increased since the earliest use of wireless, and with current and
future trends of mobile broadband and machine communications, there is no sign it is about to abate.
The management of interference is the ‘central magic’ of cellular mobile technology, enabling essentially
unlimited capacity from finite quantities of spectrum. If we look over the 50 years of wireless network
capacity evolution shown in Figure 2-1, we see that reuse – by introducing more cells – is by far the
greatest contributor. Since spectrum is a finite resource and technology is close to fundamental limits, this
trend is set to continue. The move to smaller and smaller cells is inevitable.
10000
2000
1000
Factor of
capacity
increase, 100 25 25
1960 -
2010
10
1
Spectrum Spectral More Cells
Efficiency
Femtocells can be deployed wherever consumers want them, to increase coverage or capacity. However
such flexibility presents challenges to operators to manage uncontrolled deployments.
In macrocell networks interference is traditionally managed using detailed radio planning tools, extensive
ongoing field measurements and skilled radio planners and optimisers. However this traditional approach is
not viable for femtocells, where the location of cells is variable and the numbers may far exceed the
numbers of macrocells. Nevertheless, the radio planners need to remain in ultimate control of femtocells to
ensure that they do not degrade the wider network and that they are only providing a service where it
actually enhances the experience for users – both those on the femtocells and those on the surrounding
network.
1
“From Angst to Ambivalence to Acceptance: The Story of how Cellular has Evolved to Embrace Wifi”, M
Rumney, M. Lawton, Agilent Technologies Jan 2013,http://goo.gl/4djAi
The simplest way to mitigate interference concerns is to deploy femtocells on their own dedicated carrier,
although a degree of automation is still required to manage handover between femtocell and macrocell
networks.
A dedicated carrier is not always possible or economically viable given the scarcity of spectrum suitable for
mobile communications. Fortunately femtocells can be deployed in a variety of ways with respect to sharing
of channels already used for macrocells, as shown in Figure 2-2. Each has a different trade-off between the
level of spectral reuse, and the degree of interference mitigation required.
Figure 2-2 (a) represents a dedicated channel for femtocells, which avoids any interference with the macro
but is not as spectrally efficient as a fully shared allocation shown in (b). Configuration (c) has the femtocell
and macro networks sharing one of the available channels, but also provides a dedicated macro-only
channel as a fall-back option in locations where the shared carrier suffers interference. (d) represents partial
sharing, which reduces impact of sharing on any individual macro carrier.
The ‘right’ configuration depends on the operator’s spectrum holding and in which way current channel
assignments are used. The highest overall spectral efficiency comes from maximal reuse with shared
carriers. At the time of writing, the most commonly deployed options are (a) and (c).
The access modes of a small cell define which users can use it. They can be open to all, closed to only those
on a whitelist, or a hybrid mode with priority to the whitelisted users. Figure 2-3 illustrates these modes and
provides typical examples of the types of small cells that can use them. The final choice is dependent on the
service which the operator wishes to create.
It is indicated in Figure 2-3 that in open or hybrid mode, users generally use the cell with the strongest
signal, which helps keep interference to a minimum. It is still important to ensure handover between the
cells is well managed, and work is underway in 3GPP to enhance mobility procedures for UMTS and LTE
femtocells 2. Furthermore, the potential for the user device to come in very close proximity to the femtocell
requires precautionary measures to ensure that the receiver is not overloaded with too strong a signal.
In closed mode, there may be unregistered users near to the femtocell, but not able to use it. This may
cause higher levels of interference in a number of scenarios, which are described in the following section.
2
3GPP TR 37.803 “Mobility enhancements for Home Node B (HNB) and Home enhanced Node B (HeNB)”,
http://www.3gpp.org/ftp/Specs/html-info/37803.htm
In the vast majority of femtocell usage, interference is similar to that in macro networks, and the normal
adaptation, error correction and mobility features of the air interface standards are sufficient.
However, femtocells sharing a carrier with the macro and in closed access mode may introduce a number of
scenarios where additional automated interference management techniques are needed to ensure users can
still enjoy seamless experience.
Figure 3-1 shows a femtocell deployment scenario which illustrates the ten different mechanisms (labelled
A-J) for interference which have each been studied in detail in our accompanying papers [1] for UMTS low
band and [2] for UMTS high band. The 3GPP study item on home base stations [3] also considers these
mechanisms.
The ten interference mechanisms can be grouped in to four main types as follows:
Interference from:
The principal concern is that interference from femtocells should not impact on existing macrocell layer
experience or at least that the combined user experience from the macrocell and femtocell networks in an
area is an improvement over macrocell alone. It is also important though that the femtocells deployed
provide a satisfying experience even in the presence of strong interference from nearby macrocells.
We describe each of the interference scenarios in more detail in the following sections:
Figure 3-2 Shared carrier femtocell may create ‘deadzone’ in macrocell layer
A deadzone is created because an unregistered user receives a much stronger from a closed access
femtocell, but yet cannot use it. Dead zones are created because there is too much small cell power in
relation to macrocell power. This is not such a problem for femtos placed in areas of strong macrocell
coverage. In areas of weak macrocell coverage, the size of the deadzone can be significant.
The forum has recommended that the femtocell ‘listen’ to the strength of the macrocell signal and scale its
downlink transmit power accordingly. This is one of our basic recommendations to vendors. Some vendors
have their own “secret sauce” that pushes this power management even further. Another benefit of
‘network listen’ is that the femtocell can detect the cell ID used by the macrocell, and ensure it uses a
different one. Cell IDs are used to encode transmissions and using different ones reduces interference.
Figure 3-2 illustrates this interference mechanism and the network listen based mitigation.
Note that this is an issue for closed access cells only. Open or hybrid femtocells allow all users to reselect on
to them, thus avoiding the deadzone entirely.
The other mitigation of the deadzone for closed access deployments is the use of the so-called “escape
carrier” shown in Figure 2-2 (c). Users will naturally reselect or hand-off to the other carrier in the presence
of the deadzone. We recognise that this is only suitable for operators with the luxury of the second carrier,
hence the continued relevance of downlink power management as described.
Figure 3-3 Additional Noise rise at macrocell from femtocell attached UEs
The additional UEs supported by femtocells will of course increase the general levels of power transmitted in
the network, which means an increased noise rise at the macrocell. Where the femtocell is located near to
the macrocell, high power transmissions from UEs near the edge of their femtocell coverage may cause
interference to the macrocell. Figure 3-3 illustrates these concepts.
Analysis of this scenario finds that in general femtocells actually reduce uplink power levels across the
network, as many devices attach to a nearby femtocell rather than a far off macrocell (assuming registered
users or open access). In the corner case scenario shown in Figure 3-4, the forum has recommended
limiting the maximum power for femtocell attached devices. Operators can remotely set the power cap in
UEs by broadcasting this in the configuration data from the femtocell.
Figure 3-4 Strong macrocell signals may reduce coverage area of femtocell
Where a macrocell is interfering with a small cell, unfortunately it is not possible to adjust the macrocell
output. It has been manually configured to give a certain coverage level and it is not practical to change
that every time a new femtocell is turned on. But the small cell can use the network listen feature
mentioned earlier to adjust its power to give the optimum balance between good coverage over its
designated area and interference beyond its reach. In addition to this, handover thresholds must be
correctly set to ensure femtocell attached UEs switch over to the macrocell expediently.
Figure 3-5 High power macrocell or femtocell attached devices may ‘deafen’ femtocell
receiver
Unlike macrocells which are normally located well away from the general public, it will be possible for user
devices to come in very close proximity to the femtocells. This means that the ‘minimum coupling loss’
between device and access point is much lower for femtocells than for macros, and certain provisions must
be taken to ensure the receiver does not become overloaded and blocked. This might occur in two
scenarios- firstly, a femtocell attached device placed in close proximity and the lowest transmit power is still
too high. Secondly, a device attached to a far off macrocell, and thus transmitting high power. The second
scenario would not occur in open access mode.
The forum recommends that femtocell receivers include adaptive uplink gain to mitigate effects of these
corner case scenarios and ensure the femtocell receiver does not become blocked. The adaptation should be
dynamic so that sensitivity is only reduced when needed.
It is also recommended that femtocells undergo extended testing of receiver dynamic range compared to
macrocells, due to the potential for a smaller ‘minimum coupling loss’.
Macrocell Uplink Femto served UEs at the edge of UL DCH transmit power capping
their femtocell may transmit high used to limit Femto UE power and
power and interfere with ensure handover of Femto UE to
macrocell, especially when in macrocell
large numbers
Femtocell Femtocell reselection issues as Network listen used here also to
Downlink strong macrocell signals can configure downlink power control and
prevent reselection to femtocell thresholds to strike a balance between
good femtocell coverage with minimal
macrocell deadzone
Femtocell Both Macro served UEs and Femto Adaptive receiver gain manages
Uplink served UEs near the femtocell noise rise in the femtocell and
may ‘deafen’ a femtocell if prevents blocking.
transmitting high power, or if in Extended Tests for Dynamic Range:
very close proximity Lower minimal coupling loss in
femtocell environment requires
conformance testing across a wider
dynamic range of input signals
Table 3-1 summarises the various interference management techniques recommended to address the
different corner case scenarios. In each case the operator maintains control of the overall policy, and can
strike their preferred trade-off between femtocell and macrocell performance.
To appreciate the level of performance improvement that femtocells can bring, a simulation study was
conducted by Forum members to compare user experience and network capacity in macrocell only and
macrocell+femtocell networks. Key assumptions were:
• 3 macrocells
• 24 femtocells, with one active UE each
• 10 active macrocell UEs
• In the macrocell only network, all 34 UEs connect to macrocells
• Closed access femtocells
• Shared carrier for macrocell and femtocell
Many experiments were repeated with different random UE and cell locations to build up a statistical
characterisation of the user throughput distributions shown in Figure 4-1.
Figure 4-1 Simulated user throughput for macrocell-only and macrocell+femtocell HSDPA
networks
In Figure 4-1, the full Cumulative Distribution Function (CDF) is shown on the right with a zoom in on the
worst case throughputs on the left. Firstly it is clear that user throughput is dramatically improved for all
users. In the macrocell-only network, less than 10% users achieve more than 200 kbps. Adding femtocells
increases this to 80% of users. Note that it is not just the small cell-connected users who gain a better
mobile internet experience, so do the macrocell users. With less noisy devices competing for attention,
interference is substantially reduced. The macrocell is no longer clogged up by the difficult-to-service indoor
users.
Further characteristics are summarised in Error! Reference source not found.. This shows headline
benefits of femtocells of nearly 100x to both the average user throughput and total network capacity. Such
gains are achieved through dense reuse of the mobile spectrum accompanied by the interference
management techniques described above to ensure the benefits for some do not come at the cost of others.
Separate simulation studies were conducted at 2100 MHz and for the lower 850/900 MHz frequency bands.
The lower frequencies introduce three changes to the simulations:
The simulations utilised the same extreme scenarios and produced very similar results for both the low and
high frequency bands, and concluded that the mitigation techniques identified for the 2100 MHz band were
appropriate and effective for both bands.
The 850 MHz study indicates that deployments of femtocells at lower frequency bands can serve the needs
for capacity and coverage to the same degree as higher bands. For macrocells, lower bands are sometimes
regarded as more suitable for coverage needs, while higher bands are regarded as more suitable for
capacity, arising from the different propagation characteristics and available bandwidth.
Forum studies conclude that the use of femtocell transmit power management ensures that the extra
propagation range at the lower bands is adequately offset by power reduction. The number of femtocells
required to cover a given property is also essentially identical for any band.
Femtocells have already been deployed widely in large scale commercial deployments. The following
statement from one of the early adopters confirms that interference management techniques are working
well in practice:
“We have deployed femtocells co-carrier with both the hopping channels for GSM macrocells and
with UMTS macrocells. Interference isn’t a problem.
The more femtocells you deploy, the more uplink interference is reduced”
Gordon Mansfield, Executive Director of RAN Delivery, AT&T, speaking at CTIA March 2010, quoted
at http://www.lightreading.com/blog.asp?blog_sectionid=414&doc_id=189942
The techniques that SCF has identified in its studies are also fully supported by appropriate ‘hooks’ in the
relevant standards from both 3GPP [4] and 3GPP2 [5], enabling the techniques to be fully implemented
while still enabling differentiation and further performance enhancements.
Femtocells using the same carrier and with closed access require adaptive interference management to
avoid a number of pathological scenarios.
Given these techniques, femtocells can increase network capacity by around 100 times based on simulations
of femtocells alongside macrocells in a co-channel deployment with closed access to the femtocell.
Experience from large scale commercial deployment of femtocells shows that these automated interference
management techniques work in practice. The full details of the underlying technical analysis are provided in
[2] for frequency bands above 1 GHz and [1] for bands below 1GHz.
[4] “Frequency Considerations and Effective Spectrum Management Techniques”, SCF December 2012