Content uploaded by Mahmoud H. Qutqut
Author content
All content in this area was uploaded by Mahmoud H. Qutqut on Jun 23, 2016
Content may be subject to copyright.
Mahmoud H. Qutqut and Hossam S. Hassanein
Mobility Management In Femtocells Networks
ii
Mobility Management in Wireless Broadband Femtocells
Abstract
Current Wireless Broadband Networks (WBNs) are facing several limitations and
considerations, such as poor indoor coverage, explosive growth in data usage,
and massive increase in number of WBNs’ subscribers. Various inventions and
solutions are used to enhance the coverage and increase the capacity of wireless
networks. Femtocells are seen as a key next step in wireless communication
today. Femtocells offer excellent indoor voice and data coverage. As well
femtocells can enhance the capacity and offload traffic from macrocells. There are
several issues that must be considered though to enable the successful
deployment of femtocells. One of the most important issues is mobility
management. Since femtocells will be deployed densely, randomly, and by the
millions, providing and supporting seamless mobility and handoff procedures is
essential. We present a broad study on mobility management in femtocell
networks. Current issues of mobility and handoff management are discussed.
Several research works are overviewed and classified. Finally, some open and
future research directions are discussed.
iii
Mobility Management in Wireless Broadband Femtocells
Table of Contents
List of Figures ............................................................................................................................ iv
List of Tables .............................................................................................................................. iv
List of Abbreviations .................................................................................................................. v
1 Introduction .......................................................................................................................... 1
2 Femtocells ............................................................................................................................ 2
2.1 What are femtocells? .................................................................................................... 2
2.2 Brief history and current status ................................................................................... 3
2.3 Comparison between femtocells with other coverage solutions .............................. 3
2.4 Benefits of using FBSs for users and operators: ....................................................... 4
2.5 Femtocell network architecture and functionalities ................................................... 5
2.6 Deployment configurations .......................................................................................... 6
2.7 Femtocells challenges and open issues .................................................................... 6
3 Mobility Management in Femtocell Networks .................................................................... 8
3.1 Mobility management in WBNs .................................................................................... 8
3.2 Femtocell mobility elements ...................................................................................... 10
3.3 Access control ............................................................................................................ 13
3.4 Paging .......................................................................................................................... 13
3.5 Idle mode, cell selection and cell reselection ........................................................... 13
3.6 Connected mode and handoff .................................................................................... 14
3.7 Mobility management issues ..................................................................................... 15
4 Related Work ...................................................................................................................... 17
4.1 Mobility management techniques .............................................................................. 17
4.2 Evaluation criteria ....................................................................................................... 17
4.3 Proposed mobility management schemes ................................................................ 18
5 Conclusion and Open Issues ............................................................................................ 28
References................................................................................................................................. 29
iv
Mobility Management in Wireless Broadband Femtocells
List of Figures
Figure 1: Femtocell Network Overview .............................................................................. 2
Figure 2: Femtocell Network Architecture .......................................................................... 5
Figure 3: Overview of Mobility Management Functionalities in Femtocell Networks........ 10
Figure 4: Two Tiers Network ........................................................................................... 11
Figure 5: HO Scenarios in Femtocell Networks .............................................................. 14
List of Tables
Table 1: Comparison between Femtocell, Microcell, and Distributed Antenna (DA) ......... 3
Table 2: Comparison of HO Schemes ............................................................................. 24
Table 3: Comparison of Scanning Schemes ................................................................... 26
v
Mobility Management in Wireless Broadband Femtocells
List of Abbreviations
3G
3
rd
Generation
3GPP
3
rd
Generation Partnership Project
4G
4
th
Generation
AAS
Advanced Antenna System
ACL
Access Control/CSG List
AP
Access Point
ARPU
Average Revenue Per User
BS
Base Station
CN
Core Network
CPE
Consumer Premises Equipment
CSG
Closed Subscriber Group
DA
Distributed Antenna
DL
Downlink
DSL
Digital Subscriber Line
eNB
enhanced NB
E
-
UTRAN
Evolved UTRAN
FAP
Femto AP
FBS
Femto BS
FDD
Frequency Division Duplex
Femto
-
GW
Femto Gateway
GPRS
General Packet Radio Service
GPS
Global Positioning System
GSM
Global System for Mobile Communications
HCS
Hierarchical Cell Structure
HLR
Home Location Register
HO
Handoff
HSDPA
High Speed Downlink Packet Access
HSPA
High Speed Packet Access
HSPA+
Evolved HSPA
HSUPA
High Speed Uplink Packet Access
IEEE
Institute of Electrical and Electronics Engineers
IMT
-
2000
International Mobile Telecommunications 2000
IP
Internet Protocol
ITU
International Telecommunication Union
K
m
ph
Kilometre per hour
LAI
Location Area Identify
LAU
Location Area Update
LTE
Long Term Evolution
LTE
-
A
LTE-Advanced
MAC
Media Access Control
MBN
Mobile Broadband Networks
Mbps
Megabits per second
vi
Mobility Management in Wireless Broadband Femtocells
MHz
Megahertz
MIMO
Multiple-Input Multiple-Output
MOG
Multimedia Online Gaming
NCL
Neighbour Cell List
NB
NodeB
OA&M
Operations, Administration, and Management
OFDMA
Orthogonal Frequency Division Multiplexing Access
OSG
Open Subscriber Group
PCI
Physical Cell Identifiers
PLMN ID
Public Land Mobile Networks Identity
QAM
Quadrature Amplitude Modulation
QoS
Quality of Service
QPSK
Quadrature Phase Shift Keying
RNC
Radio Network Controller
RSS
Relative/Received Signal Strength
RSSI
Received Signal Strength Indication
SAE
System Architecture Evolution
SC
-
FDMA
Single Carrier Frequency Division Multiple Access
SeGW
Security Gateway
SINR
Signal-to-interference ratio
SOHO
Small Office, Home Office
SON
Self-Organization Network
TAI
Tracking Area Identity
TDD
Time Division Duplex
UE
User Equipment
UL
Uplink
UMTS
Universal Mobile Telecommunication System
USIM
Universal Subscriber Identity Model
UTRAN
UMTS Terrestrial Radio Access Network
VLR
Visitor Location Register
VoIP
Voice over IP
WBN
Wireless Broadband Networks
WCDMA
Wideband Code Division Multiple Access
Wi
-
Fi
Wireless Fidelity
WiMAX
Worldwide Interoperability for Microwave Access
WLAN
Wireless Local Area Networks
WWAN
Wireless Wide Area Networks
1
Mobility Management in Wireless Broadband Femtocells
1 Introduction
There has been a significant interest in wireless broadband technologies over the past
two decades. Wireless Broadband Networks (WBNs) such as 3
rd
Generation (3G) and 4
th
Generation (4G) networks provide high data rates, large coverage areas, and high quality
multimedia services. However, existing 3G and 4G networks share a number of
drawbacks, including the limited cellular data capacity and poor indoor coverage. The
latter poses a major limitation in cellular usage, especially since up to 65% of voice and
90% of data traffic take place indoors [1]. There is also a tremendous growth of data
usage in cell phones, driven by the popular data-hungry applications such as Multimedia
Online Gaming (MOG), Mobile TV, Voice over IP (VoIP), Video Calling, Streaming TV,
Web2.0, Video-on-Demand, Location-Based services, social networks (Facebook,
Google+, MySpace). The aforementioned factors mandate a solution that remedies the
capacity and coverage constraints. Therefore, different solutions are proposed to solve
these issues. The solutions can range from deployment of heterogeneous networks with
Wireless Fidelity (Wi-Fi) for dual mode devices to installation of more cell sites and relay
stations, as well as signal boosters. Femtocells were also introduced as a device-
compatible and cost effective solution.
Femtocell networks are seen as a promising solution for enhanced indoor coverage and
increased network capacity, as well as offloading traffic from the macro/micro-cells.
Perhaps one of the key requirements for mass deployments and feasibility of femtocells
is mobility management. Femtocells have many unique characteristics that make mobility
management a critical and difficult process. Such challenges include random deployment
in an ad-hoc manner, working in a licensed spectrum, overlaying with macro/micro-cells
and backward compatibility requirements with the existing infrastructures and devices.
This chapter gives an overview of femtocell networks and mobility management. Also, it
presents currents problems and issues in mobility management. The chapter discusses,
reviews, and classifies several recent research efforts on mobility management in
femtocell networks.
The remainder of this chapter is organized as follows. An overview of the background
topics related to the femtocell networks is presented in Section 2. In Section 3, mobility
management and femtocell networks challenges are discussed. The related and common
research efforts are studied in Section 4, as well as classifications and comparisons of
proposed solutions. Finally, Section 5 discusses the open research issues and concludes
chapter.
2
Mobility Management in Wireless Broadband Femtocells
2 Femtocells
This section provides an overview of femtocell networks and some related aspects.
2.1 What are femtocells?
A femtocell is a cell in a cellular network that provides radio coverage and is served by a
Femto-BS (FBS)
1
. FBS also known as a Home-BS or a Femto-Access Point (FAP), is a
mini low-power BS installed by end users. FBSs are typically deployed indoors
residential, Small Office Home Office (SOHO) and enterprise to provide better coverage,
especially where access would otherwise be limited or unavailable. FBSs also offer
enhanced data capacity and offload traffic from the macro/micro networks. FBSs look like
broadband modems and some FBS manufacturers offer a choice of all in one box (DSL
modem, Wi-Fi router, and FBS) [2]. Femtocells operate in the licensed spectrum, and
have tens of meters of coverage range and can support up to ten active users in a
residential setting. FBSs connect to standard cellular phones and similar devices
through their wireless interfaces (LTE, WiMAX, HSDPA+) [3]. Traffic is then routed to the
cellular operator’s network via a broadband connection (e.g. xDSL, cable) as shown in
Figure 1.
Figure 1: Femtocell Network Overview
Femtocells can also be deployed outdoors, and can be used in urban areas and subway
stations [4, 5].
1
In this document, we use FBS to stand for the device itself (BS), and use Femtocell to refer to the coverage area that
covered by a FBS.
3
Mobility Management in Wireless Broadband Femtocells
2.2 Brief history and current status
The home base station concept was introduced by Bell Labs of Alcatel-Lucent in 1999
[16]. In 2002, Motorola announced the first 3G home base station [6]. In 2006, femtocell
as a term was introduced [17]. In late 2007, the Femto Forum was founded as a non-
profit membership organization to promote and enable femtocell technologies worldwide.
The forum changed its name to Smallcell Forum in 2012. The forum supports the
adoption of industry wide standards, regulatory and interoperability of femtocells by
telecom operators around the world [7]. Currently, Smallcell Forum includes more than
60 mobile operators and 74 vendors. Furthermore, 27 FBS vendors and more than 45
telecom providers have announced commercial launches of FBSs [7].
2.3 Comparison between femtocells with other coverage
solutions
There are many coverage solutions that have been developed to solve the problem of
indoor converge. In microcell and picocell solutions, operator installs micro-BSs (with
smaller coverage area than macrocells) to improve coverage and capacity in urban or
high density areas with poor reception. The Distributed Antenna (DA) solutions,
operator installs DA elements as signal boosters, which are connected to macro-BS via a
dedicated Fiber or Microwave link. These coverage enhancement solutions are typically
expensive and require operator’s involvement. Table 1 presents a comparison between
these solutions.
Femtocell
Microcell
Distributed
Antenna (DA)
Cost
Low
High Low
Network
Capacity
Very increased Increased Limited to BS capacity
Install
User Operator Operator
Capital
Expenditure
Purchase a FBS Installing new cell BS Antenna element and
backhaul installation
Operating
Expenditure
Provide a
broadband
connection
Electricity, site lease,
maintenance, and
backhaul
Antenna maintenance
and backhaul connection
Indoor coverage
problem
Enhance indoor
coverage
Does not entirely solve
indoor coverage
Does not solve indoor
coverage
Table 1: Comparison between Femtocell, Microcell, and Distributed Antenna (DA)
As shown in Table 1, the cost of a FBS is less than $200, which is a relatively low-cost
solution. Whereas the cost of a micro-BS is between $200,000-400,000, and that of a DA
4
Mobility Management in Wireless Broadband Femtocells
is between $200-400. Deploying femtocells will particularly increase the network capacity
via used the broadband connections for FBSs’ backhaul; where deploying more
microcells will also increase the network capacity, but not as much in femtocell
deployment, due to the limitations of the dedicated backhaul for Microcells. DAs will not
increase the capacity since they are connected to the macro-BSs. FBSs are installed as
end user devices, where micro-BSs and DAs are installed by operators. The capital
expenditure of femtocells is the hardware cost. However in microcells, the capital
expenditure is in installing new BS and its cell site, where for DA, installing of antenna
elements and backhaul installation add to the cost. The operational expenditure for
femtocells is in providing backhaul broadband connection. However, the operational
expenditures of microcells are electricity, site lease, maintenance cost, and backhaul,
and for DAs is antenna maintenance and backhaul connection. Femtocells solve the
indoor coverage problem; microcells and DAs do not entirely solve indoor coverage.
2.4 Benefits of using FBSs for users and operators
There are many potential benefits from the deployment of femtocells. These benefits are
summarized below [8, 9, 10, 2, 11].
User’s benefits:
• Improved indoor coverage for both data and voice services, since FBS is closed
to the users.
• Improved data rate capacity, because FBS uses the user’s high data rate
broadband connection as its backhaul.
• Reduced indoor cost charged (zone pricing). As the operators will offer attractive
pricing plans for indoor calls and data.
• Reduced power consumption for UEs due to the lower transmit power of FBS
when compare to macro/micro-BSs.
• Ability to offer new services, e.g., Home Gateway, Connected Home, location
based services.
• Simple deployment, as FBS works as a “plug-and-play” device.
• No need for new expensive dual mode UEs, as current UEs work with femtocells.
Operator’s benefits:
• Reduced capital expenditures, since no additional expensive macro/micro-BSs
are needed.
• Lower operational expenditures, because no new cell site, cell site backhaul or
maintenance costs are needed.
• Increased mobile usage indoors due to the low-cost fare, hence increasing the
Average Revenue Per User (ARPU).
• Reduced customer churn rate because customers will be potentially more
satisfied with the offered services through femtocells.
5
Mobility Management in Wireless Broadband Femtocells
2.5 Femtocell network architecture and functionalities
The following is a description of the main common components of a femtocell network
(shown in Figure 2).
Figure 2: Femtocell Network Architecture
FBS is a device located at the customer’s premise that interfaces with mobile devices
over-the air radio interface that functions as a BS [12].
Femto Gateway (Femto-GW) is an entry element to the operator’s core network. It acts
as concentrator to aggregate traffic from a large number of FBSs [11]. Also, the Femto-
GW could operate as a security gateway to provide authentication to allow data to/from
authorized FBSs to protect the operator’s CN from the public environment of the Internet,
and to terminate large numbers of encrypted IP data from hundreds of thousands of
FBSs. Also, there could be another element implanted separately called Security
Gateway (SeGW) does the security functions [13].
Femto Management System provides management protocols for “plug-and-play”
Operations, Administration, and Management (OA&M) of FBSs [14]. Broadband Forum
TR-069
2
has been selected as the framework for femtocell management and was widely
supported by 3GPP2 and 3GPP vendors and carriers, as the femtocell management
protocol [10, 15, 3].
2
Technical Report 069 (TR-69) is a Broadband Forum technical specification entitled CPE WAN Management
Protocol (CWMP). It defines an application layer protocol for remote management for end-user devices.
6
Mobility Management in Wireless Broadband Femtocells
2.6 Deployment configurations
There are many possible cases of deployment configurations for FBSs. The possible
configurations are classified depending on: access mode, spectrum allocation types, and
transmit power.
Access modes
An important characteristic of FBSs is their ability to control access. There are three
common access control modes: Open and Closed, and Hybrid [16].
1. Closed access mode also known as Closed Subscriber Group (CSG). In this
scenario, a FBS serves a limited number of UEs they defined before in its Access
Control List (ACL) [17]. This can be used in homes or enterprise environments.
2. Open access mode also known as Open Subscriber Group (OSG). In this
scenario, any UE can connect to the FBS without restrictions. This can be used in
hotspots, malls and airports.
3. Hybrid access mode is an adaptive access policy between CSG and OSG. In this
scenario, a portion of FBS resources are reserved for private use of the CSG and
the remaining resources are allocated in an open manner [17].
Spectrum planning
Allocation of the available spectrum in femtocell deployments can be one of following [9,
18]:
1. Dedicated spectrum: In this approach, different frequencies are assigned for
femtocells and macro/micro-cells.
2. Partial co-channel: In this approach, macro/micro-cells and femtocells share
some spectrum and the rest of the spectrum is reserved for macro/micro-cells
only.
3. Shared spectrum: In this approach, macro/micro-cells and femtocells share all
available spectrum.
Transmit power configuration
The configuration process of downlink and uplink transmit power of FBSs can be fixed
maximum or adaptive [19].
2.7 Femtocells challenges and open issues
Despite many benefits and advantages of femtocell networks, they also come with their
own issues and challenges. These issues and challenges need to be addressed for
successful mass deployment of femtocell networks. The most relevant issues include:
• Interference: unplanned deployment of a large number of FBSs introduces
interference issues for the mobile networks. Frequency interference is one of the
most crucial issues that impair femtocell deployment. Frequency interference in
7
Mobility Management in Wireless Broadband Femtocells
femtocells includes: Cross-tier and Co-tier interference [20]. In Cross-tier
interference, a FBS interferes with macro/micro-BS or vice versa. In Co-tier
interference, a FBS interferes with another neighbouring FBS or FBS user.
• Security and QoS: since FBSs use non-dedicated fixed broadband connections
(i.e. xDSL) that carry femto and non-femto traffic, managing and controlling
voice/data priority and security over third party becomes more difficult [21, 22]
unless Internet backhaul belongs to the same cellular operator.
• Location and synchronization: FBSs operate in the licensed spectrum, thus the
exact locations need to be verified, as well as inter-cell synchronization for proper
femtocell deployments [9]. Also, location information is essential to provide
tracking in emergency calls. However, Global Positioning System (GPS) that is
used in macro/micro-BSs for synchronization and location cannot be used in FBSs
due to the lack of the coverage of GPS indoor, since the typical deployed FBSs
are indoors [23].
• Integration of FBS into the CN: traffic between FBS and the CN send/receive
through broadband networks, so it is necessary to determine how FBSs integrate
with CN, with or without gateways; and what interface they want to connect FBS
with CN, or there might a need to upgrade the CN (software/hardware) to be
connected to Femto-GW. Many possible configurations are available [21].
• Self-Organization Network (SON) and auto configurations: FBS as a
Consumer Premise Equipment (CPE) are deployed as plug-and-play devices, so it
shall integrates itself into the mobile network without user intervention [5, 24].
Hence, different SON and auto configuration algorithms and techniques are
needed.
• Mobility and handoff management: considering that FBSs will be deployed
densely and by the millions [24], and may not be accessible to all users, mobility
management in femtocell networks (such as searching for FBS, handoff from/to
macro/micro-BS, access control) becomes one of the most challenging issues
[23]. (More details in Section 3.6.)
8
Mobility Management in Wireless Broadband Femtocells
3 Mobility Management in Femtocell Networks
In Section 3.1, we present principles of mobility management in wireless networks. An
overview of mobility management in femtocell networks is provided in Section 3.2.
3.1 Mobility management in WBNs
Mobility management is a set of tasks for controlling and supervising mobile User
terminals or Equipments (UE) in a wireless network to find them for delivery services, as
well as, to maintain their connections while they are moving [6]. Mobility management is
concerned with various aspects, e.g. Quality of Service (QoS), power management,
location management, handoff management, and admission control. It is one of the most
essential features in wireless communications due to the direct impact on user
experience, network performance and power consumption [5]. The two core components
of mobility management are location management and handoff management [7]. In the
following subsections, different mobility management procedures and aspects are
presented.
3.1.1 Location management
Location management includes two componenets, namely registration and paging.
Registration is the task of knowing where the UE is located to handle incoming and
outgoing calls. The registration process is performed via a database called the Home
Locations Register (HLR). The HLR contains information about the UE and its
capabilities. It also describes the home area of the user [8]. Another database is the
Visitor Location Register (VLR); it is attached to the HLR to maintain the current location
of the user. The VLR is updated whenever the user moves from one area to another area
[5]. Paging is a process that allows the network to page the UE when it is setting up a
call. The paging process is a message to be sent to the serving BS in order to get a
response from the UE before sending a call or message [8].
3.1.2 Mobile modes
1. Idle mode: This mode applies when the UE has no ongoing service (data, voice). The
UE is in this mode most of the time after turning-on and registering its location, it
monitors for page message from the networks [5]. When the UE is moving with idle
mode, it performs a reselection of BS on the way. Cell reselection helps to transfer
registration (VLR) and aims to keep camped on the best available cell during the idle
mode. UE periodically searches for a better cell according to the reselection criteria
[5]. Another mobility mechanism is selection. Cell selection is the mechanism of
selecting an appropriate cell to camp on when a UE is powered on or after having lost
coverage.
2. Connected mode: This mode is when the UE has ongoing service (data, voice).
After the UE releases its active session will switch to the idle mode. When the UE is
9
Mobility Management in Wireless Broadband Femtocells
moving with connected mode, Handoff occurs from one BS to the next. More details in
Section 2.2.3 [5].
3.1.3 Handoff management
Handoff (HO) management is the main function by which wireless network support
mobility and to maintain quality of service. HO enables the network to keep the UE’s
connection while it moves from the coverage area of one cell/sector to another [8]. HO is
the process of transferring an ongoing voice call or data session from one cell connected
to the CN to another. HO is divided into two broad categories: Hard and Soft HOs [9]. In
Hard HO, current resources are released before new resources are used. However in
Soft HO, both existing and new resources are used during the HO process.
HO initiation
HO initiation is the process of deciding when to request a HO. The four basic HO
initiation techniques are [10]:
1. Relative Signal Strength (RSS): In this technique, the Received Signal Strength
Indicator (RSSI) is measured over time, and the UE chooses the BS with the
strongest signal to handoff.
2. Relative Signal Strength with Threshold: This technique introduces a threshold
value. If the current signal is weak (less than a threshold) and the other signal is
stronger than the current signal and the threshold, then the UE will handoff.
3. Relative Signal Strength with Hysteresis: In this technique, the UE will be allowed
to handoff if the new BS is sufficiently stronger than the current BS (via Hysteresis)
4. Relative Signal Strength with Hysteresis and Threshold: This technique allows a
UE to handoff to a new BS if and only if the current signal level is below a certain
threshold and signal of the target BS is stronger than the current BS by a given
hysteresis margin.
HO decisions
There are several methods for performing HO. The primary HO decision protocols
include [11, 10]:
1. Network-Controlled HO: The network is responsible for overall HO decision and
handles the necessary RSS measurements.
2. Mobile-Controlled HO: The UE totally controls the HO process. The UE and BS
make the necessary measurements and the BS sends them to the UE. The UE then
decides when to handoff.
3. Mobile-Assisted HO: The UE makes RSS measurements and then sends them to
the network or BS to determine when to handoff. In this section, we describe the
mobility management procedures and issues in femtocell networks.
10
Mobility Management in Wireless Broadband Femtocells
3.2 Femtocell mobility elements
Nowadays, there are more than 100 million users of femtocell on more than 30 million
BSs [7]. Hence, with a large number of femtocells randomly deployed, it is difficult to treat
it as a normal macro/micro-cell. In addition, the network cannot afford broadcasting the
femtocell information as this will affect the overall performance of the network [26].
However, some identifiers and techniques for femtocells are needed to reduce the impact
and improve the mobility management of femtocell networks.
Figure 3: Overview of Mobility Management Functionalities in Femtocell Networks
3.2.1 Femtocells vs. macro/micro-cells
According to identifiers and techniques that can distinguish femtocells from macro/micro-
cells the network can be divided into two tiers [1] as shown in Figure 4. Such division
minimizes the signalling overhead across tiers and shortens the Neighbor Cell List (NCL)
that the UE scans when performing a HO. The methods proposed for such classification
are listed below.
Mobility Management in Femtocell
Networks
Femtocell
Characterizations
Distinguish
Femtocells
Finding
Neighbouring
Femtocells
Distinguish
Accessible
Femtocells
Allowed List
Access Control
Closed Subscriber
Group (CGS)
Open Subscriber
Group (OGS)
Hybrid
Paging Selection
Automatic
Manual
Re-selection Handoff
Hand-In
Hand-Out
Femto-to-Femto
HO
11
Mobility Management in Wireless Broadband Femtocells
• Hierarchical Cell Structure (HCS): HCS can be used to distinguish between the
different network cells. Each tier can be assigned individual access priority (i.e.
HCS_0, HCS_1).
• Separate femtocell PLMN ID: This technique uses a different Public Land Mobile
Network Identity (PLMN ID) for femtocells [27]. PLMN ID is an identifier for the
operator’s networks, and each operator has its own PLMN [1]. Hence femtocells
are assigned a different PLMN ID from macro/micro-cells to ensure femtocells
selection and minimize the impact on macro/micro-cells users (i.e. PLMN ID_0,
PLMN ID-1) [1]. Hence, more PLMN IDs are required for operators.
• CSG PSC/PCI: A set of Primary Scrambling Codes (PSC) in UMTS or Physical
Cell Identifier (PCI)
3
in LTE is reserved for identifying CSG cells of a specific
frequency [27].
• Dedicated CSG frequency list: Specifies the frequencies dedicated for UMTS
CSG cells only [27].
• CSG Indicator: Another approach is to use a CSG Indicator to determine whether
a femtocell is a CSG or not [26], this is used in LTE networks.
• CSG ID [1]: One or more CSG cells are identified by a unique numeric identifier
called CSG identity (CSG ID). When the UE is not authorized to access the target
femtocell a new reject message is used. The UE will then block the corresponding
CSG ID, for a configurable duration, instead of the whole frequency (in LTE
system).
• Femtocell name [27]: A text based name for the femtocell sent only by CSG and
hybrid cell. UE may display the femtocell name.
3
Physical Cell Identifier (PCI) is an ID used to identify a cell for radio purpose.
Macro/micro-Tier
Femto-Tier Femtocell
Femtocell
Femtocell
Femtocell
Macro/micro
-
cell
Figure
4
: Two Tiers Network
12
Mobility Management in Wireless Broadband Femtocells
3.2.2 Finding neighbouring femtocells
UEs can distinguish femtocells from other cellular tiers, but it is not easy to decide on
joining a neighboring femtocell due to the large numbers of FBSs. The following are
techniques that have been proposed to find femtocells.
• NCL: The NCL can be created by the FBS through self-configuration algorithms
implemented by the vendor, and it might be updated when the FBS senses any
changes of the neighboring cells or when the FBS is turned on [1].
• UE autonomous search: The objective of the autonomous search is to determine
when the UE should start searching for a femtocell to which it can have access
and whether the CSG cell is valid or not [26]. Due the limited area coverage of
femtocells, the UE only starts searching for a femtocell when the UE is in its
vicinity [26]. The autonomous search function is not specified and is left to UE
manufactures.
• Manual CSG search: This method is specified in LTE and UMTS, where a user
can find a CSG cell. On request of the user (e.g. via UE application), the UE is
expected to search for available CSG IDs by scanning all frequency bands for
CSG cells, and then the UE reports the CSG ID of the strongest or higher priority
to higher tiers [26].
3.2.3 Distinguishing accessible (CSG and Hybrid) femtocell
By knowing whether a femtocell is accessible or not, unnecessary signalling overhead
can be avoided. The following CSG related identification parameters used to identify
accessible femtocell are:
• LAI/TAI [1]: For femtocell, the Location Area Identity (LAI)/Tracking Area Identity
(TAI) of neighbours needs to be different for the purpose of user access control.
The LAI of unauthorized femtocells will be put in the UE’s Universal Subscriber
Identity Model (USIM) after it receives Location Area Update (LAU) rejections from
these femtocells.
• CSG Indicator, CSG ID, and femtocell name are also used to identify accessible
femtocell. Whereas only CSG or Hybrid cells broadcast their CSG ID or femtocell
Name.
3.2.4 Handling allowed list
The allowed list is necessary in order to check whether the UE is allowed or not to
access the target femtocell. The allowed list could be in:
• Allowed list in FBS or Femto-GW: This is a list stored locally in the FBS or
Femto-GW that contains the UEs that are allowed access [1]. The operator or
owner can manage the allowed list. This technique is used in UMTS.
13
Mobility Management in Wireless Broadband Femtocells
• Allowed list in CN: In LTE, a UE’s Allowed CSG List (ACL) [26] is provided. ACL
is the list of CSG IDs (FBSs) that the UE belongs to [26]. The ACL is stored with
the user’s subscriber information in the CN, as well as it may keep a copy in UEs.
3.3 Access control
Access control mechanisms play a vital role in HO management, as well as when a user
tries to camp on a femtocell to prevent unauthorized use of that femtocell and in
mitigating cross-tier interference. Users of femtocells in two-tier networks are classified
into [20]:
• Subscribers of a femtocell are those that are registered have the right to use it.
Femtocell subscribers may be any UEs, for instance cell phone, laptop, etc.
• Nonsubscribers are users not registered in a femtocell
Location Access Control: There are many possible locations of access control in the
femtocell network. Location of access control may be [1, 26]:
• Access control in FBS or Femto-GW: Access control is performed in FBS or
Femto-GH for femtocell while for macro/micro-cell in CN. For example in LTE,
Femto-GW shall perform access control, and FBS may optionally perform access
control.
• Access control in UE: In LTE, UE can perform the basic access control in the
registration procedure to enhance the mobility procedure.
• Access control in CN: In WiMAX and LTE, one of the CN entities (such as
admission server) checks the access of the UE to the target femtocell by the ACL
after it receives information from the femtocell.
Access control types: Femtocells support flexible access control mechanisms (as
mentioned in page 9).
3.4 Paging
For OSG and hybrid femtocells, the CN pages the UE in all cells that the UE is registered
in [1]. For CSG femtocells, there may be many CSG femtocells that the UE is registered
in but the UE is not allowed to camp on. Perhaps, the paging procedure needs to be
optimized, in term of minimizing the amount of paging messages used to page a UE in
femtocells [27]. CN and/or Femto-GW can perform the paging optimization [1].
Management of paging by the Femto-GW is left to vendor implementation [27].
3.5 Idle mode, cell selection and cell reselection
It is desirable for the UE to switch to its femtocell when the received signal from the
macro/micro-cell is strong enough to support service, even when the macro/micro-cell
can still provide reliable service [28]. Hence, cell selection and reselection in femtocell
environments are more complicated than in macro/micro-cell networks [1]. There are few
14
Mobility Management in Wireless Broadband Femtocells
alternatives to enabling cell selection and reselection in femtocells depending on the
technology used.
Cell reselection
A UE in idle mode changes (reselects) the cell it is camped at as it moves across cells.
Reselection requires parameters broadcast by the cell sites [28]. To extend battery life of
the UE (with idle mode), the UE scans other radio channels only when the signal-to-noise
ratio (SNR or S/N) of the current cell is lower than a certain threshold. In UMTS and LTE
networks, this threshold is defined by parameters called S Intersearch and S Intrasearch
[28]. Cell reselection may also be as an autonomous search function which is intended to
find CSG/Hybrid femtocell to the UE to camp on it [27]. The autonomous search function
is not specified and is left to vendor implementations [27].
Cell selection
There are two modes for network selection, manual and automatic. Automatic cell
selection is similar to that for macro/micro-cells. An extra CSG ID check is performed
when the target cell is CSG or Hybrid, to check whether the CSG/Hybrid cell is suitable
for the UE or not [28]. This technique is used in WiMAX and LTE. On the other hand, in
manual cell selection, the UE is allowed to choose its serving CSG manually. Manual
selection is not allowed in connected mode though [26, 1]. This technique is used in
UMTS and LTE.
3.6 Connected mode and handoff
There are three scenarios for HO in femtocell networks: Hand-In, Hand-Out and Femto-
to-Femto HO, as shown in Figure 5.
Figure 5: HO Scenarios in Femtocell Networks
15
Mobility Management in Wireless Broadband Femtocells
1. Hand-In takes place when a UE moves from a macro/micro-cell to a femtocell. Hand-
In is a complex scenario since it requires the UE to select an appropriate FBS, while
considering neighbour cells. As well, there is no straightforward mechanism for the
macro/micro-BS to determine the identity of the target FBS from the measurement
reports sent by the UE.
2. Hand-Out takes place when a UE moves out of a femtocell to a macro/micro-cell.
Hand-Out process is supported in femtocells almost without any changes to the
existing macro/micro-cell network or to the UE. The handling neighbor cell is easier
than Hand-In case, since the target cell is always one.
3. Femto-to-Femto HO takes place when a UE moves out of a femtocell to another
femtocell, and requires handling long NCLs.
Access control in femtocell networks makes the HO procedures more complex than in
conventional networks, especially in Hand-In and Femto-to-Femto HO [1]. In addition, in
macro/micro-cell networks, HOs are triggered when users enter the coverage area of
other cells. However, given the coverage size of open/hybrid access femtocells, this
occurs more often than in the macro/micro-cell case. Hence, different HO management
procedures are needed to allow non-subscribers to camp for longer periods at near
femtocells. In 3G femtocell networks, there is no support for soft HO [1]. In LTE Release
8, there is no support for Hand-In and Femto-to-Femto HO. However, 3GPP Release 9
supports Hand-In. 3GPP Release 10 enables Femto-to-Femto HO [1, 28].
3.7 Mobility management issues
Mobility management in femtocells should offer a seamless experience for users as they
move in and out of femtocell coverage. Since, existing cellular networks and mobile
devices have been designed without awareness of femtocells; these requirements must
be met without requiring changes to existing infrastructure or to mobile devices. Dense
deployments will cause serious issues on mobility management between the
macro/micro-cells and femtocells. The importance of mobility management in femtocell
networks is due to the following reasons:
• Large number of FBSs that are usually overlaid with macro/micro-BS coverage
• High density of FBSs
• Dynamic neighbour cell lists
• Variant access control mechanisms
• Different user preferences
• Different operator policies and requirements
The above characteristics pose the following issues:
1. Neighbour Advertisement Lists and Messages
Since large numbers of FBSs may be within the range of a single macro/micro-cell, a
long list of neighbouring FBSs would be broadcast via neighbour advertisement
16
Mobility Management in Wireless Broadband Femtocells
messages. This leads to a waste of the wireless resources and makes the process of
scanning all neighbouring femtocells time consuming [29].
2. HOs
Current macro/micro-cells share the radio frequency with potentially large numbers of
femtocells. Hence, a UE may face up continual HOs, especially when it moves around
the home or enters areas where the received signal from the macro/micro-cell is greater
than that from the femtocell [30]. In addition, leakage of coverage to the outside of a
house may occur and can lead to highly increased number of unnecessary HO of
macro/micro-cell users, which may lead to higher call dropping probability. Femtocells
also introduce complexities in Hand-In and Femto-to-Femto HO [24, 29, 31, 32].
3. HO decision parameters
In femtocell environments, creative and flexible decision parameters will influence the HO
other than existing parameters, such as serving cost, user’s status and preferences, load
balancing, [42, 44] etc. In other words, the serving cell and/or UE should decide to
handoff to a target cell based on multiple parameters. Hence, there is a need for
algorithms to optimize and adapt these and other parameters.
4. Searching for FBS in different access scenarios
In order to manage the mobility procedures in both idle and connected modes, with the
case of hybrid and CSG scenarios, there are two problems to be solved. The first
problem is how the mobile devices will find out that the target is the CSG or not. The
second problem is how to identify target CSG cell as the mobile device’s own accessible
FBS among many of FBS [33, 26].
5. Idle mode mobility procedures
Additional energy consumption should be taken into account due to the dense
deployment of femtocells and their continuous receiving and transmission signals [34].
The increase of the number of cells may result in a large increase in traffic and load of
the CN for idle mode mobility procedures [34, 35].
17
Mobility Management in Wireless Broadband Femtocells
4 Related Work
Several research efforts have been done to modify and adapt the existing mobility
management procedures in cellular networks for femtocells. In this section, we present a
survey of mobility management schemes in femtocell networks. In Section 4.1, we
describe the mobility management techniques. In Section 4.2, we define a set of
evaluation criteria. This is followed by a thorough review of proposed schemes in
different categories in Section 4.3.
4.1 Mobility management techniques
Network-based mobility management: In this type of mobility management protocols,
the network takes responsibility for all aspects of mobility management without requiring
participation from the UE in any related mobility procedures and their signalling. This
domain does not require the UE to be involved in the exchange of signalling messages
between itself and the network.
Mobile-based mobility management: This type requires the participation of the UE in
all aspects of the mobility management. However, the participation of the UE in the
mobility management and associated resources and software has become a hurdle for
standards and protocols.
Mobile-Assisted mobility management: In this type, information and measurement
from the UE are used by the network to take care of all aspects of the mobility
management.
4.2 Evaluation criteria
In order to evaluate the schemes proposed, we define a set evaluation criteria to
compare the different mobility schemes in first and second categories. These are:
• Access control type(s): This criterion indicates the type(s) of access control
supported (CSG, OSG, and Hybrid).
• HO scenario: This criterion contains the HO scenario(s) supported (Hand-In/Out,
Femto-to-Femto HO).
• HO objective(s): This criterion shows the target goal(s). The goal could be to
reduce the number of HO or unnecessary HO, decrease signalling overhead, etc.
• Additional HO parameter(s): This criterion shows the additional HO parameter(s)
used, such as QoS, user’s state, etc.
• HO latency: This criterion relates to the HO latency in proposed mechanisms.
The shorter the better.
• Signalling traffic overhead: This criterion relates to control data load. The
amount of signalling packets should be within an acceptable range.
18
Mobility Management in Wireless Broadband Femtocells
• QoS support: A HO algorithm should be reliable and the call should have good
quality after handoff, as well as it should be fast so that the user does not
experience service degradation or interruption. This criterion shows the QoS types
used, such as real-time support, packet lost, HO delay, etc.
• Special support required: This criterion discusses change required to the
infrastructure and/or the UE. The fewer the required changes the better.
4.3 Proposed mobility management schemes
Existing mobility management schemes can be categorized based on the target problem
for each scheme, such as HO schemes, reselection and scanning.
4.3.1 Handoff schemes
Schemes belonging to this category are not complete mobility management schemes.
They provide new HO protocols and algorithms in femtocells, in addition to schemes that
deal with different aspect that are related to the HO processes, such as, HO optimization
process, new HO decision, context transfer stages, reading system information in Hand-
In, etc. In this section, many protocols and algorithms are discussed. Table 2 shows a
comparison of the schemes that surveyed below based on the evaluation criteria in
Section 4.3.
Zhang et al. [29] propose a modified signalling flow of HO in LTE network with CSG
scenario for Hand-In and Hand-Out. This method is applied in the Femto-GW. The HO
algorithm is based on the user’s speed and QoS. The proposed scheme integrates the
measurement value, maximal capacity and current load of the cell as the input of HO
judgment. This algorithm does not allow the high speed user HO’s (>30 Kph) from
Macrocell to femtocell, while low speed users are allowed. The algorithm distinguishes
between real time users and non-real time users with moderate speed (>15 Kph), it
allows the real time moderate speed users to handoff. However, it does not allow
moderate speed users HOs without real time. The proposed algorithm performs better in
reducing the unnecessary HOs and the numbers of HOs compared to the traditional HO
algorithms, especially in medium and high speed users with a small penalty of signalling
overhead.
Wang et al. [24] propose two mobility management schemes applied in the Femto-GW at
Radio Network Layer (RNL) for LTE Femto-to-Femto HO. In method I, the authors
propose the Femto-GW to serve as a mobility anchor, and let it make the HO decisions.
When Femto-GW receives a HO request from the source FBS, the Femto-GW checks
the target ID. If the target cell is under its control, it will handle the HO. In method II,
Femto-GW operates as a transparent node and simply forwards the HO messages
between the FBS and MME. S-GW has to be notified with the change of end point after
HO. Moreover, Method I is more suitable for enterprise use, because it is reduced the
19
Mobility Management in Wireless Broadband Femtocells
signalling traffic with the CN. On the other hand, method II is more suitable for home use,
because more signalling messages are exchanged.
Chowdhury et al. [36] propose a signalling flow for Hand-In and Hand-Out in UMTS
networks with Call Admission Control (CAC). In the proposed signalling flow, there are
two phases: HO preparation phase (information gathering about HO candidates and
authentications, HO decision to determine the best HO candidate), and HO execution
phase. The proposed method considers the interference level as additional HO decision
parameters for Hand-In procedure, and uses the proposed CAC to reduce unnecessary
HOs. Three parameters are considered for the proposed CAC: received signal, duration
of a UE maintains the minimum required signal level, and signal to interference level. The
results show that the number of unnecessary HOs is minimized due to the proposed
CAC.
As an extension work of [36], Chowdhury and Jang [37] propose a modified signalling
flow for Hand-In and Hand-Out for small and medium scale femtocell network
deployments. The authors present the details HO call flow for these two femtocell
deployments and the proposed CAC scheme to reduce the unnecessary HO. The
proposed queuing scheme optimizes the new call blocking probability, HO call blocking
probability, and bandwidth utilization. Simulation results show that the number of
unnecessary HOs is minimized due to the proposed CAC. As well, the proposed scheme
able to provide a seamless and reliable HO for both small and medium scale
deployments.
Kim and Lee [38] propose a signalling flow for Hand-In and Hand-Out in UMTS networks
with CAC in the hybrid access mode. The proposed Hand-In procedure makes a decision
based on the new CAC. The new CAC takes into consideration the residence time in a
cell, user types, RSS level, the duration a UE maintains the signal level above the
threshold level, the signal to interference level, and the capacity that one FBS can
support. If the received signal level from the femtocell is higher than the threshold, the
Femto-GW checks whether the UE is preregistered or not. If the UE is pre-registered, the
next handover procedure is performed. If the UE is not pre-registered, UE must remain in
the femtocell area for the threshold time interval during which a signal level is higher than
the threshold signal level before continuing to the next handover procedure. Results
show that the number of unnecessary HOs is reduced via the Hybrid access CAC.
Ulvan et al. in [31] and [39], propose an adapted signalling flow for the three types of HO
based LTE networks. The proposed scheme considers the movement prediction
mechanism as an additional parameter for HO decision. This HO is a client-based HO.
Reactive and Proactive HO (PHO and RHO) procedures are proposed to initiate the HO,
since the HO procedure may be initiated by FBS, macro/micro-cell, and UE. In RHO, the
HO is trigged when the UE almost lose its serving cell signal or the most probable
20
Mobility Management in Wireless Broadband Femtocells
position of UE is predicted. RHO aims to delay the HO as long as possible to prevent the
frequent and unnecessary HO, and reduce the generated overhead of HO. However, in
PHO, the HO may occur any time before the level of RSSI for current BS reaches the HO
threshold via estimate of a specific position before the UE reaches that position. After the
UE discovers the new target cell RSSI, the UE calculates the time remaining before the
normal HO is triggered, then the HO triggers before the HO threshold. RHO is expected
to minimize packet loss (PL) and latency in HO.
Ellouze et al. [40] propose a modified HO procedure between WiMAX macro-BSs and
FBSs. The proposed HO scheme takes QoS and load balancing into account in two
ways, first, by limiting the break time connection caused by the scanning interval.
Second, the HO selection procedure decides either to connect the user to FBS or other
BSs according to the user’s QoS profile. The proposed solution reduces the HO delay,
and balances the load over the network.
De Lima et al. [41] propose a stochastic association mechanism for Hand-In with OSG.
The proposed system introduces a new distributed HO procedure, which does not
depend on any centralized coordination. The proposed solution uses a modified multi-
stage Dutch auction to autonomously coordinate and prioritize bidding FBSs. As well, the
solution uses a stochastic process to separate candidate FBSs to reduce the probability
of collision. Macrocell user plays an ‘auctioneer’, while target FBS play ‘bidders’. A
stochastic election process is incorporated to the selection process to reduce the
chances of multiple bidders. Femtocell users faced some QoS degradation
Xu et al. [42] propose a user’s state and signal-to-interference ratio (SINR) based Hand-
In algorithm for 3G networks, to overcome the drawbacks of the large asymmetry
transmit power between FBSs and macro-BSs in two-tier networks. The proposed
algorithm uses SINR to avoid the asymmetry transmit power, and user’s state to reduce
the unnecessary HOs. The authors add user state such as velocity, QoS, with SINR as
HO decision parameter. On the one hand, the results show that the new algorithm cuts
down the number of unnecessary HOs due to taking the user’s state into account. On the
other hand, the total number of HOs is increased.
Shaohong et al. [43] propose two HO algorithms for 3G networks using the mathematical
concept of ‘set’. In Velocity and Signal HO algorithm (VSHO), they consider the velocity
and RSS of the UE, hence, the frequent HO of high speed UEs are avoided. In the other
improved algorithm called Unequal HO algorithm (UHO), the scheme considers the
difference between macro-BSs and FBSs to consider the issue that the UE receives a
higher signal strength from a FBS in a house than from the macrocell outside. In other
words, UHO sets a higher signal level limit for Femtocells than macrocell to serve as a
chief BS. The results of the two proposed algorithms show that HO probabilities
decreased for high speed users, and the total throughput of macrocell networks is
21
Mobility Management in Wireless Broadband Femtocells
increased. The above algorithms do not represent the power asymmetry by adding a
constant value to received signal from FBSs. In addition, the user state is not considered
as a factor for HO. Hence, the Hand-In process may initiate with no guarantee of QoS
from the femtocells, and then it may lead to HO failure.
Wu et al. [44] propose Hand-In and Hand-Out procedures for LTE femtocells. The
authors consider a group of parameters for the HO decision which are interference level,
RSS, user’s velocity, available bandwidth and QoS level. The Hand-In has two kinds of
procedures. The First is for CSG users where the UE shall chose the most appropriate
target FBS. The second is for non-CSG users, if a non-CSG UE causes too much
interference; it can handoff to FBS to minimize interference. This HO is different from the
normal situation, because the HO is triggered by FBS. The proposed solution does not
consider the co-tier interference.
Becvar and Mach [45] propose an adaptive hysteresis margin for HO for LTE networks.
The proposed solution utilizes the reported metrics (RSSI or CINR) for the dynamic
adaptation of an actual value of hysteresis margin according to the position of the user in
a cell. The hysteresis margin decreases with UE’s moving closer to the cell border. This
proposed solution shows reduction of redundant HO by mainly focusing on avoids Ping-
Pong effects. However, this is not an appropriate way to prevent unnecessary HOs
caused by femtocell visitor.
In another paper, Becvar and Mach [<becvar11>] propose an enhanced HO decision for
Hand-In procedure with OSG and hybrid scenarios. The proposed scheme takes into
account the delay and capacity of FBS’s backbone as an additional decision parameter
to achieve an acceptable level of QoS for femtocell users. Also, the authors consider the
time duration that spent by users in femtocell coverage as a decision parameter to
reduce the number of unnecessary and frequent HOs. Results show that signalling
overhead is increased, and the HO latency remains within the acceptable range.
Moon and Cho [46, 47] propose a modified HO decision algorithm for Hand-In procedure
in LTE networks based on RSS. They combine the value RSS from the serving macro-
BS and a target FBS to derive a reasonable HO criterion using the concept of
combination factor, and takes into consideration that the UE has an ability to detect
neighbouring femtocells. Results show that there is an enhancement of the assignment
probability to the femtocell while keeping the same level of the number of HOs. In the
problem of the asymmetry transmit power for the previous two proposed algorithms, the
user’s state is not considered as a factor for Hos; hence, the HOs may initiate with no
guarantee of QoS from the femtocells, and then it may lead to HO failure.
Jeong et al. [48] propose a smart Hand-In procedure to reduce the number of
unnecessary HO of temporary visitors that stay a short time in the femtocell. They keep
22
Mobility Management in Wireless Broadband Femtocells
UEs connected to a macrocell rather than conducting a Hand-In when staying a short
time in a femtocell, based on next movement pattern analysis. The proposed scheme
applies a locations prediction algorithm to identify temporary femtocell visitors while the
users move along random movement patterns. Results show that the proposed scheme
reduces the number of unnecessary Hos.
In order to mitigate the interference problem that may take place when a non CSG user
comes to a femtocell, Fan and Sun in [32] propose a method for access and HO
management for OFDAM femtocell networks. In CSG scenario, when a UE comes near a
FBS, its serving BS will check the UE’s ID, if within the allowed list for the target FBS, the
BS informs the FBS to start the HO procedure. Otherwise, the BS should notify the FBS
to start the proposed proactive interference management procedure. As well, the authors
propose a hybrid access to the same situation. After a non CSG UE enters a femtocell, a
FBS measures the UE’s signal strength and decides whether the potential interference
caused by the UE is above the interference threshold or not. If so, the FBS will request a
HO procedure from the serving BS for the UE, and informs that this is an avoid
interference HO. The CSG scenario reduces the unnecessary HOs and signalling load.
However in the Hybrid scenario, the number of HOs is increased.
To solve the same problem presented in [32], Li et al. in [49] propose a pseudo HO
based on the scheduling information exchange method, subchannel and power
adaptation to avoid collision interference in LTE/-A networks. The pseudo HO is executed
in the Radio Access Network (RAN), not referring to MME, which significantly reduced
signalling overhead. When the UE tries to camp on a CSG FBS, if the UE belongs to this
femtocell, the regular HO is triggered; or else, the pseudo HO is triggered. The FBS will
set up and maintain a table that contains the ID of non-CSG called pseudo-HO users.
Discussion
Although there are several proposed solutions for HO in femtocells, most of the solutions
have targeted only one or two parts of the HO procedure, such as HO preparation, HO
decision parameter, HO signalling, Hand-In algorithm. A few solutions have proposed a
comprehensive HO procedure. Table 2 provides a detailed comparison between the
schemes. Zhang [40], Wang [34], Chowdhury [47], Chowdhury & Jang [48], Kim [49],
Ulvan [42, 50], and Wu [55] provide schemes for the signalling flow of the HO process
with different additional parameters to reduce the number of unnecessary HOs. For
example, Wang [34] supports CSG and OSG scenarios in the Femto-to-Femto HO. The
scheme uses the user speed, QoS, and load balancing as additional parameters for the
HO decision. The HO latency and signalling overhead are increased due to additional
gateway (Femto-GW) that is installed.
Ellouze [51], de Lima [52], Xu [53], Shaohong [54], Becvar [56], and Moon [57, 58]
provide HO algorithms and decision parameters with more details to enhance the HO
23
Mobility Management in Wireless Broadband Femtocells
process and reduce the number of unnecessary HOs. In [52], the scheme just targets the
OSG scenario to enhance the Hand-In process. The HO process triggers by using the
CINR as an additional parameter for the HO. The proposed scheme enhances the
selection process of a target FBS to Hand-In, where there is an obvious degradation of
the QoS. Reference [51] provides Hand-In/Out algorithm under the OSG scenario. This
scheme is a complete solution due to inclusion of scanning and selecting procedures
which are the two main processes before the HO process, in addition to the proposed HO
procedures.
24
Mobility Management in Wireless Broadband Femtocells
Access
Control
Type(s)
HO
Scenario
(s)
HO Objective(s) Additional HO
Parameter(s)
HO
Latency
Signalling
Overhead
QoS
Support
Special
support
required
Zhang [29]
CSG In, Out Reduce no. of
unnecessary HO
User’s
velocity, QoS,
load balancing
High High Real-time Femto-GW
Wang [24]
Method I
All F-F Enhance HO N/A Medium
Low
No Femto-GW
Wang [24]
Method II
All F-F Enhance HO N/A High High No Femto-GW
Chowdhury
[36]
All In, Out Reduce no. of
unnecessary HO
CAC,
interference
level, time
duration
4
High Medium No Femto-GW
Chowdhury
& Jang [37]
All In, Out Reduce no. of
unnecessary HO
CAC,
interference
level, time
duration
Low Medium
Call/HO
blocking
probability
Small Depl.:
new server
Medium Depl.:
Femto-GW
Kim & Lee
[38]
Hybrid In, Out Reduce no. of
unnecessary HO
Hybrid access
CAC and other High Medium No Femto-GW
Ulvan [31,
39]
OSG All Prevent frequent &
unnecessary HOs
Movement
prediction
Medium
in RHO,
low in
PHO
Low in
RHO
Minimized
PL in PHO Femto-GW
Ellouze [40]
OSG In, Out
Achieve load
balancing &
enhance QoS
Load
balancing,
QoS
Low Low Yes No
De Lima
[41]
OSG In Enhance the HO CINR High Medium Real-time No
Xu [42]
OSG In
Solve the
asymmetry transmit
power in Hand-In
User’s state,
SINR Medium High Minimized
PL No
Shaohong
[43]
OSG In
Cut down the no. of
unnecessary HO &
increase throughput
User velocity High High in
UHO No No
Wu [44]
Hybrid In, Out
Reduce no. of
unnecessary &
failure HOs
Velocity,
bandwidth,
QoS,
Interference
High High Real-time Femto-GW
Becvar &
Mach [45]
OSG,
Hybrid All Reduce of
redundant HO CINR High N/A N/A No
Becvar &
Mach [50]
OSG,
Hybrid In Eliminate redundant
HOs
FBS
backbone,
time spent in a
FBS
Medium
High Yes
Modifications
in backbone
management
Moon & Cho
[46, 47]
OSG In
Enhance
assignment
probability
Combine RSS
from serving &
target BSs
High
HO
failure
High N/A No
Jeong [48]
OSG In Reduce no. of
unnecessary HO
Mobility
pattern
prediction
N/A Medium Yes New server
Fan [32]
CSG,
Hybrid All
Interference
mitigation & reduce
no. of HOs
Interference
threshold N/A Low No Femto-GW
Li [49]
CSG,
Hybrid All Avoid collision
interference
Interference
threshold N/A Low No No
Table 2: Comparison of HO Schemes
4
Duration of a UE maintains the minimum required signal level.
F-F: Femto-to-Femto, In: Hand-In, Out: Hand-Out
25
Mobility Management in Wireless Broadband Femtocells
Summary
A complete HO solution should take into account a number of factors in order to be a
comprehensive and reliable. For instance, an additional HO parameter must be
considered for the HO decision and triggering process in the femtocells, viz. the user
state (e.g. velocity, preference) in Shaohong [43] and Zhang [29]. Also, the operator’s
preference, bandwidth and load balancing should take place in any HO process as in
Ellouze [40] and Wu [44]. The HO latency and signalling overhead must be minimized as
in Ellouze [40].
4.3.2 Scanning and Selection Schemes
As aforementioned, scanning is the process used to find a cell for the UE to camp on.
Schemes in this category propose mechanisms and algorithms to improve the scanning
and selection processes.
Nam et al. [51] propose a network-assisted FBS management scheme in Mobile WiMAX
networks using a triangulation mechanism and FBS monitoring scheme to reduce the
number of scanning operations, as well as the size of neighbour advertisement
messages. The proposed scheme uses a FBS monitoring mechanism to provide UE with
the nearest FBS information under the OSG scenario. The authors assume that every
FBS has two interfaces. One is used to communicate with the attached UE, while the
other is used to monitor signals of candidate UE. Results show that the proposed
scheme improves scanning performance and to reduces wasting air resources.
Han et al. [52] propose an automatic generation scheme of neighbouring BS lists for
femtocell networks under the OSG scenario. This scheme is utilized by a UE to specify a
target FBS to handoff. A neighbour list automatically generates by jointly utilizing the
measurement of multiple neighbour BSs to include the all the neighbouring BSs for a
proper HO. The proposed scheme operates in three steps. First, a BS measures the RSS
of the neighbouring BSs locally and reports the measurement results. Then the BS
requests the other BSs’ measurements. Second, a BS reconstructs the topology of the
identified neighbours, and then the topology is used to find hidden neighbouring FBSs.
Third, a BS discovers hidden neighbouring BSs with the support of other identified
neighbours. These three steps can repeat periodically or be triggered whenever the
network topology changes. The scheme shows acceptable results regardless the
shadowing effects.
Jung et al. [53] propose a scanning scheme to reduce unnecessary scanning procedures
for an accessible FBS (CSG) to reduce the power consumption. The scheme uses an
adaptive threshold with a margin for RSSI. Using thresholds within a serving macrocell,
the target region is separated into smaller regions; hence the UE only scans for the FBS
26
Mobility Management in Wireless Broadband Femtocells
within a small region satisfying triggering conditions. Results show that the proposed
scheme can reduce scanning time and power consumption.
Chowdhury et al. [54] propose an optimized NCL for Femto-to-Femto HO and Hand-In in
dense femtocell networks. The proposed algorithm considers the received signal level
from FBSs; open or closed scenarios; detected frequencies from the serving FBS and the
neighbour FBSs, and location information (using SON capabilities of the FBSs) for the
optimal neighbour FBS. The authors try to reduce power consumption for scanning many
FBSs and the Media Access Control (MAC) overhead. Femtocells are categorized in two
categories. First category contains the FBSs from which the received signals are greater
than or equals to a threshold level. The FBSs that are in the second category from which
the received signals are less than a threshold level or the serving FBS and the neighbour
FBSs use the same frequency. The results show that the proposed scheme is able to
maintain a minimum number (not optimal) of neighbour FBS list for the Femto-to-Femto
HO.
Kwon and Cho [64] propose a load based cell selection algorithm for faulted HO. The
proposed cell selection algorithm allows each UE to select a target femtocell based on
the basis of information of other UEs that have previously made their selection. The
proposed scheme minimizes HO blocking probability and achieves load balancing
between neighbouring cells when the FBS generates a fault. In contrast, the scheme
increases signalling overhead.
Access Control Scheme Objective(s)
Nam
[
51
]
OSG Reduce scanning overhead
Han
[
52
]
OSG Generate automatic neighbouring list
Jung
[
53
]
CSG Minimize unnecessary scanning
Chowdhury
[
54
]
OSG, CSG
Optimize the NCL
Kwon
[64]
OSG
Balance load among cells; minimize HO blocking probability
Table 3: Comparison of Scanning Schemes
4.3.3 Reselection schemes
Reselection schemes are used to allow the UE to change its serving cell to another,
without having an active session.
In [55], the authors target the issue of when a UE searches for a femtocell. The UE
needs to scan the entire femtocell spectrum in order to switch from macrocell to
femtocell. The authors propose a cache scheme for femtocell reselection. The proposed
scheme considers the UE’s movement history by storing the cell information of the
recently visited FBSs. The aim is to obtain the most recently visited order of FBSs that
27
Mobility Management in Wireless Broadband Femtocells
have been stored in the cache. The scheme seems useful in the OSG scenario with a
large number of FBSs; otherwise it is inefficient.
The authors in [34] propose two dynamic idle mode procedures for femtocells. The first
procedure activates the FBS only to serve active calls from its registered users. This is
achieved by a low power “sniffer” capability in the FBS that allows the detections of an
active call from a UE to the underlying macro/micro-cell based on a measured rise in
noise that activates the FBS upon request. The second procedure reduces the pilot
power and adjusts the cell’s reselection thresholds when it is not serving an active call.
The authors in [56] address the issue of discovering a 3G femtocell in multiple
frequencies, and the impact of this issue on the UE battery life, as well as the effect of
different cell reselection parameters on capacity offloads. For instance, with good
macrocell coverage, a UE may never initiate searches and would remain on the
macrocell even when it is in the vicinity of its own FBS. The authors improve the idle
mobility procedure to enable UEs to discover and camp on FBSs via three potential
techniques: search threshold optimization, beacon-based approach for enhanced
reselection procedure, and UE enhancement for enhanced the idle mode to the UE.
28
Mobility Management in Wireless Broadband Femtocells
5 Conclusion and Open Issues
This chapter presents a comprehensive study of mobility management in femtocell
networks. Issues in Handoff and other mobility management procedures in femtocells are
identified. Several research efforts are presented and classified.
Building efficient mobility mechanisms will play a vital role for successful deployment of
femtocells and for providing seamless services. We highlight some open problems and
issues that can be derived from our study of the research efforts discussed in this chapter
as follows.
• The HO decision parameters must be adaptive and flexible based on the situation.
For example, the HO decision parameter for the Hand-In in existing schemes is
typically based on the user preferences, CINR or RSSI. There are cases where
other factors must be considered as well. For instance, a UE may receive a
stronger RSSI from a macrocell while it is under the coverage of a femtocell,
hence, the UE will not handoff to the femtocell. This is due to the large asymmetry
in transmitting power. Therefore, techniques to optimize and adapt multiple HO
parameters depending on the situations need to be addressed.
• The service interruption time caused by reading system information of a target
CSG FBS during Hand-In and Femto-to-Femto HO should be minimized. This is
because UEs cannot receive data while reading system information. Existing
schemes do not address such issue in a satisfactory manner.
• Mobile-based or mobile-assisted HOs will play a key role in femtocell networks to
meet user needs. Some research work has been proposed on proactive HO in
femtocells networks. These solutions focus on satisfying user needs only and
ignore other networking requirements leading to unexpected results (e.g. whether
they are also interested in service quality or cost, etc.).
• Mobile femtocells on transit systems (e.g., buses and trains) will become more
prominent in the future. These will typically be used to aggregate user traffic and
relay it to the macrocell networks or to other access networks. Means of offloading
this aggregated traffic of these fast moving femtocells are needed. Hence,
modified protocols for HO and location management for mobile femtocells need to
be developed.
• Methods and techniques that assist in managing and updating the network
topology are essential for effective mobility procedures between macro/micro-cells
and femtocells and among femtocells. All macro/micro-BSs and FBSs have to be
aware if a FBS enters or leaves their coverage, hence changing the mobility
conditions. For UEs to perform Handoff and cell searching in a more efficient way,
the UE and/or FBS should acquire network topology.
29
Mobility Management in Wireless Broadband Femtocells
References
[1] Z. Jie and D. Guillaume, Femtocells: technologies and deployment. UK: WILEY, 2010.
[2] C. Patel, M. Yavuz, and S. Nanda, “Femtocells [industry perspectives],” Wireless
Communications, IEEE, vol. 17, pp. 6 –7, Oct. 2010.
[3] D. Knisely and F. Favichia, “Standardization of femtocells in 3GPP2,” Communications
Magazine, IEEE, vol. 47, pp. 76 –82, Sept. 2009.
[4] I. Guvenc, S. Saunders, O. Oyman, H. Claussen, and A. Gatherer, “Femtocell networks,”
EURASIP Journal on Wireless Communications and Networking, 2010.
[5] S. Yeh, S. Talwar, S. Lee, and H. Kim, “WiMAX femtocells: A perspective on network
architecture, capacity, and coverage,” Communications Magazine, IEEE, vol. 46, pp. 58 –
65, Oct. 2008.
[6] S. Ortiz, “The wireless industry begins to embrace femtocells,” Computer, vol. 41, pp. 14 –
17, July 2008.
[7] Smallcell Forum, http://www.smallcellforum.org/
[8] F. Chiussi, D. Logothetis, I. Widjaja, and D. Kataria, “Femtocells [guest editorial],”
Communications Magazine, IEEE, vol. 48, pp. 24 –25, Jan. 2010.
[9] M. Neruda, J. Vrana, and R. Bestak, “Femtocells in 3G mobile networks,” in Systems,
Signals and Image Processing (IWSSIP), 16th International Conference on, 2010.
[10] J. Boccuzzi and M. Ruggiero, Femtocells: design & application. US: McGraw-Hill, 2011.
[11] G. Korinthios, E. Theodoropoulou, N. Marouda, I. Mesogiti, E. Nikolitsa, and
G. Lyberopoulos, “Early experiences and lessons learned from femtocells,”
Communications Magazine, IEEE, vol. 47, pp. 124 –130, Sept. 2009.
[12] D. Knisely, T. Yoshizawa, and F. Favichia, “Standardization of femtocells in 3GPP,”
Communications Magazine, IEEE, vol. 47, pp. 68 –75, Sept. 2009.
[13] K. Elleithy and V. Ra, “Femto cells: Current status and future directions,” International
Journal of Next-Generation Networks (IJNGN), vol. 3, Mar. 2011.
[14] “WiMAX femtocells.” White Paper, ARICENT, 2010.
[15] E. Seidel and E. Saad, “LTE Home Node Bs and its enhancement in release 9.” White
Paper, nomor research, May 2010.
30
Mobility Management in Wireless Broadband Femtocells
[16] S. Choi, T. Lee, M. Chung, and H. Choo, “Adaptive coverage adjustment for femtocell
management in a residential scenario,” in Management Enabling the Future Internet for
Changing Business and New Computing Services, vol. 5787 of Lecture Notes in Computer
Science, pp. 221–230, Springer Berlin / Heidelberg, 2009.
[17] L. Mohjazi, M. Al-Qutayri, H. Barada, K. Poon, and R. Shubair, “Deployment challenges of
femtocells in future indoor wireless networks,” in GCC Conference and Exhibition (GCC),
IEEE, pp. 405 –408, Feb. 2011.
[18] F. Hasan, H. Siddique, and S. Chakraborty, “Femtocell versus WiFi - a survey and
comparison of architecture and performance,” in Wireless Communication, Vehicular
Technology, Information Theory and Aerospace Electronic Systems Technology (Wireless
VITAE),1st International Conference on, pp. 916 –920, May 2009.
[19] H. Claussen, L. Ho, and L. Samuel, “An overview of the femtocell concept,” Bell Labs
Technical Journal, vol. 13, pp. 221–245, Mar. 2008.
[20] G. de la Roche, A. Valcarce, D. Lopez-Perez, and J. Zhang, “Access control mechanisms
for femtocells,” Communications Magazine, IEEE, vol. 48, pp. 33 –39, Jan. 2010.
[21] V. Chandrasekhar, J. Andrews, and A. Gatherer, “Femtocell networks: A survey,”
Communications Magazine, IEEE, vol. 46, pp. 59 –67, Sept. 2008.
[22] R. Ellouze, M. Gueroui, and A. Alimi, “Femtocells QoS management with user priority in
Mobile WiMAX,” in Wireless Communications and Networking Conference (WCNC), IEEE,
pp. 84 –89, Mar. 2011.
[23] R. Kim, J. K., and K. Etemad, “WiMAX femtocell: Requirements, challenges, and solutions,”
Communications Magazine, IEEE, vol. 47, pp. 84 –91, Sept. 2009.
[24] L. Wang, Y. Zhang, and Z. Wei, “Mobility management schemes at radio network layer for
LTE femtocells,” in Vehicular Technology Conference (VTC Spring), IEEE 69th, pp. 1 –5,
Apr. 2009.
[25] “Femtocell access point fixed-mobile convergence for residential, smb, and enterprise
markets.” Research Report, ABI Research, 2010.
[26] A. Golaup, M. Mustapha, and B. Patanapongpibul, “Femtocell access control strategy in
UMTS and LTE,” Communications Magazine, IEEE, vol. 47, pp. 117 –123, Sept. 2009.
[27] G. Horn, “3GPP femtocells: Architecture and protocols.” Report, QUALCOMM Incorporated,
Sept. 2010. Available online.
[28] P. Humblet and A. Richardson, “Femtocell radio technology.” White Paper, Airvana, May
2010. Available online.
31
Mobility Management in Wireless Broadband Femtocells
[29] H. Zhang, X. Wen, B. Wang, W. Zheng, and Y. Sun, “A novel handover mechanism
between femtocell and macrocell for LTE based networks,” in Communication Software and
Networks (ICCSN), Second International Conference on, pp. 228 –231, Feb. 2010.
[30] “Femtocells - the gateway to the home.” White Paper, Motorola, May 2008. Available
online.
[31] A. Ulvan, R. Bestak, and M. Ulvan, “The study of handover procedure in LTE-based
femtocell network,” in Wireless and Mobile Networking Conference (WMNC), Third Joint
IFIP.
[32] Z. Fan and Y. Sun, “Access and handover management for femtocell systems,” in Vehicular
Technology Conference (VTC Spring), IEEE 71st, pp. 1 –5, May 2010.
[33] H. Kwak, P. Lee, Y. Kim, N. Saxena, and J. Shin, “Mobility management survey for Home-
eNB based 3GPP LTE systems,” Journal of Information Processing Systems, vol. 4,
pp. 145 –152, Dec. 2008.
[34] H. Claussen, I. Ashraf, and L. Ho, “Dynamic idle mode procedures for femtocells,” Bell Labs
Technical Journal, vol. 15, no. 2, pp. 95–116, 2010.
[35] Y. Li, G. Su, P. Hui, D. Jin, L. Su, and L. Zeng, “Multiple mobile data offloading through
delay tolerant networks,” in Proceedings of the 6th ACM workshop on Challenged networks,
CHANTS, (New York, NY, USA), pp. 43–48, 2011.
[36] Z. Chowdhury, W. Ryu, E. Rhee, and Y. Jang, “Handover between macrocell and femtocell
for UMTS based networks,” in Advanced Communication Technology (ICACT), 11th
International Conference on, Feb. 2009.
[37] M. Chowdhury and Y. Jang, “Handover control for WCDMA femtocell networks,” The
Journal of Korea Information and Communication Society, vol. abs/1009.3779, Sept. 2010.
[38] J. Kim and T. Lee, “Handover in UMTS networks with hybrid access femtocells,” in
Advanced Communication Technology (ICACT0, the 12th International Conference on.
[39] A. Ulvan, R. Bestak, and M. Ulvan, “Handover scenario and procedure in LTE-based
femtocell networks,” in The Fourth International Conference on Mobile Ubiquitous
Computing, Systems, Services and Technologies (UBICOMM), 2010.
[40] R. Ellouze, M. Gueroui, and A. Alimi, “Macro-femto cell handover with enhanced QoS in
Mobile WiMAX,” in Wireless Telecommunications Symposium (WTS), pp. 1 –6, Apr. 2011.
[41] C. de Lima, K. Ghaboosi, M. Bennis, A. MacKenzie, and M. Latva-aho, “A stochastic
association mechanism for macro-to-femtocell handover,” in Signals, Systems and
Computers (ASILOMAR), Conference Record of the Forty Fourth Asilomar Conference on,
pp. 1570 –1574, Nov. 2010.
32
Mobility Management in Wireless Broadband Femtocells
[42] P. Xu, X. Fang, J. Yang, and Y. Cui, “A user’s state and SINR -based handoff algorithm in
hierarchical cell networks,” in Wireless Communications Networking and Mobile Computing
(WiCOM), 6th International Conference on, pp. 1 –4, Sept. 2010.
[43] W. Shaohong, Z. Xin, Z. Ruiming, Y. Zhiwei, F. Yinglong, and Y. Dacheng, “Handover study
concerning mobility in the two-hierarchy network,” in Vehicular Technology Conference
(VTC Spring), IEEE 69th, pp. 1 –5, Apr. 2009.
[44] S. Wu, “A new handover strategy between femtocell and macrocell for LTE-based network,”
in Ubi-Media Computing (U-Media), 4th International Conference on, pp. 203 –208, July
2011.
[45] Z. Becvar and P. Mach, “Adaptive hysteresis margin for handover in femtocell networks,” in
Wireless and Mobile Communications (ICWMC), 2010 6th International Conference on,
pp. 256 –261, Sept. 2010.
[46] J. Moon and D. Cho, “Novel handoff decision algorithm in hierarchical macro/femto-cell
networks,” in Wireless Communications and Networking Conference (WCNC), IEEE, pp. 1
–6, Apr. 2010.
[47] J. Moon and D. Cho, “Efficient handoff algorithm for inbound mobility in hierarchical
macro/femto cell networks,” Communications Letters, IEEE, vol. 13, pp. 755 –757, Oct.
2009.
[48] B. Jeong, S. Shin, I. Jang, N. Sung, and H.Yoon, “A smart handover decision algorithm
using location prediction for hierarchical macro/femto-cell networks,” in Vehicular
Technology Conference (VTC Fall), IEEE, pp. 1 –5, Sept. 2011.
[49] H. Li, X. Xu, D. Hu, X. Chen, X. Tao, and P. Zhang, “Pseudo-handover based power and
subchannel adaptation for two-tier femtocell networks,” in Wireless Communications and
Networking Conference (WCNC), IEEE, pp. 980 –985, Mar. 2011.
[50] Z. Becvar and P. Mach, “On enhancement of handover decision in femtocells,” in Wireless
Days (WD), IFIP, pp. 1 –3, Oct. 2011.
[51] J. Nam, W. Seo, D. Kum, J. Choi, and Y. Cho, “A Network-Assisted Femto Base Station
Management Scheme in IEEE 802.16e System,” in Consumer Communications and
Networking Conference (CCNC), 7th IEEE, pp. 1 –2, Jan. 2010.
[52] K. Han, S. Woo, D. Kang, and S. Choi, “Automatic neighboring bs list generation scheme
for femtocell network,” in Ubiquitous and Future Networks (ICUFN), Second International
Conference on, pp. 251 –255, June 2010.
[53] B. Jung, J. Moon, and D. Cho, “Scanning time reduction based on adaptive threshold in
hierarchical cellular networks,” in Personal, Indoor and Mobile Radio Communications,
IEEE 20th International Symposium on, pp. 2861 –2865, Sept. 2009.
33
Mobility Management in Wireless Broadband Femtocells
[54] M. Chowdhury, M. Bui, and Y. Jang, “Neighbor cell list optimization for femtocell-to-
femtocell handover in dense femtocellular networks,” in Ubiquitous and Future Networks
(ICUFN), Third International Conference on, pp. 241 –245, June 2011.
[55] K. Lee, J. Lee, Y. Yi, I. Rhee, and S. Chong, “Mobile data offloading: How much can WiFi
deliver?,” in Proceedings of the 6th International COnference (Co-NEXT), Nov. 2010.
[56] F. Meshkati, Y. Jiang, L. Grokop, S. Nagaraja, M. Yavuz, and S. Nanda, “Mobility and
capacity offload for 3G UMTS femtocells,” in Global Telecommunications Conference
(GLOBECOM), IEEE, pp. 1 –7, Dec. 2009.
