Conference PaperPDF Available

Handoff procedure for heterogeneous wireless networks

Authors:

Abstract and Figures

Handoff is an important issue when considering mobility in heterogeneous telecommunication networks. During the past ten years, telecommunication networks have emerged as a central strategic component in various fields. Today, added flexibility of the heterogeneous wireless networks in the form of robust inter-technology mobility management schemes and sophisticated algorithms is becoming more important. Also data applications and IP protocols have become more important players when designing future architechtures. This paper gives a novel approach to procedures, algorithms and metrics involved in handoff in heterogeneous wireless networks. A network layer Mobile IP based handoff procedure between WLAN and GPRS is presented as a case study example.
Content may be subject to copyright.
Handoff Procedure for Heterogeneous Wireless Networks
M. Ylianttila, R. Pichna
1
, J. Vallström
1
, J. Mäkelä,
Centre for Wireless Communications
Department of Electrical Engineering
University of Oulu
Finland
A. Zahedi2, P. Krishnamurthy, and K. Pahlavan
Center for Wireless Information Network Studies
Department of Electrical and Computer Engineering
Worcester Polytechnic Institute
Worcester, MA 01609 USA
1 Currently with Nokia Ltd.
2 Currently with 3-COM
ABSTRACT
Handoff is an important issue when considering mobility
in heterogeneous telecommunication networks. During
the past ten years, telecommunication networks have
emerged as a central strategic component in various
fields. Today, added flexibility of the heterogeneous
wireless networks in the form of robust inter-technology
mobility management schemes and sophisticated
algorithms is becoming more important. Also data
application and IP protocols have become more
important players when designing future architechtures.
This paper gives a novel approach to procedures,
algorithms and metrics involved in handoff in
heterogeneous wireless networks. A network layer Mobile
IP based handoff procedure between WLAN and GPRS is
presented as a case study example.
1. INTRODUCTION
During the past ten years, telecommunication networks
have emerged as a central strategic component in various
fields. They are needed to form the worldwide
infrastructure needed to support educational purposes,
economic development, scientific research, and social
interaction between people in various fields. The
proliferation of high-speed local data networking and new
multimedia services has driven the broadband
telecommunication networks to the focus of research,
development, and standard activities worldwide. One
accelaring factor in the IT markets is and will be the
concept of mobility [1]. Mobility can be defined as a
possibility for a nomadic user to use his network
resources freely in any place and in any time. He can
access remote databases and mail-boxes by using
lightweight GUI devices, e.g., laptops or palm pilot,
anywhere and any time. Incorporating mobility into
broadband systems requires many considerations in every
layer of the communication: power control in the physical
layer, traffic management in the data link layer, mobility
management in the network layer and communication
optimizations in the transport and application layer, just to
mention few examples. Combining the servives of
telecommunication and data networks brings value
addition for both of them by increasing the usability and
scalability. Both operators and software industry are
looking for solutions which can be implemented with
minimal changes to existing infrastructures. Despite of the
hard work of standardization organizations and
committees, there are many competing standards in the
area of wireless communications. This means, that in
order the nomadic user can take a full advantage of the
available network services, he needs to roam between
different technologies.
Figure 1: Inter-technology mobility between two wireless
technologies (GSM and WLAN).
There are several possible levels to solve this problem:
above TCP/IP level, transport level (TCP), network level
(IP) and underlying protocol level [4]. A special inter-
technology roaming protocol layer can be inserted
between the application and the transport protocol layer.
Representatives of this approach are the project On The
Move [5], and Mobile TCP/IP[6]. There is another
category of approaches that offers modifications of
existing protocols sitting above the TCP/IP stack. One of
them is the work on the X-interface mobility [7]. Mobility
gateway can be used to provide mobility in the transport
MH
AP
Subnet 130.231.25
130.231.25.10 (eth0, WaveLAN )
130.231.25.14
130.231.25.20
CH
130.231.25.16
RT
130.231.25.15
HA
194.142.3.x
(ppp0, GSM Data)
2,4GHz
890-960 MHz
BSC
MSC
BTS
PSTN
PM
10631061
194.197.68.15
Internet
FW
Subnet 194.142.3
level. Good examples of this approach are Indirect TCP
[8] and MSOCKS [9]. Solutions to inter-technology
handoff can be found also in layers below IP [4]. In the
network layer most widespread approach is the IETF
Mobile IP. Network layer implementations have been
done in projects like Monarch [10], MosquitoNet [11] and
Daedalus/Barwan [12]). Here we focus on network layer
handoff strategies, and we show how Fuzzy Logic and
Neural Networks class of algorithms can be used to
control Inter-technology handoff procedure. WLAN and
GPRS are used as a case study example. Section 2
presents different scenarios for interconnecting the
services of WLAN and GPRS. In section 3, a general
level handoff procedure for inter-tech roaming is
presented. In section 4 potential metrics for handoff
decision are discussed, and summary and conclusions are
presented in section 5.
2. INTER-TECH MOBILITY SCENARIOS
As presented in [12], there can be seen many different
scenarios to implement inter-tech roaming. Figure 2
shows five different scenarios that enable the
implementation of inter-technology (inter-tech) roaming
between WLAN and GPRS.
Internet
GGSN
SGSN
GSM BSS LAN
AP VAP
3
1
2
4
MG 5
Figure 2: Overview of different scenarios for Inter-
technology roaming between GPRS and WLAN.
The first two scenarios consider implementation of the
WLAN as a base station in the GPRS network. In these
architectures GPRS is a master network and WLAN plays
the slave network. The third scenario considers
exploiting the Mobile-IP for implementation of inter-tech
roaming. The fourth scenario considers implementation
of the GPRS network as an access point in the WLAN.
The fifth scenario considers using a mobility gateway to
interconnect the two networks. In the fourth scenario
WLAN is the master network and GPRS is a slave for the
inter-technology roaming. In the third and the fifth
scenario GPRS and WLAN are interconnected as peers in
a larger network. It is preferable to reduce, as far as
possible, major changes to existing networks and
technologies especially at the lower layers such as MAC
and physical layers. This will ensure that existing
networks will continue to function as before. Scenarios 3
and 5 were preferred, since they treat GPRS and WLAN
as peer networks, without any modification to GPRS or
WLAN specific protocols.
Application
TCP
IP
SNDCP
LLC
RLC
MAC
PHY
RLC
MAC
PHY
LLC
BSSGP
Fr. Rel
L1 bis L1 bis
Fr. Rel
BSSGP
LLC
SNDCP
IP
GTP
UDP
IP
BB L2
BB L1
IP
GTP
UDP
IP
BB L2
BB L1 PDN L1
PDN
L2
802 LLC
802.11 MAC
802.11 PHY
802 LLC
802.11 MAC 802.3 MAC
802.11 PHY 802.3 PHY 802.3 PHY
802.3 MAC 802.3 MAC
802.3 PHY
802 LLC 802 LLC
IP
IPIP
802 LLC
802.3 MAC
802.3 PHY
PDN
L2
PDN L1
Mobile
Base Station
Access Point
SGSN GGSN
on the LAN
Home or
Foreign
Agent
Figure 3: Interconnecting WLAN and GPRS by using
Mobile IP (Scenario 3).
The service scenario was that when a wideband local area
service is available, the terminal will use it. When it is
not available the terminal roams to a lower speed
underlay wireless data service. In our test case the user
has an application running on the terminal and moves out
of the range of the local wireless network. The mobile
data service should notice this and fall back from the
wireless LAN mode to an underlay low speed mobile data
service (GPRS or CDPD). A daemon program is needed
in the terminal to monitor the network resources. When
one network is not available it should adjust the routing
table and trigger the M-IP registration procedure.
Application
TCP
IP
SNDCP
PHY
MAC
PHY
Fr. Rel
PVC
IP
PVC PHY
802 LLC
802.11 MAC
802.11 PHY
802 LLC
802.11 MAC 802.3 MAC
802.11 PHY 802.3 PHY 802.3 PHY
802.3 MAC 802.3 MAC
802.3 PHY
802 LLC 802 LLC
IP
IPIP
802 LLC
802.3 MAC
802.3 PHY
PDN
L2
PDN L1
Mobile
MDBS
Access Point
MD-IS
on the LAN
Mobility
Gateway
MDLP
MAC
MDLP
PVC PVC
Frame Relay
PVC
Fr. Rel
MDLP
SNDCP 802 LLC
802.3 MAC
Figure 4: Interconnecting WLAN and CDPD by Mobility
Gateway (Scenario 5).
Figures 3 and 4 illustrate the protocols involved in
selected scenarios.
3. HANDOFF PROCEDURE
Figure 5 shows the block diagram of the handoff
procedure for inter-tech mobility between WLAN and
GPRS. The handoff is two-fold: handoff from WLAN to
GPRS and handoff from GPRS to WLAN. Figure 7
describes the interactions among the algorithm, metrics,
and the outcomes of executing the algorithm. There are
three items in need of detailed explanation: the HO
architecture (step-by-step description of the procedure to
relate the elements of the network), HO metrics, and the
HO decision mechanism and algorithm. These issues are
addressed in the following.
Figure 5: Handoff procedure for inter-tech mobility.
The general architecture describes the functions of
elements of the network when we have a HO from
WLAN to GPRS or the other way around. First we
describe the step-by-step procedure to go from the WLAN
connection to the GPRS connection. As shown in Figure 6
the following stages occur while the mobile moves away
from the coverage of the WLAN within the GPRS
coverage:
1. Signal received from the AP in the WLAN is strong
2. Signal from the AP becomes weak
3. HO algorithm at the mobile host decides to handoff
from WLAN to GPRS
4. Start/update Mobile IP: Foreign Agent in the MH
gets activated using the Mobile-IP protocol and the
mobile host uses the new IP address.
5. The Home Agent in the WLAN is informed about
the new IP address through the Mobile-IP protocol .
In the reverse situation when the MH is connected to the
GPRS and it realizes that a WLAN is available the
following stages will occur:
1. Signal from the WLAN does not exist.
2. MH detects a signal from the AP of a WLAN
3. HO algorithm decides on the HO.
4. Update Mobile IP: FA in the MH is deactivated by
the Mobile-IP protocol and the regular IP address is
used.
5. HA in the WLAN is informed by the MH through
the M-IP protocol.
Bea con p erio dical ly
1. S tron g sig nal
AP
1
Ho me A gent
LA N
2. W eak sign al
Rou ter
INTERN ET
GPR S
Net work
BTS
3. H and off P roce dure
4. A ctiv ate F orei gn A gent
5. I nform Ho me A gen t
Figure 6: Stages in Inter-tech handoff.
Figure 7 shows the HO mechanism. The overall block
diagram shows the mechanism. The most important item
is the algorithm. We have studied Fuzzy Logic (FL) and
Neural Network (NN) class of algorithms [13,14].
Figure 7: Handoff mechanism.
HO algorithm (Figure 7) gives one of the three possible
action outputs: A, B and C. For a mobile host working
for the underlay network (WLAN) the output A (Relax)
indicates that the link quality of the underlay network is
satisfactory and there is no handoff imminent. Output B
Working in WLAN Working in GPRS
Prepare Ready
for
GPRS
Prepare
Handover
to GPRS
Near
WLAN?
Prepare
Wait
N(sec) Wait
M<N(sec)
Handover
to WLAN
ABC
Yes
Yes
No
No
CB A
HO Algorithm
HO Algorithm
Measurements
Monitoring
Stable
WLAN Unstable
WLAN Poor
WLAN
A B C
Algorithm
(Alert) indicates that there is an increased possibility of
handoff in the near future and gives the system a chance
to prepare for a handoff procedure. The C (Handoff)
output indicates that a handoff is needed and the handoff
procedure to the overlay network (GPRS) is invoked. For
simplicity reasons, this diagram does not show the
possibility of failure, as there was an assumption made
that the overlay network is always available.
A mobile host working for the overlay network (GPRS) is
regularly checking if it is close to the underlay network
(WLAN). The optimum would be to compare
geographical coordinates first. Whatever is the resolution
of the checking, if MH realizes that the underlay network
could be available, it regularly checks for the presence of
WLAN pilot and invokes the fuzzy or neural net engine to
evaluate the quality of the connection. The output of this
engine is the same as in the previous case. The difference
in action from the previous case is that if the connection
indicator is unstable (B), no handoff preparations are
made. MH waits till the link is stable (A) and then a
handoff to the underlay network is performed. Action
output C indicates that no handoff should be performed.
The difference between these is given by the fact that a
user working for the overlay network does not have to be
worried about loosing the connection. Intra-overlay
handoff (within GPRS) is guaranteed by the lower GPRS
layers. Therefore, the user working for the overlay layer
just occasionally checks for the availability of the
underlay network hopefully aided by the location
information to avoid needless search if the MH is, e.g., in
a different country. The finer the location information
becomes the more a useless search for the underlay
network can be avoided. If the location information can
be given by geographical coordinates, MH can store the
underlay network coordinates, e.g., at the time when it left
it last time and start searching for it as soon as it
approaches this area. Also a user-controlled handoff
management is possible, i.e., the user clicks a button in
the screen of laptop when he arrives to his office
environment and otherwise the mobility management
works automatically.
The WLAN to GPRS handoff triggering algorithm is
more crucial for the reliable operation of the system, since
a MH moving away from the underlay network coverage
may suddenly experience a severe degradation of the
service and has to handoff very fast to maintain the higher
layer connection. To facilitate features, such as power
saving by powering down unused interface cards, loading
of interface drivers on demand, the proposed algorithm
allows for an alert, or possibly several levels of alert,
which would enable the system to prepare for an
upcoming handoff. For these reasons the triggering
algorithm is decomposed into two parts, overlay (WLAN)
and underlay (GPRS). To enable a reuse of the code, both
parts use the same HO engine; just the frequency of its
invocation and acting on its output is different.
4. METRICS
There are several handoff metrics that are being employed
in various voice and data mobile networks. Clearly, the
received signal strength is the most popular handoff
metric. However, using only the RSS need not be
optimum as discussed in [13]. A combination of RSS and
some parameters like hysteresis margin, traffic, velocity
have been used with benefits and disadvantages in a
variety of handoff algorithms. We are primarily
concerned with handoff between heterogeneous systems
such as GPRS and IEEE 802.11 networks. A priority type
handoff is required and the RSS is only one metric that
may be employed for handoff decision. Handoff metrics
are closely related to the type of algorithm that may be
ultimately employed to make a handoff decision. We
suggest studying a broad class of handoff metrics
including the RSS, beacon packets, SNR (or noise
indicator), BER, and packet error rates (or
retransmissions) with parameters such as hysteresis
margin and traffic for our project. A second reason for
suggesting a broad class of metrics is the fact that the
algorithms that we are studying, based on neural networks
and fuzzy logic systems can make highly sensitive
decisions based on a large number of inputs. The decision
is made on a pattern match between inputs and previously
observed situations. Consequently, the pattern match may
be made better by employing particular subsets of handoff
metrics as the input. Since the handoff that is being
considered is for data networks, the sensitivity of
performance to the metrics and algorithms will also have
to be investigated. Future work is to concentrate on
investigating the performance and sensitivity of NN and
FL algorithms to different HO metrics.
5. SUMMARY
In this paper we have discussed about procedures,
algorithms and metrics involved in handoff in
heterogeneous wireless networks. A general level
procedure was given for inter-technology roaming
between WLAN and GPRS. However, this procedure can
be applied with any two wireless systems, i.e., WLAN
and UMTS. The main objective of the Wireless LANs for
UMTS (WiLU) project was to explore wideband
technologies for UMTS and develop a network
demonstrator for that purpose. The particular emphasis
was on incorporating WLAN technology into the UMTS.
Since the wideband wireless services will not have a
comprehensive coverage we envision that WLANs should
be embedded into wide area wireless data networks. In
another words, when a wideband wireless data service is
available it will be used by the terminal and when it is not
available the terminal should seamlessly switch to a lower
speed wireless data service offered by UMTS. The focus
for implementation was to incorporate new triggers,
entities or protocols that operate at the management layer,
network or higher layers to enable inter-tech roaming that
will be transparent to the mobile user to the extent
possible. A general level procedure was shown and
potential metcis was discussed for handoff in a
heterogeneous environment.
ACKNOWLEDGMENTS
The acknowledgments are due to the project team of
WiLU (Wireless LAN for UMTS) at the CWC (Centre for
Wireless Communications) in University of Oulu, Finland
and at CWINS (Center for Wireless Information Network
Studies) in Worchester Polytechnic Institute, USA. The
authors would like to thank TEKES, Nokia and Sonera
Ltd. for supporting financially this project.
REFERENCES
1. J.M. Holzman, M. Zorzi, "Advances in Wireless
Communications", Kluwer Academic Publishers 1998,
382 p.
2. G. Brasche and B. Walke, “Concepts, Services, and
protocols of the new GSM Phase 2+ General Packet
Radio Service”, IEEE Communications Magazine,
August 1997.
3. R. Pichna, J. Vallström, M. Ylianttila, J. Talvitie,
“Current Wireless Inter-technology Roaming Efforts”,
Proceedings of the 1st Euromicro Summer School in
Mobile Computing, Oulu, Finland, August 1998, pp.
119-131.
4. Meggers et al., 1996, “Roaming Between GSM and
Wireless LAN”, ACTS Mobile Telecommunications
Summit, Granada, Spain, vol. 2, Nov. 27- 29, pp. 828 -
834.
5. X. Qu, J. Xu Yu, and R. Brent, “Implementation of a
portable-IP system for mobile TCP/IP”, ANU
Computer Science Technical Report, November 1997
http://cs.anu.edu.au/techreports/1997/TR-CS-97-
19.html
6. T. Richarson, F. Benett, G. Mapp, A. Hopper,
“Teleporting in and X Window System Environment,”
IEEE Personal Communications, Vol.1, No.3, pp. 6-
12, 3rd qtr, 1994.
7. A. Bakre and B.R. Badrinath, “Implementation and
Performance Evaluation of Indirect TCP,” IEEE
Trans. Comp., Vol. 46, No.3, March 1997.
8. D.Maltz, P. Bhagwat: MSOCKS: An Architecture for
Transport Layer Mobility, to appear in IEEE
INFOCOM 98.
9. A. Hills and D. Johnson, “A Wireless Data Network
Infrastructure at Carnegie Mellon University”. IEEE
Personal Communications, 3(1):56-63, February 1996.
10. M. Baker, X. Zhao, S. Cheshire, J. Stone, “Supporting
Mobility in MosquitoNet,” Proc. of 1996 USENIX
Technical Conference, January 1996.
11. S. Seshan, H. Balakrishnan, R. Katz, “Handoffs in
Cellular Wireless Networks: The Daedalus
Implementation and Experience,” Kluwer
International Journal on Wireless Communication
Systems, 1996.
12. P. Krishnamurthy, K. Pahlavan, A. Zahedi, J.
Vallström, J. Talvitie, R. Pichna, and M. Ylianttila,
"Handoff in 3G Non-Homogeneous Mobile Data
Networks", European Microwave Week, Netherlands,
October 1998.
13. N. Tripathi, “Generic Adaptive Handoff Algorithms
using Fuzzy Logic and Neural Networks”, Ph.D
Thesis, Virginia Polytechnic Institute and State
University, August 1997, 262 p.
14. A. Hatami, P. Krishnamurthy, K. Pahlavan, J. Mäkela,
R. Pichna, M. Ylianttila, J. Talvitie, "Fuzzy Logic,
Neural Network and Other Algorithms for Handoff in
Wireless Networks", to appear in Proc. of Third
International ICSC Symposium on Fuzzy Logic and
Applications, Rochester, June 1999.
... The primary technical objectives in the design of a seamless vertical handoff system are the balance among handoff latency, power consumption, and wasted bandwidth [3, 6]. A special inter-technology roaming protocol layer can be inserted between the application and the transport protocol layer [16]. This has been done in projects such as On The Move [31] and Mobile TCP/IP [24]. ...
... Three possibilities have been identified so far: no-coupling, loose coupling, tight coupling (at SGSN/GGSN level) [5, 12, 16]. ...
... The backward MAHO (mobile assisted handover) with signalling diversity is adopted in the paper, as it is considered to be the optimum strategy in minimizing the signalling load and delays, and allowing a relatively simple MT. In [16] it is shown how fuzzy logic and neural networks class of algorithms can be used to control inter-technology handoff procedure. WLAN and GPRS are used as a case study example. ...
Article
Next generation mobile networks, commonly referred to as 4G, are envisaged as a multitude of heterogeneous systems interacting through a horizontal IP-centric architecture. This paper presents an overview of the issues arising in such hybrid systems and international research work, with particular emphasis on quality of service (QoS) and vertical handoff. An outline of our research is presented, that is, optimising the handover process in proportion to both delay and throughput, by measuring QoS parameters and link characteristics at appropriate places in the networks, and drawing out decisions upon the profitability/necessity of a vertical handoff.
... A MIH based Velocity Optimized Seamless Handover Mechanism (VOSHM) for WiMAX a network is described in [23]. It shows how the handover probability value is affected by the velocity of the mobile node. ...
... It shows how the handover probability value is affected by the velocity of the mobile node. If an MT is associated with the WLAN for the longest possible duration, the user throughput is improved and even during the transition period, the RSS oscillates around the receiver sensitivity level [23]. ...
Article
Full-text available
WiMAX is a highly reliable and cost effective technology. A Mobile Mesh Network (MMN) involves communication between mobile nodes and mesh routers. Based on the Sponsor Node to adjacent node signal power Ratio (SNDR) value of each node, the decision for inter-cluster or intra-cluster handover is made. TCP Round Trip Timer (RTT) method is used to reduce congestion in the Base Station (BS) during handover. The congestion at the BS is detected and the contention window size is dynamically computed based on the contention RTT in the MAC layer and congestion RTT in the transport layer. The proposed system outperforms the existing system in terms of Throughput, delay, computation and message overheads.
... The AGAL protocol must solve a similar problem: location of the ad-hoc mobile host that corresponds to the mobile host with which communication is occurring over GPRS. One area where there is more leverage from existing work, such as [6] [7] [2] [8] [9], is the algorithm for determining when to handoff from one network to another. [7] makes the decision based on signal strength, while [6] considers data rate as a more accurate measure of the switching time. ...
... One area where there is more leverage from existing work, such as [6] [7] [2] [8] [9], is the algorithm for determining when to handoff from one network to another. [7] makes the decision based on signal strength, while [6] considers data rate as a more accurate measure of the switching time. [8] [9] consider efficient methods for avoiding congestion control in TCP during handoff. ...
Article
The explosive increase in the number of mobile users in the present day scenario has made it increasingly tough from the provider's point of view to provide each of the users with a bandwidth high enough for high quality support of services such as multimedia, gaming etc. In this paper we propose Ad-hoc/GPRS Alternate Link (AGAL) protocol which aims at shifting as many as possible mobile users to high-speed , inexpensive ad-hoc mode whenever they are in range of each other. In AGAL, a mulithomed mobile client has both a GPRS/GSM and 802.11 ad-hoc link. Whenever the communicating mobile users are able to detect each other through their ad-hoc interface, they shift over to the higher quality ad-hoc link which also helps ease up the load on the WWAN network. Through simulation we show that AGAL greatly improves user throughput without significant increase of overhead.
... Different procedures, algorithms and metrics related to heterogeneous wireless environment are discussed in [125]. The paper presents a handover algorithm usable when using two access networks: a cellular WWAN network and an accompanying WLAN network. ...
... The schemes seem feasible. In Ylianttila et al. (1999), a framework for vertical handoff method using fuzzy logic in heterogeneous wireless networks is proposed. In Majlesi and Khalaj (2002), one handoff algorithm using fuzzy logic is proposed for hybrid networks built using mobile networks and WLANs. ...
Article
Full-text available
In wireless and mobile networks ensuring smooth handoff and tracing exact location of destination node and that too without wasting much time and bandwidth of constrained network is much essential. Several related techniques have been proposed in literature. Each has its own advantages/disadvantages. In this paper, we have proposed, first, a fuzzy logic based handoff management scheme that also handles some basic network management activities like call transferring after allotting a suitable wireless or radio link to the incoming node. It also proposes a location management scheme based on simple heuristics technique like time and probability.
Article
There are some corrections to the text on the following pages: 1. In page 371, instead of LMP, it should be AS: When MH moves out of the service area of the current AS, L2 and/or L3 handover should be executed initially. 2. In page 373, instead of MMH, it should be MH. Inter-Subnet Handover - When a MH moves from one subnet to another, within the same GW, an IP handover takes place. This is known as L3 Handover. IPv6 protocol stack in MH is notified by a local event that triggers IP handover. 3. In page 376 and 377, the unit of velocity is ‘m/s’ instead of ‘ms’. Velocity of the mobile node is varied from 2 m/s to 40 m/s and the packet loss rate is observed during L2 and L3 handover. Velocity of the mobile node is varied from 2 m/s to 40 m/s and the handover latency is observed during layer 2 and layer 3 handover.
Conference Paper
Fast and seamless handover are weighty issues in 3G systems. Based on UMTS and WLAN Converging architecture, a joint vertical handover decision strategy (jVHOD) was proposed for heterogeneous wireless networks. Also, several significant metrics were evaluated through a vertical handover performance model, which was validated by computer simulation. The analytical results showed that jVHOD had better performance than the conventional in terms of the number of handoffs and available bandwidth, but a little longer packet delay. To design optimal vertical handover strategies the application requirement threshold should be properly adjusted by an appropriate trade-off of related handover performance metrics.
Chapter
Full-text available
The ideas and basic principles of multiuser receivers for code-division multiple-access (CDMA) systems are summarized and reviewed. Linear multiuser receiver formulation and combination of MAI suppression with multipath and antenna combining is the main contribution of the chapter. The multiuser equalization can be performed either after maximal ration combining or there can be a separate equalizer in each rake receiver branch. If the latter choice is made, the correlation of noise in the equalizer outputs needs to be taken into consideration when selecting the combining weights. Efficient bit error probability (BEP) evaluation methods for such linear receivers with uncoded transmission are also presented. Thirdly, chip equalizers for downlink are described. Their major advantage is that they are applicable also in CDMA systems where the period of spreading sequences is significantly longer than the data symbol period.
Chapter
The wireless data communication industry has experienced fast development in the past few years. With the finalization of new series of IEEE 802.11 and ETSI BRAN HIPERLAN standards, new features have been integrated into the conventional wireless LAN, which was introduced as an alternative of fixed LAN. New emerging technologies, such as HomeRF and Bluetooth, are becoming new impetus for the fast expansion of the market. In this paper, we present an overview of the current status and future trends of wireless data communication systems.
Article
Full-text available
Network protocols in cellular wireless data networks must update routes as a mobile host moves between cells. These routing updates combined with some associated state changes are called handoffs. Most current handoff schemes in wireless networks result in data loss or large variations in packet delivery times. Unfortunately, many applications, such as real-time multimedia applications and reliable transport protocols, adapt to long term estimates of end-to-end delay and loss. Violations and rapid fluctuations of these estimates caused by handoff processing often result in degraded performance. For example, loss during handoff adversely affects TCP performance [4], and high packet loss and variable delays result in poor real-time multimedia performance. In this paper, we describe a multicast-based protocol that eliminates data loss and incurs negligible delays during a handoff. The basic technique of the algorithm is to anticipate a handoff using wireless network information in the form of received signal strengths and to multicast data destined for the mobile host to nearby base stations in advance. This routing, combined with intelligent buffering techniques at the base stations, enables very rapid routing updates and eliminates data loss without the use of explicit data forwarding. We have implemented this protocol using IP Multicast and Mobile IP-like routing. In our implementation, handoffs typically take between 8 and 15 ms to complete and result in no data loss.
Article
Full-text available
For almost three years the European Telecommunications Standards Institute (ETSI) has been working on the specification of the general packet radio service (GPRS) for the Global System for Mobile Communications (GSM). In the course of 1997 standardization will come to an end, and a first introduction of GPRS by a service provider or network operator is likely to take place in 1999. The new service will accommodate data connections with variable bit rates and high bandwidth efficiency, and thus offers the possibility to attract a wide range of new applications to GSM networks. In this article possible applications and elementary concepts and service characteristics of GPRS are explained in detail. Furthermore, a medium access control protocol that conforms to the draft standard proposal for the GPRS air interface is introduced and analyzed. Thereafter, the simulation model is described and simulation results presented. The performance analysis carried out shows reasonable performance even under high load conditions
Article
Full-text available
The goal of the MosquitoNet project is to provide continuous Internet connectivity to mobile hosts. Mobile hosts must be able to take advantage of the best network connectivity available in any location, whether wired or wireless. We have implemented a mobile IP system that supports seamless switching between different networks and communication devices. In contrast to previous approaches to mobile IP, we believe mobile hosts should not assume any explicit mobility support from the networks they visit, aside from basic Internet connectivity. This decision places extra responsibilities on the mobile hosts themselves. In this paper, we describe the design and implications of such a system. Measurements of our implementation show that the inherent overhead to switch networks (below 10ms) is insignificant compared to the time required to bring up a new communication device.
Chapter
This chapter discusses the basics of wireless communications, such as signal multiplexing, filtering, and modulation. The sharing of a communications channel by multiple users is known as multiplexing. There are a number of methods to successfully locate and receive particular signals and reject others, including frequency-division multiplexing, time-division multiplexing, spatial multiplexing, code-division multiplexing, and directional multiplexing. Any signal that is periodic in time can be regarded as being composed of sinusoids of differing frequencies or described either as a function of time or as a function of frequency by taking its Fourier transform. The orthogonality of differing frequencies makes it possible to extract a signal from a complex mixture of signals of different frequencies, which is generally known as filtering. To transmit information, waves must be modulated, resulting in a finite bandwidth for a given data rate and thus a finite amount of spectrum to share between differing uses and users. The chapter examines various modulation schemes including binary phase-shift keying (BPSK), quadrature-amplitude modulation (QAM), on-off keying (OOK), and amplitude-shift keying (ASK). The chapter also examines two specialized modulation techniques, orthogonal frequency-division multiplexing (OFDM), and pulse radio, of increasing interest in the modern wireless world.
Article
E#cient hando# algorithms cost-e#ectively enhance the capacity and Qualityof Service (QoS) of cellular systems. This research presents novel approachesfor the design of high performance hando# algorithms that exploit attractivefeatures of several existing algorithms, provide adaptation to dynamic cellularenvironment, and allow systematic tradeo#s among di#erent system characteristics.
Conference Paper
Mobile nodes of the future will be equiped with multiple network interfaces to take advantage of overlay networks, yet no current mobility systems provide full support for the simultaneous use of multiple interfaces. The need for such support arises when multiple connectivity options are available with different cost, coverage, latency and bandwidth characteristics, and applications want their data to flow over the interface that best matches the characteristics of the data. We present an architecture called transport layer mobility that allows mobile nodes to not only change their point of attachment to the Internet, but also to control which network interfaces are used for the different kinds of data leaving from and arriving at the mobile node. We implement our transport layer mobility scheme using a split connection proxy architecture and a new technique called TCP splice that gives split connection proxy systems the same end to-end semantics as normal TCP connections
Article
In order to support mobile computing research, including the development of software which will allow seamless access to multiple wireless data networks, a wireless data network infrastructure is being built at Carnegie Mellon University. This infrastructure will allow researchers and other members of the campus community to use mobile computers to gain access to data networks while they are on-campus, or off-campus in the greater Pittsburgh area. The infrastructure will initially include two different types of wireless networks, a low-bandwidth wide area system and a high-bandwidth local area system, each of which will provide access to the campus computer network. Since the campus network is called “Andrew” (after Andrew Carnegie and Andrew Mellon), the new wireless infrastructure has been dubbed “Wireless Andrew”. The article describes the Wireless Andrew infrastructure being built. An overview of the infrastructure and the characteristics of the two types of wireless networks used are presented. Each of these networks is then described in more detail. The technology used, its implementation in the wireless network infrastructure, and deployment experience are discussed
Article
With the advent of small portable computers and the technological advances in wireless communications, mobile wireless computing is likely to become very popular in the near future. Wireless links are slower and less reliable compared to wired links and are prone to loss of signal due to noise and fading. Furthermore, host mobility can give rise to periods of disconnection from the fixed network. The use of existing network protocols, which were developed mainly for the high bandwidth and faster wired links, with mobile computers thus gives rise to unique performance problems arising from host mobility and due to the characteristics of wireless medium. Indirect protocols can isolate mobility and wireless related problems using mobility support routers (MSRs) as intermediaries, which also provide backward compatibility with fixed network protocols. We present the implementation and performance evaluation of I-TCP, which is an indirect transport layer protocol for mobile wireless environments. Throughput comparison with regular (BSD) TCP shows that I-TCP performs significantly better in a wide range of conditions related to wireless losses and host mobility. We also describe the implementation and performance of I-TCP handoffs
Article
This article describes the Wireless Andrew infrastructure we are building. We begin with an overview of the infrastructure and the characteristics of the two types of wireless networks we are using. Next, we describe in more detail each of these networks. We discuss the technology used, our implementation of it in our wireless network infrastructure, and our current deployment experience. Finally, we summarize the lessons we have learned and present conclusions. Overview of Infrastructure