This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.
D. Camps-Mur, X. Perez-Costa, V. Marchenko and S. Sallent Ribes. On Centralized
Schedulers for 802.11e WLANs: Distribution vs Grouping of Resources Allocation. Under Submission.
X. Perez-Costa and D. Camps Mur. IEEE 802.11e QoS and Power Saving feature: Overview and Analysis of Combined Performance. Accepted for publication at IEEE Wireless Communications Magazine.
A. Zubow, D. Camps-Mur, X. Pérez-Costa, P. Favaro. Greedy Scheduling Algorithm (GSA) - Design and Evaluation of an Efficient and Flexible WiMAX
OFDMA Scheduling Solution. Elsevier Computer Networks Journal (CN), abstract pdf-preprint link
WiMAX is one of the most promising technologies to provide broadband wireless access in the near future. In this paper we focus on the study of the combined performance of a WiMAX Base Station MAC downlink scheduler and OFDMA packing algorithm which mainly determine the usage efficiency of the available radio resources. We design and analyze an efficient and flexible solution, greedy scheduling algorithm (GSA), and evaluate its performance as compared to several relevant alternative solutions. Specifically, we analyze performance differences with respect to efficiency, flexibility to provide per subscriber station burst shape preferences, interference mitigation and computational load. Our results show that GSA achieves a performance similar to the one of the competing approaches considered in terms of efficiency, even better in some cases, and significantly outperforms them in flexibility to provide per subscriber station burst shape preferences, interference mitigation and computational load. As a conclusion, the proposed GSA solution is a promising candidate to maximize the utilization of the available WiMAX radio resources at a low computational cost while at the same time being able to fulfill a wide range of requirements based on operators’ preferences and/or network environment specifics.
G. Hiertz, D. Denteneer, L. Stibor, Y. Zang, X. Pérez Costa, B. Walke. The 802.11 Universe. IEEE Communications Magazine, Volume 48, Issue 1, January 2010.
abstract pdf link
The introduction of IEEE's 802.11 standards has enabled a mass market, with a huge impact in the home, office, and public areas. Today, laptops, PCs, printers, cellular phones, VoIP phones, MP3 players, Blu-Ray players, and many more devices incorporate wireless LAN technology. With low-cost chipsets and support for high data rates, 802.11 has become a universal solution for an ever increasing application space. As a direct consequence of its high market penetration, several amendments to the basic 802.11 standard have been developed or are under development. They fix technology issues or add functionality expected to be required by future applications. In this article we overview the emerging 802.11 standard and address the technical context of its extensions. The article highlights its finalized amendments and those under development.
D. Camps Mur, X. Perez-Costa and S. Sallent Ribes. An Adaptive Solution for Wireless LAN
Distributed Power Saving Modes. Elsevier Computer Networks Journal (CN), Volume 53, Issue 18, December 2009. abstract pdf-preprint link
The current trend to incorporate the Wireless LAN technology in increasingly smaller mobile devices poses new challenges, in terms of QoS and power consumption requirements, for the design of such devices. IEEE 802.11 and 802.11e define mechanisms to address these challenges but do not provide any guidance about how to design the algorithms required to make use of them to achieve an optimum performance. The work presented in this paper focuses on the design of an adaptive algorithm for the distributed power saving mechanisms of Wireless LANs when facing the challenge of providing QoS to devices in a power save mode. Our main contributions are (i) an analytical model that captures the dependencies between the QoS experienced and the configuration of the Wireless LAN distributed power saving mechanisms, (ii) a generic algorithm based on the steepest descent method that, using only information available in the MAC layer, adapts to the applications’ characteristics and configures a power saving mechanism in order to provide a satisfactory QoS experience, (iii) an analysis of the convergence properties ofthe proposed algorithm that provides the optimum values of its configurable parameters, and (iv) a thorough simulative study that demonstrates the suitability of our adaptive solution in todays typical Wi-Fi deployments and its advantages in front of existent solutions in the state of the art.
X.Pérez-Costa, D.Camps-Mur and Albert Vidal. On the Distributed Power Saving Mechanisms of Wireless LANs 802.11e U-APSD vs 802.11 Power Save Mode. Elsevier Computer Networks Journal (CN), Volume 51, Issue 9, June 2007. abstract pdf-preprint link
The integration of wireless LAN technology in mobile devices such as cellular phones, PDAs or laptops has become a user need due to its popularity in providing high speed wireless Internet access at a low cost. Such devices though should meet users’ expectations with regard to QoS, i.e., guarantee a reasonable voice quality when VoIP is used, and power saving efficiency, i.e., standby and calling times should be similar to the ones of cellular phones. The IEEE 802.11e standard, which extends the 802.11 wireless LAN MAC layer with QoS and power saving enhancements, should be the most appropriate solution to address users’ wishes in those devices.
In this paper, we focus on the 802.11e functionality likely to be included in mobile devices in the short-term, EDCA for QoS and U-APSD for power saving, and evaluate the performance improvements and associated costs of two possible configurations of U-APSD as compared to the 802.11 power save mode. In addition, the dependency between the QoS and power saving enhancements obtained with U-APSD and the available channel capacity is analyzed considering three different scenarios: 802.11b, 802.11b+g and 802.11g. The evaluation is based on our proposed implementation of U-APSD: Static U-APSD (SU-APSD).
The main conclusions that can be drawn from our results are that U-APSD significantly outperforms the 802.11 power save mode in all considered performance metrics and that the performance enhancements obtained with U-APSD areindependent of the available channel capacity.
X.Pérez-Costa, D.Camps-Mur and T.Sashihara. Analysis of the Integration of IEEE 802.11e Capabilities in Battery Limited Mobile Devices. IEEE Wireless Communications Magazine (WirComMag), special issue on Internetworking Wireless LAN and Cellular Networks, Volume 12, Issue 6, December 2005. abstract pdf link
IEEE 802.11 wireless LANs (WLANs) is
widely recognized as a complementary technology
to cellular access networks in hot-spot areas
due to its lower cost and higher data rates. Different
interworking approaches are being studied,
but one common element is considered by
all of them: mobile devices can include capabilities
for connecting through both access technologies,
allowing the best option to be chosen
depending on the availability at a specific
moment. However, several challenges need to be
addressed in order to achieve seamless integration
of WLAN and cellular systems in such a
mobile device. The IEEE 802.11e standard,
which defines mechanisms to provide QoS in a
WLAN, represents a basic step toward this integration,
since it provides the means to support
key applications such as voice over IP. The IEEE
802.11 power-save mode (PSM) is another necessary
element for devices with severe battery
limitations (e.g., cellular phones) in order to
ensure reasonable battery duration. The resulting
performance when both QoS and power-saving
mechanisms are used together, however, is
uncertain and requires further study. We analyze
the implications of the interaction of the 802.11
PSM with 802.11e QoS mechanisms by determining
if the desired QoS is still provided,
detecting functionality conflicts, and quantifying
the impact of the PSM upon the 802.11e QoS
efficiency and system performance. The evaluation
is performed via simulation.
V.Marques, X.Pérez-Costa, R.L.Aguiar, M.Liebsch and A. M. O. Duarte. Evaluation of a Mobile IPv6-based Architecture supporting User Mobility QoS and AAAC in Heterogeneous Networks. IEEE Journal on Selected Areas in Communications (JSAC), special issue on Wireless Overlay Networks Based on Mobile IPv6, Volume 23, No. 11, November 2005. abstract pdf link
This paper presents a Mobile IPv6-based overlay network architecture for heterogeneous environments, designed entirely based on IPv6, that aims to be implemented seamlessly irrespectively of the supporting network infrastructure. All transmission technologies are handled at the physical and data-link layers, imposing IPv6-based protocols for all higher layer communications and signaling. The architecture builds on Mobile IPv6 including improved fast handover, and integrates quality-of-service and authentication, authorization, accounting, and charging control per user. The most critical issues of the proposed architecture, mainly related to the handover process, were subject of a performance evaluation via ns-2 simulations. Finally, a field trial of the system was implemented, overlaying part of the GEANT infrastructure between Madrid and Stuttgart, which results are presented here.
C.Bettstetter, H.Hartenstein and X.Pérez-Costa. Stochastic Properties of the Random Waypoint Mobility Model. ACM/Kluwer Wireless Networks (WINET), special issue on Modeling and Analysis of Mobile Networks, vol. 10, no. 5, September 2004. abstract pdf-preprint link
The random waypoint model is a commonly used mobility model for simulations of wireless communication networks. By giving a formal description of this model in terms of a discrete-time stochastic process, we investigate some of its fundamental stochastic properties with respect to: (a) the transition length and time of a mobile node between two waypoints, (b) the spatial distribution of nodes, (c) the direction angle at the beginning of a movement transition, and (d) the cell change rate if the model is used in a cellular-structured system area.
The results of this paper are of practical value for performance analysis of mobile networks and give a deeper understanding of the behavior of this mobility model. Such understanding is necessary to avoid misinterpretation of simulation results. The movement duration and the cell change rate enable us to make a statement about the "degree of mobility" of a certain simulation scenario. Knowledge of the spatial node distribution is essential for all investigations in which the relative location of the mobile nodes is important. Finally, the direction distribution explains in an analytical manner the effect that nodes tend to move back to the middle of the system area.
X.Pérez-Costa, M.Torrent-Moreno and H.Hartenstein. A Performance Comparison of Mobile IPv6, Hierarchical Mobile IPv6, Fast Handovers for Mobile IPv6 and their Combination. ACM SIGMOBILE Mobile Computing and Communications Review (MC2R), Volume 7, Issue 4, October, 2003. abstract pdf-preprint link
Mobile IP, the current IETF proposal for IP mobility support, represents a key element for future All-IP wireless networks to provide service continuity while on the move within a multi-access environment. We conducted a performance evaluation of Mobile IPv6 and its proposed enhancements, i.e., Fast Handovers for Mobile IPv6, Hierarchical Mobile IPv6 and our proposed combination of them, using the network simulator ns-2 for the case of a 'hot spot' deployment scenario. The simulation scenario comprises four access routers and up to 50 mobile nodes that communicate in accordance with the IEEE 802.11 wireless LAN standard. The study provides quantitative results of the performance improvements obtained by the proposed enhancements as observed by a single mobile user with respect to handoff latency, packet loss rate and achieved bandwidth per station. As a complementary part of the study, the signaling load costs associated with the performance improvements provided by the enhancements has been analyzed. The simulation environment allowed us also to investigate the behavior of the protocol in extreme cases, e.g., under channel saturation conditions and considering different traffic sources: CBR, VoIP, Video and TCP transfers. While some simulation results corroborate the intention of the protocols specifications, other results give insights not easily gained without performing simulations. This study provides a deep understanding of the overall performance of the various protocols and supports the design process of a Mobile IPv6-based network when a decision of whether it is appropriate to implement any of the proposed Mobile IPv6 enhancements has to be made.
X.Pérez-Costa, H.Hartenstein and C.Bettstetter. Towards a Mobility Metric for Reproducible and Comparable Results in Ad Hoc Networks Research. ACM SIGMOBILE Mobile Computing and Communications Review (MC2R), special feature on MobiCom 2003 posters, Volume 7, Issue 4, October 2003. abstract pdf poster link
The performance of a mobile ad hoc network depends on
its ability to adapt to changes in the network topology. It is
therefore important to understand the relation between node
mobility and topology dynamics. While the input parameters
of a mobility model allow for generating different simulation
scenarios (e.g., by selecting the speed, pause time,
system area, and number of nodes), the relation to the actually
generated degree of topology dynamics is usually not
immediate. Thus, a translation of the mobility model to the
resulting topology dynamics is required. A natural way to do
this is to introduce the topology change rate (TCR), defined
as the number of link changes per time unit as observed by a
single node. If we measure the network performance versus
the TCR instead of versus ‘incidental’ input parameters of
a mobility model, a higher level of abstraction of the results
will be achieved, and the reproducability and comparability
of the results will be enhanced.
This short paper derives in an analytical manner the TCR
in a scenario where nodes move according to the random
waypoint (RWP) model [5], being the most heavily used mobility
model in reseach on ad hoc networks. Recent insights
into RWP mobility have shown some unexpected behavior,
e.g., with respect to the spatial node distribution [2] and the
average velocity over time [8]. These results do not mean
that RWP as a mobility model is ‘invalid’, but show the
importance of proper use due to thorough understanding of
the model. This contribution now provides the link between
RWP mobility and the corresponding TCR.
X.Pérez-Costa and H.Hartenstein. A Simulation Study on the Performance of Mobile IPv6 in a WLAN-Based Cellular Network. Elsevier Computer Networks Journal (CNJ), special issue on The New Internet Architecture, September 2002. abstract pdf-preprint link
We performed a simulative evaluation of Mobile IPv6 (MIPv6) via ns-2 for a scenario comprising up to three access routers and 30 mobile nodes that communicate in accordance with the IEEE 802.11 wireless LAN standard. We compare standard MIPv6 to a fast handoff procedure and analyze route optimization issues. We present results with respect to handoff latency, packet loss, end-to-end delay, signaling load, channel utilization, and bandwidth per station, and show how various traffic types like UDP CBR, VoIP, and TCP are affected by the handoffs. While some simulation results corroborate the intention of the protocol specification, other results give insights not easily gained without performing simulations. For example, we learned that (i) the fast handoff mechanism almost fully eliminates packet losses but only improves handoff latency in specific cases, (ii) the signaling load introduced by MIPv6 does not severely affect the performance, and (iii) in a shared environment like IEEE 802.11, traffic type can affect the handoff rate since in high-load situations router advertisements might get lost more frequently thereby 'irritating' the handoff decision mechanism. As a challenge for future simulations we have identified the design of a random mobility model that allows good control of the handoff rate.
X. Pérez-Costa, M. Mezzavilla, J.R.B de Marca and J. Arauz. E-Diophantine: An Admission Control Algorithm for WiMAX Networks. Accepted for publication at IEEE Wireless Communications and Networking Conference (WCNC 2010).
WiMAX is one of the most promising technologies
to provide broadband wireless access in the near
future. In this paper we identify a key element for the
performance of a WiMAX network, the DL-MAP packing
algorithm, which mainly determines the usage efficiency
of the available radio resources and investigate potential
differences that could appear between WiMAX equipment
vendors in the maximum capacity of the system due
to the packing approach used. Our results show that
the performance of simple DL-MAP packing algorithms
might be significantly outperformed by more complex
ones resulting in a clear differentiation factor among
manufacturers.
M. Einhaus, A. Maeder and X. Pérez Costa. A Zone Assignment Algorithm for Fractional
Frequency Reuse in Mobile WiMAX Networks. Accepted for publication at IFIP Networking 2010.
A. Maeder, M. Einhaus, X. Pérez Costa and L. Fan. IEEE 802.16m for IMT-Advanced: The Next Step
in Wireless MAN Evolution. In Proceedings of 9th Workshop on IP: Joint ITG and Euro-NF Workshop "Visions of Future Generation Networks" (EuroView 2009), Würzburg, Germany, July 2009. abstract pdf link
The next step in the evolution of the IEEE WirelessMAN
project is IEEE 802.16m. Building upon IEEE 802.16e
Rev2 mobile WiMAX, this standard in development is one of
the contenders in the IMT-Advanced programme of the ITU.
Designed for high data rates and low packet latencies, IEEE
802.16m has the potential to become a key enabler of the future
wireless Internet. We provide a comprehensive overview of the
current status, development plan and main technical features,
including advanced concepts like Femtocells, relays and self-organizing
networking.
P. Serrano, P. Patras, X. Pérez Costa, B. Gloss and D. Chieng. A MAC Layer Abstraction for Heterogeneous Carrier Grade Mesh Networks. In Proceedings of ICT Mobile Summit, Santander, Spain, July 2009. abstract pdf link
Providing carrier grade services to a large number of mobile users is
becoming an important challenge for wireless network operators. One promising solution
for offering cost-efficient alternatives compared to classical cellular approaches
is the use of wireless mesh networks along with the use of heterogeneous radio technologies.
In this paper we propose a MAC abstraction layer to lessen the management
burden of heterogeneous radio technologies. This abstraction layer is intended to hide
the complexity and specifics of different wireless interfaces, this way supporting the
use of a single set of routing and capacity handling mechanisms.
X. Pérez-Costa, P. Favaro, A. Zubow, D. Camps-Mur and J. Arauz. On the Challenges for the Maximization of Radio Resources Usage in WiMAX Networks. In Proceedings of IEEE Consumer and Communications Networking Conference (CCNC) Broadband Wireless Access (BWA) Workshop, Las Vegas, USA, January 2008. abstract pdf link
WiMAX is one of the most promising technologies
to provide broadband wireless access in the near
future. In this paper we identify a key element for the
performance of a WiMAX network, the DL-MAP packing
algorithm, which mainly determines the usage efficiency
of the available radio resources and investigate potential
differences that could appear between WiMAX equipment
vendors in the maximum capacity of the system due
to the packing approach used. Our results show that
the performance of simple DL-MAP packing algorithms
might be significantly outperformed by more complex
ones resulting in a clear differentiation factor among
manufacturers.
X. Pérez-Costa and D. Camps Mur. A Protocol Enhancement for IEEE 802.11 Distributed Power Saving Mechanisms: No Data Acknowledgment. In Proceedings of IST Mobile Summit 2007 Conference, Budapest, Hungary, July 2007. abstract pdf link
Mobile devices including Wireless LAN functionality are becoming
increasingly popular in society. The wide range of products available
in the market target different customer needs but most of them should
meet two main requirements: QoS support for differentiating real-time
services from non real-time and power saving functionality to achieve an
operating time according to users’ expectations. The devices presenting
the most challenging technical issues to meet these two requirements are
dual-mode handsets (Cellular/WLAN) because of their mandatory support
of a strict QoS demanding application, VoIP, and their small device size
which severely limits the battery capacity. The focus of our work in this
paper is the design and evaluation of an enhancement of the distributed
Wireless LAN power saving mechanisms defined in the IEEE 802.11 and
802.11e standards, No Data Acknowledgment (NDACK). The objective of
NDAck is to increase the battery lifetime of Wireless LAN mobile devices
while providing the required QoS. The protocol improvement designed has
been implemented in OPNET to evaluate the performance enhancements
obtained. Our results show that i) NDAck significantly reduces the power
consumption of stations running real-time applications, ii) the larger the
power consumption due to the congestion in the wireless channel the larger
the power consumption reduction with NDAck and iii) NDAck results in a
considerable QoS improvement for real-time applications.
X.Pérez-Costa, D.Camps-Mur, J. Palau, D. Rebolleda and S. Akbarzadeh. OAS-APSD: Design and Evaluation of an Overlapping Aware Scheduled Automatic Power Save Delivery Algorithm. In Proceedings of European Wireless Conference (EW), Paris, France, April 2007.
abstract pdf link
Mobile devices including Wireless LAN functionality are becoming
increasingly popular in society. The wide range of products available
in the market target different customer needs but most of them should
meet two main requirements: QoS support for differentiating real-time services
from non real-time and power saving functionality to achieve an operating
time according to users’ expectations. The devices presenting the
most challenging technical issues to meet these two requirements are dualmode
handsets (Cellular/WLAN) because of their mandatory support of a
strict QoS demanding application, VoIP, and their small device size which
severely limits the battery size. The focus of our work in this paper is the
evaluation of the capability of a specific IEEE 802.11e power saving mechanism,
Scheduled Automatic Power Save Delivery (S-APSD) in combination
with HCF Controlled Channel Access (HCCA) to efficiently address
the challenges posed by the requirements of Wireless LAN-capable mobile
devices. In order to do so, we designed an Overlapping Aware S-APSD algorithm
(OAS-APSD) and evaluated its performance in combination with
HCCA as compared to the basic 802.11 power save mode using EDCA.
X.Pérez-Costa and D.Camps-Mur. AU-APSD: Adaptive IEEE 802.11e Unscheduled Automatic Power Save Delivery. In Proceedings of IEEE International Conference on Communications (ICC), Istanbul, Turkey, June 2006. abstract pdf link
The integration of the wireless LAN technology in
mobile devices such as cellular phones or PDAs has become
a user need due to its popularity for providing high speed
wireless Internet access at a low cost. Such devices though should
meet users’ expectations with regard to QoS, e.g., guarantee a
reasonable voice quality when VoIP is used, and power saving
efficiency, e.g., standby and calling times should be similar to
the ones of cellular phones. IEEE 802.11e defines QoS and
power saving enhancements that should allow the wireless LAN
technology address users’ wishes in such specific devices. Our
focus is the study of the distributed power saving mechanism of
802.11e, i.e., U-APSD, as compared to the legacy 802.11 power
saving mode in order to assess its suitability for solving the
challenges of the upcoming mobile devices requirements. Our
contributions are as follows. We provide first an overview of the
U-APSD functionality. Then, we describe in detail our proposed
implementation of the U-APSD mechanism, Adaptive U-APSD
(AU-APSD), a generic solution that requires only information
available at the MAC layer. Finally, we quantify the performance
improvements that are obtained with our proposed AU-APSD
implementation as compared to the legacy 802.11 power save
mode.
X.Pérez-Costa, A.Vidal and D.Camps-Mur. SU-APSD: Static IEEE 802.11e Unscheduled Automatic Power Save Delivery. In Proceedings of European Wireless Conference (EW), Athens, Greece, April 2006. abstract pdf link
The integration of the wireless LAN technology
in mobile devices as cellular phones or PDAs has become
a user need due to its popularity in providing high
speed wireless Internet access at a low cost. Such devices
though should meet users’ expectations with regard
to QoS, e.g., guarantee a reasonable voice quality when
VoIP is used, and power saving efficiency, e.g., standby
and calling times should be similar to the ones of cellular
phones. IEEE 802.11e defines QoS and power saving
enhancements that should allow the wireless LAN technology
address users’ wishes in such specific devices. However,
there are several questions that need to be answered in order
to assess whether the additional complexity introduced
by these mechanisms would be justified by a relevant performance
improvement. With our work we contribute to
answer these questions by i) identifying the likely 802.11e
functionality to be implemented in mobile devices in the
short-term (EDCA+U-APSD) ii) proposing a specific implementation
of the U-APSD mechanism (SU-APSD) and
iii) evaluating the performance improvement obtained as
compared to 802.11 standard power save mode.
X.Pérez-Costa, K.Heinze, A.Banchs and S.Sallent-Ribes. Analysis of Performance Issues in an IP-based UMTS Radio Access Network. In Proceedings of ACM Workshop on Modeling, Analysis and Simulation of Wireless and Mobile Systems (MSWiM), Montreal, Canada, October 2005. abstract pdf-preprint link
The substitution of ATM transport by IP in future UMTS Radio Access Networks (UTRAN) introduces several performance challenges that need to be addressed to guarantee the feasibility of its deployment. The significant increase of the overhead requires of header compression and multiplexing methods to achieve a usage of the UTRAN resources similar to the ATM one. Additionally, the specific UTRAN transport needs require the adaptation of standard packet scheduling mechanisms to efficiently use the network resources while providing the required QoS. Our results show that, applying header compression plus multiplexing techniques and taking into account the specific UTRAN synchronization requirements for QoS scheduling, very significant performance improvements can be obtained.
X.Pérez-Costa and D.Camps-Mur. APSM: Bounding the Downlink Delay for 802.11 Power Save Mode. In Proceedings of IEEE International Conference on Communications (ICC), Seoul, Korea, May 2005. abstract pdf link
The popularity of wireless LANs, due to its provision
of high speed wireless Internet access at a low cost and
the cheap chipset price, has resulted in a strong trend toward
the integration of this technology in the upcoming all-in-one
mobile devices that could include for instance cellular, wireless
LAN and personal digital assistant (PDA) capabilities. Such
devices though require of power saving mechanisms in order to
guarantee a reasonable battery duration. The 802.11 standard
provides a power save mode that reduces the wireless LAN
technology power consumption, however, this mode can result
in downlink delays (AP to station) unacceptable for the QoS
of some applications, e.g., VoIP. To overcome this problem, we
propose an adaptive power save mode algorithm (APSM) that
adapts the data frames MAC downlink delay of a certain station
according to the downlink frame interarrival time observed at
the AP MAC layer. We conducted an evaluation of our proposal
with respect to downlink delay, power efficiency and signaling
load using the OPNET simulator and compared its performance
with 802.11 standard power save mode and two different static
alternatives. The results show the effectiveness of our algorithm
in providing a soft upper bound of the MAC downlink delay while
significantly decreasing the power consumption and requiring a
signaling load similar to the one of standard power save mode.
M.Liebsch, X.Pérez-Costa.Utilization of the IEEE802.11 Power Save Mode with IP Paging, In Proceedings of IEEE International Conference on Communications (ICC), Seoul, Korea, May 2005. abstract pdf link
Mobile communication systems increasingly adopt
Internet Protocol solutions for transport of control and data
traffic. To optimize scalability of the mobile communication infrastructure
as well as to save scarce radio bandwidth and mobile
energy resources, mobile devices can enter a power save mode and
refrain from sending superfluous location information towards
the network in case the device is idle. Such systems utilize paging
to locate and reactivate mobile devices in power save mode, as
well as to initiate re-establishment of routing information in the
network. The well accepted wireless LAN standard IEEE802.11
supports mechanisms for power saving while a mobile terminal
is associated with a particular access point. But paging control
of mobile devices in power save mode beyond the scope of
a single access point is not part of the standard IEEE802.11
application. In this paper, we propose a mechanism to integrate
and utilize existing IEEE802.11 power save mode with an IP
paging architecture and protocol without the need to modify
the IEEE802.11 standard. In addition to the IP paging protocol
operation, the proposal covers an appropriate addressing and
identification scheme, which can be utilized for other access
technologies as well. An analytical evaluation of energy and
paging delay costs allows estimation of the proposed mechanism’s
efficiency, which might be sub-optimal for high load conditions,
but appropriate to support migration scenarios towards future
heterogeneous access mobile communication networks.
R.Schmitz, H.Hartenstein, T.Melia, X. Pérez-Costa and W.Effelsberg. The Impact of Wireless Radio Fluctuations on Ad Hoc Network Performance (Der Einfluss von Schwankungen der Uebertragungsreichweite auf die Leistungsfaehigkeit von Ad Hoc Netzwerken). In Proceedings of the 18th DFN Workshop on Communication Networks (DFN-Arbeitstagung über Kommunikationsnetze). Düsseldorf, Germany, June 2004. abstract pdf link
Die Leistungsfaehigkeit eines drahtlosen, mobilen Ad-Hoc Netzwerkes wird im wesentlichen durch die Faehigkeit der Protokolle bestimmt, auf Aenderungen der Netzwerktopologie effektiv zu reagieren. Die fuer diese Netzwerke charakteristische Eigenschaft der Dynamik in der Netztopologie basiert im wesentlichen auf der Mobilitaet der Knoten und auf Schwankungen der Funkreichweite. Waehrend der Mobilitaetsaspekt bereits umfassend untersucht wurde, werden die Eigenschaften des drahtlosen Mediums in Modellen meist idealisiert mittels konstantem Funkradius evaluiert. Solche Ergebnisse werden den tatsaechlichen Begebenheiten, wie Erfahrungen in Feldtests belegen, nicht gerecht.In dieser Arbeit untersuchen wir den Einfluss von Mobilitaet, von zeitabhaengigen Schwankungen der Uebertragungsreichweite und der Kombination beider Faktoren auf die Netzwerk-Leistungsmerkmale Topologieaenderungsrate (Topological Change Rate, TCR) und Linkstabilitaet. Die temporaeren Fluktuationen des Funkradius werden mittels eines Gauss-Markov Prozesses erster Ordnung modelliert, waehrend die Auswirkungen der Mobilitaet basierend auf dem Random Waypoint Mobilitaetsmodell untersucht werden. Die Ergebnisse unserer vergleichenden Simulationsstudie zeigen, dass Fluktuationen der Uebertragungsreichweite zusaetzlich zur Mobilitaet die Leistungsmetriken eines Ad-Hoc Netzwerkes erheblich beeinflussen. Daher empfehlen wir, Fluktuationen der Radiowellenausbreitung bei dem Entwurf von Ad-Hoc Protokollen und Ad-Hoc Modellierung nicht zu vernachlaessigen.
X.Pérez-Costa, A.Banchs, J.Noguera and S.Sallent-Ribes. Optimal Radio Acess Bearer Configuration for Voice over IP in 3G UMTS networks. In Proceedings of the 5th European Wireless (EW), Barcelona, February 2004. abstract pdf link
The Radio Access Bearer (RAB) is the entity responsible
for transporting radio frames of an application over
the access network in UMTS. The parameters of a RAB, namely
the maximum bandwidth and the allowed frame sizes, can be
configured according to the requirements of the application using
it. In this paper we take up the issue of identifying the optimal
RAB configuration for VoIP applications. For this purpose, we
have developed our own simulator, which models accurately all
the aspects that have an impact onto the RAB configuration,
and have evaluated the voice quality resulting from different
RAB configurations. Based on the results obtained, we propose
two new RAB configurations for VoIP, one when Robust Header
Compression (RoHC) is used and another when it is not used.
M.Torrent-Moreno, X.Pérez Costa and S.Sallent-Ribes. A Performance Study of Fast Handovers for Mobile IPv6. In Proceedings of the IEEE Conference on Local Computer Networks (LCN), Bonn, Germany, October 2003. abstract pdf link
We conducted a simulative evaluation of the overall performance
of Fast Handovers for Mobile IPv6 in comparison
with the baseline Mobile IPv6 using the network simulator
ns-2 for a ‘hot spot’ deployment scenario. The simulation
scenario comprises four access routers and up to 50 mobile
nodes that move randomly and communicate in accordance
with the IEEE 802.11 wireless LAN standard. The study
provides quantitative results of the QoS improvements obtained
by FMIPv6 with respect to handoff latency, packet
loss rate and bandwidth per station. The simulation environment
allowed us also to investigate the behavior of the
protocol in extreme cases, e.g. under channel saturation
conditions and considering different traffic sources: CBR,
VoIP, Video and TCP transfers. As a complementary part
of the study, the signaling load costs associated to the performance
improvements provided by the enhancement proposal
was analysed. While some simulation results corroborate
the intention of the protocol specification, other results
give insights not easily gained without performing simulations.
X.Pérez-Costa, H.Hartenstein, and C.Bettstetter. Towards a Mobility Metric for Reproducible and Comparable Results in Ad Hoc Networks Research. In Proceedings of ACM annual international conference on Mobile Computing and Networking (MOBICOM), poster session, San Diego, USA, September 2003. abstract pdf poster link
The performance of a mobile ad hoc network depends on
its ability to adapt to changes in the network topology. It is
therefore important to understand the relation between node
mobility and topology dynamics. While the input parameters
of a mobility model allow for generating different simulation
scenarios (e.g., by selecting the speed, pause time,
system area, and number of nodes), the relation to the actually
generated degree of topology dynamics is usually not
immediate. Thus, a translation of the mobility model to the
resulting topology dynamics is required. A natural way to do
this is to introduce the topology change rate (TCR), defined
as the number of link changes per time unit as observed by a
single node. If we measure the network performance versus
the TCR instead of versus ‘incidental’ input parameters of
a mobility model, a higher level of abstraction of the results
will be achieved, and the reproducability and comparability
of the results will be enhanced.
This short paper derives in an analytical manner the TCR
in a scenario where nodes move according to the random
waypoint (RWP) model [5], being the most heavily used mobility
model in reseach on ad hoc networks. Recent insights
into RWP mobility have shown some unexpected behavior,
e.g., with respect to the spatial node distribution [2] and the
average velocity over time [8]. These results do not mean
that RWP as a mobility model is ‘invalid’, but show the
importance of proper use due to thorough understanding of
the model. This contribution now provides the link between
RWP mobility and the corresponding TCR.
X.Pérez-Costa and M.Torrent-Moreno. A Performance Study of Hierarchical Mobile IPv6 from a System Perspective. In Proceedings of the IEEE International conference on Communications (ICC), Anchorage, Alaska, May 2003. abstract pdf link
We performed a simulative evaluation of standard Mobile
IPv6 in comparison with Hierarchical MIPv6 via ns-2 for a ‘hot spot deployment’
scenario. The simulation scenario comprises four access routers
and up to 30 mobile nodes that move randomly and communicate in accordance
with the IEEE 802.11 wireless LAN standard. The study collected the
performance metrics of all mobile nodes from the system. As data traffic
video, VoIP and TCP sources were considered. The goal of the study was
to obtain quantitative results of the improvements provided by HMIPv6
with respect to handoff latency, packet loss, signaling load and bandwidth
per station as well as an indication of the number of users that could be
accommodated depending on the traffic source. Moreover, we performed
a ’stress-test’ of the protocol to investigate the behavior of the protocol in
extreme cases, e.g. under channel saturation conditions. In addition to the
quantitative results provided, the simulations taught us insights on the protocol
performance not easily gained without performing simulations. E.g.,
we learned that i) in our scenario a low HMIPv6 signaling load reduction
outside of the micro-mobility domain implies a significant increase within
it, ii) under high saturation conditions we can expect a better performance
of HMIPv6 in latency terms but not in packet losses or bandwidth and iii)
the consideration of network coverage user unawareness impacts the performance
results.
X.Pérez-Costa, M.Torrent-Moreno and H.Hartenstein. A Simulation Study on the Performance of Hierarchical Mobile IPv6. In Proceedings of the International Teletraffic Congress (ITC), Berlin, Germany, August 2003. abstract pdf link
We performed a simulative evaluation of Hierarchical MIPv6 in comparison with standard
MIPv6 using the network simulator ns-2 for a ‘hot spot deployment’ scenario. The simulation
scenario comprises four access routers and up to 50 mobile nodes that move randomly and
communicate in accordance with the IEEE 802.11 wireless LAN standard. The study provides
quantitative results of the improvements provided by HMIPv6 with respect to handoff latency,
packet loss, signaling load and bandwidth per station. The simulation environment allowed us
also to investigate the behavior of the protocol in extreme cases, e.g., under channel saturation
conditions, and considering different traffic sources: CBR, Video, VoIP and TCP.
A.Banchs, X.Pérez-Costa, D.Qiao. Providing Throughput Guarantees in IEEE 802.11e Wireless LANs. In Proceedings of the International Teletraffic Congress (ITC), Berlin, Germany, August 2003. abstract pdf link
In this paper, we propose ARME (Assured Rate
MAC Extension), an extension of the IEEE 802.11 MAC protocol
to provide throughput guarantees. The proposed extension relies
on the Distributed Coordination Function (DCF) with a modified
algorithm for the computation of the Contention Window (CW).
Best Effort service (with no throughput guarantee) is supported
by the functionality of the current 802.11 standard in such a way
that legacy IEEE 802.11 terminals behave as Best Effort terminals
in ARME. The performance of the proposed extension has been
extensively evaluated through simulation; simulation results show
that IEEE 802.11 devices using ARME behave well for different
types of traffic and different source rates.
C.Bettstetter, H.Hartenstein, and X.Pérez-Costa. Stochastic Properties of the Random Waypoint Mobility Model: Epoch Length, Direction Distribution, and Cell Change Rate. In Proceedings of ACM MobiCOM Workshop on Modeling, Analysis and Simulation of Wireless and Mobile Systems (MSWiM), Atlanta, USA, September 2002). abstract pdf-preprint link
The random waypoint model is a commonly used mobility model for simulations of wireless communication networks. In this paper, we present analytical derivations of some fundamental stochastic properties of this model with respect to: (a) the length and duration of a movement epoch, (b) the chosen direction angle at the beginning of a movement epoch, and (c) the cell change rate of the random waypoint mobility model when used within the context of cellular networks. Our results and methods can be used to compare the random waypoint model with other mobility models. The results on the movement epoch duration as well as on the cell change rate enable us to make a statement about the 'degree of mobility' of a certain simulation scenario. The direction distribution explains in an analytical manner the effect that nodes tend to move back to the middle of the system area.
M.Liebsch, X.Pérez, R.Schmitz, A.Sarma, J.Jähnert, S.Tessier, M.Wetterwald, I.Soto.Solutions for IPv6-based Mobility in the EU Project Moby Dick. In Proceedings of the World Telecommunications Congress (WTC), Paris, September 2002. abstract pdf link
Mobile Internet technology is moving towards
a packet-based or, more precisely, IPv6-based network.
Current solutions on Mobile IPv6 and other related QoS
and AAA matters do not offer the security and quality that
users have come to take for granted. The EU IST project
Moby Dick has taken on the challenge of providing a
solution that integrates QoS, mobility and AAA in a
heterogeneous access environment. This paper focuses on
the mobility part of the project, describes and justifies the
handover approach taken, shows how QoS-aware and
secure handover is achieved, and introduces the project’s
paging concept. It shows that a transition to a fully
integrated IP-RAN and IP-Backbone has become a
distinct option for the future.
X.Pérez-Costa, R.Schmitz, H.Hartenstein, M.Liebsch. A MIPv6, FMIPv6 and HMIPv6 Handover Latency Study: Analytical Approach. In Proceedings of the IST Mobile & Wireless Telecommunications Summit (IST Summit), Thessaloniki, Greece, June 2002. abstract pdf link
Enhancement of basic Mobile IPv6 as IP based mobility
management with respect to handover latency it’s being
studied within the framework of the EU IST project
Moby Dick. This analytical study compares the
handover latency of different IP mobility management
schemes, considering the different approaches and
combinations of them, as currently discussed within the
IETF: basic MIPv6, FMIPv6 as support for fast
handover and HMIPv6 as an approach allowing
hierarchies of mobility agents. This study has been
performed to assess the most appropriate approach for
the functional specification and the implementation,
specially with respect to the implementation for the
project’s field trial.
A.Banchs, M.Radimirsch, and Xavier Pérez. Assured and Expedited Forwarding Extensions for IEEE 802.11 Wireless LAN. In Proceedings of the tenth IEEE/IFIP International Workshop on Quality of Service (IWQoS), Miami Beach, USA, May 2002. abstract pdf link
In this paper, we propose DIME (DIffServ MAC Extension),
an extension of the IEEE 802.11MAC protocol to support
Differentiated Services. The proposed extension consists of two optional
modules: the Expedited Forwarding (EF) and the Assured
Forwarding (AF). The Expedited Forwarding extension (DIMEEF)
reuses the Interframe space of the Point Coordination Function
(PCF) of the IEEE 802.11 standard in a distributed manner,
while the Assured Forwarding extension (DIME-AF) relies on the
Distributed Coordination Function (DCF) with a modi£ed algorithm
for the computation of the Contention Window (CW). Best
Effort is supported by the functionality of the current 802.11 standard
in such a way that legacy IEEE 802.11 terminals behave as
Best Effort terminals in the DIME architecture. While the performance
of the Assured Forwarding extension has been thoroughly
evaluated by the authors elsewhere [1], this paper concentrates on
the overall architecture and the performance of the Expedited Forwarding
extension.
A. Banchs, X. Pérez. Distributed Weighted Fair Queuing in 802.11 Wireless LAN. In proceedings of the IEEE International Conference on Communications (ICC), New York, USA, April 2002. abstract pdf link
With Weighted Fair Queuing, the link’s bandwidth
is distributed among competing flows proportionally to their
weights. In this paper we propose an extension of the DCF function
of IEEE 802.11 to provide weighted fair queuing in Wireless
LAN. Simulation results show that the proposed scheme is able
to provide the desired bandwidth distribution independent of the
flows aggressiveness and their willingness to transmit. Backwards
compatibility is provided such that legacy IEEE 802.11 terminals
receive a bandwidth corresponding to the default weight.
A.Banchs, X. Pérez. Providing Throughput Guarantees in IEEE 802.11 Wireless LAN. In Proceedings of the IEEE Wireless Communications and Networking Conference (WCNC), Orlando, USA, March 2002. abstract pdf link
In this paper, we propose ARME (Assured Rate
MAC Extension), an extension of the IEEE 802.11 MAC protocol
to provide throughput guarantees. The proposed extension relies
on the Distributed Coordination Function (DCF) with a modified
algorithm for the computation of the Contention Window (CW).
Best Effort service (with no throughput guarantee) is supported
by the functionality of the current 802.11 standard in such a way
that legacy IEEE 802.11 terminals behave as Best Effort terminals
in ARME. The performance of the proposed extension has been
extensively evaluated through simulation; simulation results show
that IEEE 802.11 devices using ARME behave well for different
types of traffic and different source rates.
A. Banchs, X. Pérez, M. Radimirsch, H. Stuettgen. Service Differentiation Extensions for Elastic and Real-Time Traffic in 802.11 Wireless LAN. In Proceedings of the IEEE Conference on High Performance Switching and Routing (HPSR), Dallas, USA, May 2001. abstract pdf link
QoS in wireless networks has a special relevance due to the scarce bandwidth available in such networks. This contribution addresses this issue by extending the MAC protocol of the IEEE 802.11 wireless LAN standard. The extension is divided in two steps. In the first step, real-time traffic is distinguished from elastic traffic by a priority scheduling approach in order to meet the delay requirements of, for example, voice communication. In the second step, service differentiation is introduced for elastic traffic, based on a relative differentiation model. In this model, a high priority service always receives a higher throughput than a low priority one. The proposed architecture has been validated via simulation. Results for real-time traffic show that the proposed approach leads to sufficiently low delays if admission control is properly applied. Elastic traffic achieves the desired differentiation in all simulated scenarios.
A.Banchs, X.Pérez, M.Radimirsch, and S.Sallent. Service Differentiation Extensions for IEEE 802.11. In 11th Workshop on Local and Metropolitan Area Networks (LANMAN), Boulder, USA, March 2001. abstract pdf
The world of data communication has undergone many changes over the last few years. Probably
the most important one is the convergence of voice, video and data communication under the roof
of the Internet Protocol (IP) suite. Originally, IP was designed to support elastic services, i.e. data
applications like file transfer, electronic mail and remote terminal. Elastic services are tolerant of
delays and, even though they benefit from increasing data rates in terms of user satisfaction, still work
at low data rates. Voice services, in contrast, require a certain minimum rate and suffer significantly
from high delay and delay variation. In wired networks, the delay issue can be solved by (1) bandwidth
overprovisioning and (2) service differentiation. Bandwidth overprovisioning is not possible in radio
networks and, thus, innovative solutions for service differentiation are necessary. This contribution
addresses this issue by extending the MAC protocol of the IEEE 802.11 wireless LAN standard.
The extension is divided in two steps. In the first step, real-time services are distinguished from
elastic services by a priority scheduling approach in order to meet the hard requirements of e.g. voice
communication. In the second step, service differentiation is introduced for elastic services. Elastic
service differentiation can be on a per-service basis, e.g. to differentiate between an interactive service
like web access and a non-interactive one like electronic mail, on a per-user basis, e.g. to differentiate
services for managers and other employees in a company, or even on a per-device basis, e.g. for
different sensors in a production facility. The fact that the two-step extension proposed is designed in
different and independent modules gives the manufacturer the option to omit one of them, therewith
simplifying the migration effort from the current IEEE 802.11 standard.
The real-time extension is a distributed scheme which redefines the Point Coordination Function
(PCF) of the current IEEE 802.11 standard. The original PCF is not widely supported in current
products, and the only requirement of our solution is that the original PCF must not be used in
a network together with the extension presented here. We propose a scheme to resolve contention
among real-time stations based on two Elimination Bursts. This scheme provides a constant residual
collision rate almost independent from the number of contending stations. Admission control is a key
aspect of the real-time extension. In order to meet the requirement of real-time traffic for low delay,
the amount of traffic using this service must be kept sufficiently low. In addition, admission control
for real-time traffic avoids the starvation of lower priority traffic.
Applying admission control, however, does not match the nature of elastic services. Elastic services
do not need a specific capacity but rather as much capacity as possible. Therefore, we propose a
relative differentiation model in which a high priority service always receives a higher throughput
than a low priority one. Due to the relative nature of the scheme, admission control can be omitted
and differentiation can always be achieved, independent of the incoming load.
The elastic service differentiation model is based on the notion of a share. The share reflects the
priority assigned to a service such that the throughput experienced by the service is proportional to
the share assigned to it. As an example, assume eight services, six with a share of 1 and two with a
share of 2, then the total of shares is 10. Each of the two high priority services should get 20% of the
total throughput and the remaining six should get 10% each.
We propose an algorithm that dynamically computes the size of the Contention Window (CW) of
the IEEE 802.11 DCF mode. With this algorithm, each station computes its relative throughput W
which is equal to its measured transmission rate divided by the share assigned to its service. This
value is included in the header of each transmitted packet. Whenever a station observes a packet
from another station with a W different from its own W, this station modifies its own CW by a small
amount such that the difference between its own W and the observed W is reduced. In this way,
the W of all the stations converge toward a common value and the desired differentiation is achieved.