-
[show abstract]
[hide abstract]
ABSTRACT: The Intelligent Network (IN) architecture promises easy development and fast deployment of sophisticated communication services.
In this paper, we present the results of our research in the area of Intelligent Multimedia Services. One of the main results
has been the evaluation of novel multimedia services operating over different network platforms. One of the new services being
designed is a Multimedia News-on-Demand application allowing users to browse through items of current affairs and select a
specific article for viewing. This service is supplemented by an Intelligent Multimedia Gateway providing interConnectivity
between the public service telephone network and the Internet. The study of this service has proved useful in gaining insight
on the way novel intelligent multimedia services across network platforms will benefit from the use of IN technology. A performance
evaluation of the service has also proved valuable to identify the major system and traffic parameters to consider when designing
and deploying IN multimedia services across multiple networks.
06/2006: pages 250-264;
-
[show abstract]
[hide abstract]
ABSTRACT: More efficient transmission of coded speech over packet switching systems can be achieved by channel coding. The redundancy introduced by channel coding to recover lost or impaired packets results in an increased traffic and an increased queuing loss. Hence there is a critical tradeoff between channel coding rate and queuing loss. We show that the channel loss rate, the traffic level and congestion rate affect the optimum channel coding rate.
Communications, 2002. ICC 2002. IEEE International Conference on; 02/2002
-
[show abstract]
[hide abstract]
ABSTRACT: With the rapid development of network technology, deployment of real-time applications such as video, voice and virtual reality over network become more popular than ever. However, the best effort model, existing over the Internet cannot satisfy the QoS requirements of real-time applications. In order to make Internet quality of service (QoS) capable, the IETF proposed a series of protocols and algorithms, some of them forming what is known as "differentiated services." In this paper, we focus on analyzing the performance of real-time applications over a Diffserv-capable network, especially when real-time traffic aggregation and link congestion case happen in the core router. Based on extensive testing results we propose a novel scheduling scheme, which can solve the problem generated by EF traffic aggregation and link congestion. Based on our findings, we conclude that with proper resource management and traffic engineering, differentiated services can provide satisfactory QoS to real-time applications with strict delay, delay variation, packet loss and throughput constraints.
Electrical and Computer Engineering, 2002. IEEE CCECE 2002. Canadian Conference on; 02/2002
-
[show abstract]
[hide abstract]
ABSTRACT: The cost minimization of virtual private networks (VPN) that use the resources of an underlying transport network is the key factor for their successful implementation. The investigation in this paper is focused on 'network based VPNs', where the operation of the VPN is outsourced to an Internet Service Provider (ISP). The interest in such solutions is generated both by customers seeking to reduce support costs and by ISPs seeking new revenue sources. Solving the cost minimization would allow Internet service providers (ISPs) to define and deploy new VPN services. The basic building block of VPN is the tunnel. A tunnel operates as an overlay across the backbone, and the traffic sent through the tunnel is opaque to the underlying backbone. A VPN end point can terminate multiple tunnels or forward packets between different tunnels. Different tunnels can share the same physical link and traffic belonging to the same VPN tunnel can be carried along different physical links. The multiplexing and management of the VPN tunnels is made possible by core routers supporting the VPN of the underlying network. The novelty of the presented work is the network flow model of VPN mapping on the underlying ISP network. We assume that the VPN topology, the topology of the ISP network and the total utilization cost for all underlying links are known parameters. Based on this, we propose a network management system based on a network flow optimization in order to define the minimal cost link allocation for the VPN tunnels. We provide a simulation of the proposed optimal establishment of VPN tunnels and performance evaluation of the simulation results.
Electrical and Computer Engineering, 2002. IEEE CCECE 2002. Canadian Conference on; 02/2002
-
[show abstract]
[hide abstract]
ABSTRACT: Digital subscriber line (DSL) technology is being widely used to provide high-speed Internet access over the local loop. Nowadays, efforts are going on to efficiently integrate voice and data services over DSL. One of the main challenges remains the design and development of simple scheduling schemes able to guarantee the QoS requirements of voice and data services over the access segment. This paper describes and evaluates various traffic scheduling mechanisms designed specifically for the integration of voice and data services over DSL. A comparative evaluation of the various traffic scheduling algorithms is done in terms of delay and bandwidth efficiency. Our results show that an algorithm that prioritizes the voice traffic over data traffic while ensuring a minimum bandwidth reservation for data traffic can efficiently integrate voice and data.
Electrical and Computer Engineering, 2002. IEEE CCECE 2002. Canadian Conference on; 02/2002
-
[show abstract]
[hide abstract]
ABSTRACT: The ATM ABR service category was initially designed to support non-real-time applications, e.g., data file transfer. However it is possible for ABR to support real-time applications (such as distributed multimedia applications) by taking advantage of the inherent traffic control capability of ABR and the adaptation ability of real-time applications. This paper consists of two parts. In the first part we review the principles of operation of the FASTRAC, an ABR control mechanism. This analysis allows us to better understand the adaptation ability of real-time applications, a feature deemed essential to take advantage of the inherent traffic control capability of ABR while guaranteeing a reasonable QoS required by the end application. In the second part we explore the use of ABR service to support real-time video transmission using FASTRAC. Our simulation results show that FASTRAC can support both real-time video applications and non real-time applications.
Electrical and Computer Engineering, 2002. IEEE CCECE 2002. Canadian Conference on; 02/2002
-
[show abstract]
[hide abstract]
ABSTRACT: Transmission of G.729A coded speech over packet switching networks
including forward error correction is considered. Three buffer
management policies are applied: simple, random and privileged. The
privileged policy plays an important role to achieve quasi-uniform
packet loss among multiple users. Its effect on spreading the loss among
various sources and in avoiding consecutive or closely timed losses for
any source is described. In this paper, the subjective quality
improvement by correctly adapting the loss pattern to the sensitivity of
the G.729A algorithm is described
Communications, 2001. ICC 2001. IEEE International Conference on; 02/2001
-
[show abstract]
[hide abstract]
ABSTRACT: The growth in cellular networks has motivated research on
broadband access technologies for the support of diverse services over
wireless networks. Wireless asynchronous transfer mode (WATM) has been
proposed as solution for next generation wireless and personal
communication networks and ATM Adaptation Layer 2 (AAL2) has been
identified suitable for the support of low data rate and delay sensitive
voice traffic generated in cellular networks. We study the performance
of AAL2 over a burst error wireless channel. Analytical expressions for
performance measures of interest are derived for the Gilbert-Elliott
(1960, 1963) channel and numerical results are presented to demonstrate
the effect of burst errors on the AAL2 frame structure and the AAL2
performance
Electrical and Computer Engineering, 2001. Canadian Conference on; 02/2001
-
[show abstract]
[hide abstract]
ABSTRACT: A traffic prediction algorithm is evaluated. It assumes that
network traffic is realistically modeled using alpha-stable long-range
dependent stochastic processes. This work analyzes the sensitivity of
the prediction algorithm with respect to inaccuracies introduced when
extracting the model parameters
Communications, Computers and signal Processing, 2001. PACRIM. 2001 IEEE Pacific Rim Conference on; 02/2001
-
[show abstract]
[hide abstract]
ABSTRACT: Reliable prediction of network traffic allows for the
implementation of more efficient resource management schemes. In a
previous work, reported by some of the same authors of this paper, a
novel algorithm for linear prediction of network traffic was introduced
and evaluated. That algorithm assumed that traffic statistics can be
modeled using alpha-stable long-range-dependent stochastic processes.
The relevant prediction algorithm was based on the minimum dispersion
criterion, whose resulting equations were solved in a
processing-efficient but approximate manner. More recent work has proved
that in most of the cases the coefficients so obtained produce a robust
and acceptable performance. Nevertheless, further studies suggest that
the accuracy of the linear prediction can be enhanced if needed. This
work identifies where this can be done, proposes some optimization
procedures and provides some numerical examples. Our results show that,
when incorporating optimization, the gain in performance is quite
remarkable for network traffic exhibiting strong long-range dependence
Info-tech and Info-net, 2001. Proceedings. ICII 2001 - Beijing. 2001 International Conferences on; 02/2001
-
[show abstract]
[hide abstract]
ABSTRACT: Wireless asynchronous transfer mode (WATM) has been proposed as a transport solution for the next generation of wireless and personal communication networks. ATM adaptation layer 2 (AAL2) has been identified as suitable for the support of low data rate and delay sensitive voice traffic generated in those cellular networks. Wireless ATM is of particular interest for future tactical networks, which will employ ATM technology in the wide area, such as the NATO POST-2000 tactical communications systems. AAL2 as defined by ITU-T standards does not perform very efficiently over wireless links because of the limited protection against errors and because of the way AAL2 packets are delineated within the ATM payload. A novel delineation mechanism is proposed for AAL2 to be used over high BER channels, such as wireless links, while preserving interoperability with standard AAL2.
Military Communications Conference, 2001. MILCOM 2001. Communications for Network-Centric Operations: Creating the Information Force. IEEE; 02/2001
-
[show abstract]
[hide abstract]
ABSTRACT: We study the performance requirements for the transport of MPEG
video streams over ATM networks. A control scheme is introduced to limit
the propagation of impairments through MPEG-based video sequences as a
result of cell losses. The scheme assigns different slice sizes to the
video frames based on their type and order of appearance within the
group of pictures. The main aim of the proposed scheme is to improve the
robustness of the video delivery process at a minimum cost. Our results
demonstrate the effectiveness of this scheme in reducing the propagation
of impairments due to the loss of information during the transmission of
MPEG video streams over ATM networks
IEEE Transactions on Broadcasting 07/2000; · 1.70 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: Personal communications services (PCS) key objective is to provide
ubiquitous communication services. The success of PCS will depend on the
network resources to ensure the quality-of-service required by service
subscribers. Performance evaluation studies of different system
configurations and implementation alternatives will play a major role in
the planning and dimensioning of successful PCS networks. We undertake
the analysis of some major implementation issues of a PCS architecture
based on the IN system architecture. Different system configurations and
traffic scenarios are considered in the study. The overall system
performance is evaluated in terms of utilization rate of the key network
elements and the end-to-end response time of the signaling network for
various IN services
Communications, 1999. ICC '99. 1999 IEEE International Conference on; 02/1999
-
[show abstract]
[hide abstract]
ABSTRACT: A technique for the fast simulation of broadband communications systems is proposed, which is based on regenerative importance sampling techniques. Our algorithm is applicable to estimating the probability of rare events when modeling the offered traffic using fractional stable noise (FSN) processes (including fractional Brownian noise as a particular case), which have been proved to be able to capture both the long-range dependence and the burstiness of today's aggregate network traffic. An exact description of FSN processes is given, as well as an approximation that allows for the application of importance sampling techniques. The results obtained for a simple example are also included
Simulation Conference Proceedings, 1999 Winter; 02/1999
-
[show abstract]
[hide abstract]
ABSTRACT: The connection oriented data service (CO service) has been defined
by the IEEE 802.6 committee to support variable-bit-rate (VBR) data
services over DQDB MANs. The main aim of the CO service is to be able to
guarantee the QoS required by a wide variety of VBR services. We derive
an adaptive bandwidth allocation control mechanism for the DQDB CO data
service. Simulation results demonstrate the effectiveness of the
proposed mechanism in providing the guarantees on the throughput
required by the applications under changing network conditions
Communications, Computers and Signal Processing, 1999 IEEE Pacific Rim Conference on; 02/1999
-
[show abstract]
[hide abstract]
ABSTRACT: The connection-oriented data service has been previously introduced in the IEEE 802.6 standard to support variable bit rate applications for which a certain guarantee on quality of service is required. So far, the proper tuning of the parameters of this service remains an open issue. In this work, we undertake the study of this service with the main objective of defining the needed mechanisms to properly set-up the parameters of the connection oriented service according to various quality of service requirements
Electrical and Computer Engineering, 1998. IEEE Canadian Conference on; 06/1998
-
[show abstract]
[hide abstract]
ABSTRACT: In this study, we have undertaken the modeling and performance
evaluation of a PCS network supporting IN services. We have taken into
account the interconnectivity between the PCS network and the PSTN.
Furthermore, we have accurately modeled the underlying signaling
protocol, namely the SS7 and IS-41. The study has considered additions
to the IS-41 protocols aiming to provide security mechanisms to limit
the fraud present in PCS networks. The results show that when the IS-41
security mechanisms are not implemented, the network is able to support
considerably higher arrival rates of register, mobile call originating
and mobile call delivery service requests that in the case when the
security mechanisms are implemented. The quality of the service provided
by the system has been evaluated in terms of the end-to-end delay
experienced by the users requests for the aforementioned services. The
results also showed that the implementation of security mechanisms in
PCS will require faster databases to accommodate the same capability
achieved by the actual IS-41 implementations (no security). The main
objective has been to stress the important role played by the databases
within the IN architecture
Intelligent Network Workshop, 1998. IN '98. Proceedings., 7th IEEE; 06/1998
-
[show abstract]
[hide abstract]
ABSTRACT: The intelligent network (IN) architecture promises easy development and fast deployment of sophisticated telecommunication services. However, the centralization of service logic programs, call processing database, and other housekeeping functions at the service control point (SCP) node may create some bottleneck situations. The authors present queuing models for the SCP and study the behavior of the IN centered around the SCP by simulation. It is illustrated that such models can help identify bottlenecks and plan the capacity of the SCP. Although the models are not complex, useful insights into the system can be gained. The models can be reused in other studies, on the design and evaluation of congestion control schemes for IN
Intelligent Information Systems, 1997. IIS '97. Proceedings; 01/1998
-
[show abstract]
[hide abstract]
ABSTRACT: ATM switches with multicast and broadcast capabilities are required in order to support B-ISDN multiparty calls and to provide local area network interconnection to B-ISDN. The GAUSS ASE is an ATM switch that uses the broadcast bus architecture but does not provide multicast capability. It has a number of favorable properties, namely the broadcast bus, output queuing, self-routing method and a limited increase in its internal speed. Its architecture can be easily modified to provide multicast and broadcast capabilities. This paper presents the simulation results and analysis of the design modification to the GAUSS switch proposed by Rein J.F. de Vrie (see Proceedings of IEEE GLOBECOM, p.211-17, 1990)
Electrical and Computer Engineering, 1997. IEEE 1997 Canadian Conference on; 06/1997
-
[show abstract]
[hide abstract]
ABSTRACT: In this paper we give a general description of the intelligent
network (IN) and study a service control point (SCP) architecture in
which multiple call processors participate to IN call/service
processing. The SCP is the core physical entity of the IN architecture
and is dedicated to process IN calls under stringent real-time
requirements. We show, by means of simulations, that with the increasing
complexity and amount of IN services stored in local disk-based
databases, a variety of parameters and resources result in poor
performance. The complexity of a service is described as the amount of
service processing logic data required to complete an IN call. We also
present some possible solutions in order to alleviate the poor
performance when local disk-based databases are used
Electrical and Computer Engineering, 1997. Engineering Innovation: Voyage of Discovery. IEEE 1997 Canadian Conference on; 06/1997
