MPNT (Multicast Proxies NeTwork) is an overlay architecture that provides multicast services to unicast or multicast clients. It is multimedia-oriented, as it implements RTP, RTCP and SIP protocols. Nevertheless, it can be used to broadcast any type of data. MPNT is proxy-based, i.e. its nodes are not end-users but proxies: they act like relays for end-users, providing them standard unicast or multicast access. For scalability and adaptivity reasons, these proxies can be interconnected using different topologies. The first implemented topology is spanning tree based. It is very efficient in terms of switching delay (no routing) but the global delay variance can be important. To remedy this, we implemented an alternate hypercube-based topology. Indeed, hypercubes have several interesting properties that ensure a better scalability. Nevertheless, this logical hypercube optimization is valid provided that interconnecting links are comparable. To adapt the architecture to the network heterogeneity, we implemented a hybrid topology exploiting both spanning tree and hypercube advantages. The objective of this paper is to compare these topologies using HYRPAN, a MPNT-specific network simulator we developed for this purpose.
{"title":"Comparative study of MPNT topological models","authors":"Karim Sbata, R. Benaini, P. Vincent","doi":"10.1109/ICW.2005.37","DOIUrl":"https://doi.org/10.1109/ICW.2005.37","url":null,"abstract":"MPNT (Multicast Proxies NeTwork) is an overlay architecture that provides multicast services to unicast or multicast clients. It is multimedia-oriented, as it implements RTP, RTCP and SIP protocols. Nevertheless, it can be used to broadcast any type of data. MPNT is proxy-based, i.e. its nodes are not end-users but proxies: they act like relays for end-users, providing them standard unicast or multicast access. For scalability and adaptivity reasons, these proxies can be interconnected using different topologies. The first implemented topology is spanning tree based. It is very efficient in terms of switching delay (no routing) but the global delay variance can be important. To remedy this, we implemented an alternate hypercube-based topology. Indeed, hypercubes have several interesting properties that ensure a better scalability. Nevertheless, this logical hypercube optimization is valid provided that interconnecting links are comparable. To adapt the architecture to the network heterogeneity, we implemented a hybrid topology exploiting both spanning tree and hypercube advantages. The objective of this paper is to compare these topologies using HYRPAN, a MPNT-specific network simulator we developed for this purpose.","PeriodicalId":255955,"journal":{"name":"2005 Systems Communications (ICW'05, ICHSN'05, ICMCS'05, SENET'05)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115598642","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this paper, we present a performance assessment of one-way resource reservation protocols in optical burst switched (OBS) mesh networks. The performance analysis considers five resource reservation protocols, Just-In-Time (JIT), JumpStart, JIT+, Just-Enough-Time (JET) and Horizon, and focuses on the following topologies: rings, degree-three chordal rings, degree-four chordal rings, degree-five chordal rings, degree-six chordal rings, mesh-torus, NSFNET, ARPANET and the European Optical Network (EON). It is shown that when the nodal degree increases from 2 to around 3, the largest gain is observed for degree-three chordal rings (slightly less than three orders of magnitude) and the smallest gain is observed for the ARPANET (less than one order of magnitude). On the other hand, when the nodal degree increases from 2 to around 4, the largest gain is observed for degree-four chordal rings (with a gain between four and five orders of magnitude) and the smallest gain is observed for the EON (with a gain less than one order of magnitude). When the nodal degree increases from 2 to around 5 or 6, the gain is between four and six orders of magnitude. These results clearly show the importance of the way links are connected in OBS networks, since, in this kind of networks, burst loss probability is a key issue. Moreover, the performance of the five protocols is very close for those topologies.
{"title":"One-way resource reservation protocols for IP over optical burst switched mesh networks","authors":"J. Rodrigues, M. Freire, P. Lorenz","doi":"10.1109/ICW.2005.62","DOIUrl":"https://doi.org/10.1109/ICW.2005.62","url":null,"abstract":"In this paper, we present a performance assessment of one-way resource reservation protocols in optical burst switched (OBS) mesh networks. The performance analysis considers five resource reservation protocols, Just-In-Time (JIT), JumpStart, JIT+, Just-Enough-Time (JET) and Horizon, and focuses on the following topologies: rings, degree-three chordal rings, degree-four chordal rings, degree-five chordal rings, degree-six chordal rings, mesh-torus, NSFNET, ARPANET and the European Optical Network (EON). It is shown that when the nodal degree increases from 2 to around 3, the largest gain is observed for degree-three chordal rings (slightly less than three orders of magnitude) and the smallest gain is observed for the ARPANET (less than one order of magnitude). On the other hand, when the nodal degree increases from 2 to around 4, the largest gain is observed for degree-four chordal rings (with a gain between four and five orders of magnitude) and the smallest gain is observed for the EON (with a gain less than one order of magnitude). When the nodal degree increases from 2 to around 5 or 6, the gain is between four and six orders of magnitude. These results clearly show the importance of the way links are connected in OBS networks, since, in this kind of networks, burst loss probability is a key issue. Moreover, the performance of the five protocols is very close for those topologies.","PeriodicalId":255955,"journal":{"name":"2005 Systems Communications (ICW'05, ICHSN'05, ICMCS'05, SENET'05)","volume":"64 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125868319","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper provides an overview of methods for multimedia performance assessments during the deployment of UMTS networks. It is shown how performance assessments can be embedded in the deployment process. Investigations are based on reviews of the radio planning and system parameter settings given by the network operator To gain additional insight into specific areas of interest, these are complemented by measurement campaigns employing onsite drive tests. Several metrics characterizing radio networks and their service quality are emphasized for different multimedia scenarios, including voice, data and video applications. Examples and general observations of typical performance assessments obtained from numerous projects with different network operators are addressed.
{"title":"Multimedia performance assessments in deployed UMTS networks","authors":"R. Weber","doi":"10.1109/ICW.2005.59","DOIUrl":"https://doi.org/10.1109/ICW.2005.59","url":null,"abstract":"This paper provides an overview of methods for multimedia performance assessments during the deployment of UMTS networks. It is shown how performance assessments can be embedded in the deployment process. Investigations are based on reviews of the radio planning and system parameter settings given by the network operator To gain additional insight into specific areas of interest, these are complemented by measurement campaigns employing onsite drive tests. Several metrics characterizing radio networks and their service quality are emphasized for different multimedia scenarios, including voice, data and video applications. Examples and general observations of typical performance assessments obtained from numerous projects with different network operators are addressed.","PeriodicalId":255955,"journal":{"name":"2005 Systems Communications (ICW'05, ICHSN'05, ICMCS'05, SENET'05)","volume":"202 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123034367","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
L. Isaksson, S. Chevul, M. Fiedler, J. Karlsson, P. Lindberg
Motivated by the central role and the importance of network throughput for the performance of net-worked applications and thus for user experience, this paper investigates the process of user-perceived throughput. This throughput is measured on rather small time scales and interpreted with aid of summary statistics, histograms and autocorrelation coefficients. The results including GPRS, UMTS and WLAN measurements reveal a clear influence of the network, seen from variations of application-perceived throughput on the one-second time scale, which has to be considered when choosing the right kind of network for a specific task. The different statistical metrics considered will later be used in our research to maintain the QoS of the service decided by the user. While theoretical throughputs are yielded as expected in UMTS, both GPRS and WLAN do not reach the promised throughput values.
{"title":"Application-perceived throughput process in wireless systems","authors":"L. Isaksson, S. Chevul, M. Fiedler, J. Karlsson, P. Lindberg","doi":"10.1109/ICW.2005.28","DOIUrl":"https://doi.org/10.1109/ICW.2005.28","url":null,"abstract":"Motivated by the central role and the importance of network throughput for the performance of net-worked applications and thus for user experience, this paper investigates the process of user-perceived throughput. This throughput is measured on rather small time scales and interpreted with aid of summary statistics, histograms and autocorrelation coefficients. The results including GPRS, UMTS and WLAN measurements reveal a clear influence of the network, seen from variations of application-perceived throughput on the one-second time scale, which has to be considered when choosing the right kind of network for a specific task. The different statistical metrics considered will later be used in our research to maintain the QoS of the service decided by the user. While theoretical throughputs are yielded as expected in UMTS, both GPRS and WLAN do not reach the promised throughput values.","PeriodicalId":255955,"journal":{"name":"2005 Systems Communications (ICW'05, ICHSN'05, ICMCS'05, SENET'05)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124631279","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Applications in mobile ad hoc network are growing rapidly. But these applications have maximum utility only when mobiles can be used anywhere at anytime. However, one of the greatest limitations of such networks, is finite power of batteries. In order, to maximize the battery life of the nodes, we must minimize their energy consumption. The energy can be managed in various levels. Here, we propose an implementable energy-saving initialization protocol. It is the result of a combination of an average case analysis and a randomized approach to initialize ad hoc networks. Its implementation shows an effective gain in terms of energy saving.
{"title":"An energy-saving algorithm for the initialization of single hop mobile ad hoc networks","authors":"D. Karimou, J. Myoupo","doi":"10.1109/ICW.2005.26","DOIUrl":"https://doi.org/10.1109/ICW.2005.26","url":null,"abstract":"Applications in mobile ad hoc network are growing rapidly. But these applications have maximum utility only when mobiles can be used anywhere at anytime. However, one of the greatest limitations of such networks, is finite power of batteries. In order, to maximize the battery life of the nodes, we must minimize their energy consumption. The energy can be managed in various levels. Here, we propose an implementable energy-saving initialization protocol. It is the result of a combination of an average case analysis and a randomized approach to initialize ad hoc networks. Its implementation shows an effective gain in terms of energy saving.","PeriodicalId":255955,"journal":{"name":"2005 Systems Communications (ICW'05, ICHSN'05, ICMCS'05, SENET'05)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130741606","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Distributed Hash table (DHT) based peer-to-peer (P2P) algorithms are very promising for their efficient routing performance. However, most commercial P2P systems do not adapt DHT algorithms due to reasons like routing information maintenance overhead, considerable network traffic and increased P2P system complexity in a highly dynamic environment. The dynamics of peer participation (churn) is an inherent property of P2P systems that should be incorporated in the design and evaluation of DHT-based structured P2P overlay networks. In this paper, we study the effect of variable churn on DHT updation and lookup latency due to non-uniform bandwidth nodes randomly joining and leaving the P2P network. We further study the effect of non-uniform node bandwidth in handling churn related DHT communications, flash crowd requests and study the tradeoff between bandwidth and latency for evaluating DHT performance during variable churn. We employ a modified Pareto distribution to determine probability of node liveness based on its current lifetime in the P2P network.
{"title":"Bandwidth and latency model for DHT based peer-to-peer networks under variable churn","authors":"Praveen Kumar, G. Sridhar, V. Sridhar","doi":"10.1109/ICW.2005.31","DOIUrl":"https://doi.org/10.1109/ICW.2005.31","url":null,"abstract":"Distributed Hash table (DHT) based peer-to-peer (P2P) algorithms are very promising for their efficient routing performance. However, most commercial P2P systems do not adapt DHT algorithms due to reasons like routing information maintenance overhead, considerable network traffic and increased P2P system complexity in a highly dynamic environment. The dynamics of peer participation (churn) is an inherent property of P2P systems that should be incorporated in the design and evaluation of DHT-based structured P2P overlay networks. In this paper, we study the effect of variable churn on DHT updation and lookup latency due to non-uniform bandwidth nodes randomly joining and leaving the P2P network. We further study the effect of non-uniform node bandwidth in handling churn related DHT communications, flash crowd requests and study the tradeoff between bandwidth and latency for evaluating DHT performance during variable churn. We employ a modified Pareto distribution to determine probability of node liveness based on its current lifetime in the P2P network.","PeriodicalId":255955,"journal":{"name":"2005 Systems Communications (ICW'05, ICHSN'05, ICMCS'05, SENET'05)","volume":"208 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115742880","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
For several years, researchers in the Internet pricing research area have been proposing various pricing mechanisms. To analyze and evaluate their proposals, they usually make use of analytical based approaches with different assumptions on the environment (sometimes simplistic assumptions); this makes it difficult, or rather impossible, to compare proposals using the same realistic assumptions on the environment. Sometimes, they use also simulations to evaluate their proposals; however, these simulations are very specific and cannot be used for other proposals. We believe that there is a need for a simulation framework that can be used to evaluate different pricing schemes. Such a framework will en able the evaluation of Internet pricing schemes using the same realistic environment (e.g., network and traffic). In this paper, we present the. design and implementation of an Internet pricing simulator using J-Sim, a java-based network simulation tool. The objective of the framework is to ease the process of evaluating Internet pricing schemes. For a proof of concept, we evaluate a pricing scheme, we developed in the past, using the proposed framework.
{"title":"A simulation framework for evaluating Internet pricing mechanisms","authors":"Y. Khene, A. Hafid, N. Agoulmine","doi":"10.1109/ICW.2005.17","DOIUrl":"https://doi.org/10.1109/ICW.2005.17","url":null,"abstract":"For several years, researchers in the Internet pricing research area have been proposing various pricing mechanisms. To analyze and evaluate their proposals, they usually make use of analytical based approaches with different assumptions on the environment (sometimes simplistic assumptions); this makes it difficult, or rather impossible, to compare proposals using the same realistic assumptions on the environment. Sometimes, they use also simulations to evaluate their proposals; however, these simulations are very specific and cannot be used for other proposals. We believe that there is a need for a simulation framework that can be used to evaluate different pricing schemes. Such a framework will en able the evaluation of Internet pricing schemes using the same realistic environment (e.g., network and traffic). In this paper, we present the. design and implementation of an Internet pricing simulator using J-Sim, a java-based network simulation tool. The objective of the framework is to ease the process of evaluating Internet pricing schemes. For a proof of concept, we evaluate a pricing scheme, we developed in the past, using the proposed framework.","PeriodicalId":255955,"journal":{"name":"2005 Systems Communications (ICW'05, ICHSN'05, ICMCS'05, SENET'05)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124507773","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
An important problem in WDM network design is to construct a logical topology and determine an optimal routing over that topology. Mixed integer linear program (MILP) formulations to generate optimal solutions for this problem are computationally intractable, even for moderate sized networks. A standard approach is to decouple the problem of logical topology design and the problem of routing on this logical topology. Heuristics for finding the logical topology exist and a straightforward linear program (LP), based on the node-arc formulation is normally used to solve the routing problem over a given logical topology. Such LP formulations become computationally infeasible for large networks. In this paper, we present a new formulation for optimally routing traffic over a given logical topology. Our formulation is based on the arc-chain representation and minimizes the congestion of the network. We have used an implicit column generation technique and exploited the special generalized upper bounding structure of our formulation to efficiently solve the problem.
{"title":"An efficient algorithm for routing in WDM networks","authors":"Y. Aneja, S. Bandyopadhyay, A. Jaekel, S. Dey","doi":"10.1109/ICW.2005.23","DOIUrl":"https://doi.org/10.1109/ICW.2005.23","url":null,"abstract":"An important problem in WDM network design is to construct a logical topology and determine an optimal routing over that topology. Mixed integer linear program (MILP) formulations to generate optimal solutions for this problem are computationally intractable, even for moderate sized networks. A standard approach is to decouple the problem of logical topology design and the problem of routing on this logical topology. Heuristics for finding the logical topology exist and a straightforward linear program (LP), based on the node-arc formulation is normally used to solve the routing problem over a given logical topology. Such LP formulations become computationally infeasible for large networks. In this paper, we present a new formulation for optimally routing traffic over a given logical topology. Our formulation is based on the arc-chain representation and minimizes the congestion of the network. We have used an implicit column generation technique and exploited the special generalized upper bounding structure of our formulation to efficiently solve the problem.","PeriodicalId":255955,"journal":{"name":"2005 Systems Communications (ICW'05, ICHSN'05, ICMCS'05, SENET'05)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126821226","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this paper, we present an analytical study targeted at statistically capturing the loss behavior of a RED queue. We utilize a finite-state Markov chain model. Starting from recursive equations of the model, we derive equivalent closed-form equations. We numerically validate the matching of recursive and closed-form equations. Further, we apply our model to monitor the average RED queue size in a number of sample topologies illustrating their practicality. Based on our results, we argue that our model can adapt to the changing network conditions.
{"title":"A finite-state Markov chain model for statistical loss across a RED queue","authors":"Mohit B. Singh, H. Yousefi’zadeh, H. Jafarkhani","doi":"10.1109/ICW.2005.7","DOIUrl":"https://doi.org/10.1109/ICW.2005.7","url":null,"abstract":"In this paper, we present an analytical study targeted at statistically capturing the loss behavior of a RED queue. We utilize a finite-state Markov chain model. Starting from recursive equations of the model, we derive equivalent closed-form equations. We numerically validate the matching of recursive and closed-form equations. Further, we apply our model to monitor the average RED queue size in a number of sample topologies illustrating their practicality. Based on our results, we argue that our model can adapt to the changing network conditions.","PeriodicalId":255955,"journal":{"name":"2005 Systems Communications (ICW'05, ICHSN'05, ICMCS'05, SENET'05)","volume":"106 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122679450","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
M. Khadraoui, B. Hirsbrunner, F. Meinkohn, D. Khadraoui, M. Courant
The main purpose of the research and development activities presented in this paper is related to the content development techniques for the ITV. It deals with the development of a real-time immersive and interactive TV show based on a DVB-MHP technology. It is basically a TV quiz that can be played from home thanks to virtual avatars that simulate the behaviors of the players. In fact, the immersive and interactive dimensions require a new gaming concept that renders the viewers as active and participative actors. This work has been achieved under RTI-TVS ESA (European Space Agency) project
{"title":"Interactive TV Show Based on Avatars","authors":"M. Khadraoui, B. Hirsbrunner, F. Meinkohn, D. Khadraoui, M. Courant","doi":"10.1109/ICW.2005.53","DOIUrl":"https://doi.org/10.1109/ICW.2005.53","url":null,"abstract":"The main purpose of the research and development activities presented in this paper is related to the content development techniques for the ITV. It deals with the development of a real-time immersive and interactive TV show based on a DVB-MHP technology. It is basically a TV quiz that can be played from home thanks to virtual avatars that simulate the behaviors of the players. In fact, the immersive and interactive dimensions require a new gaming concept that renders the viewers as active and participative actors. This work has been achieved under RTI-TVS ESA (European Space Agency) project","PeriodicalId":255955,"journal":{"name":"2005 Systems Communications (ICW'05, ICHSN'05, ICMCS'05, SENET'05)","volume":"235 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133466578","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：