Master’s thesis is dedicated to the issue of network element. In the first part are described technological aspects as the internal structure of network elements and control mechanisms that provide differentiaton of services and support their quality requirements. In the first chapter are listed control mechanisms from most simple like FIFO to the more sophisticated like CBWFQ. Active queue managements are not missing and they are described in the section Riadiaci člen. Significant role in the differentiation of services have processes marking and pakets classification. The second chapter is devoted most promising standard DiffServ. Contains section devoted to architecture, paket marking, code point DSCP and the final part is devoted to the types of per-hop behavior and it is expedited forwarding EF and technique assured forwarding AF. The third chapter gives an overview common models of neural networks, their properties and assess their suitability for deployment in network elements. The routher design itself and its structure in programming environment Matlab/Simulink is the subject of the fourth chapter. Except used bloks from Simulink library is described their setting and function in the wiring. Consequently results are reviewied and conclusions drawn.

Wireless networks are undergoing an unprecedented revolution in the last decade. With the explosion of delay-sensitive applications in the Internet (i.e., online gaming and VoIP), latency becomes a major issue for the development of wireless technology. Taking advantage of the significant decline in memory prices, industrialists equip the network devices with larger buffering capacities to improve the network throughput by limiting packets drops. Over-buffering results in increasing the time that packets spend in the queues and, thus, introducing more latency in networks. This phenomenon is known as “bufferbloat”. While throughput is the dominant performance metric, latency also has a huge impact on user experience not only for real-time applications but also for common applications like web browsing, which is sensitive to latencies in order of hundreds of milliseconds. Concerns have arisen about designing sophisticated queue management schemes to mitigate the effects of such phenomenon. My thesis research aims to solve bufferbloat problem in both traditional half-duplex and cutting-edge full-duplex wireless systems by reducing delay while maximizing wireless links utilization and fairness. Our work shed lights on buffer management algorithms behavior in wireless networks and their ability to reduce latency resulting from excessive queuing delays inside oversized static network buffers without a significant loss in other network metrics. First of all, we address the problem of buffer management in wireless full-duplex networks by using Wireless Queue Management (WQM), which is an active queue management technique for wireless networks. Our solution is based on Relay Full-Duplex MAC (RFD-MAC), an asynchronous media access control protocol designed for relay full-duplexing. Compared to the default case, our solution reduces the end-to-end delay by two orders of magnitude while achieving similar throughput in most of the cases. In the second part of this thesis, we propose a novel design called “LearnQueue” based on reinforcement learning that can effectively control the latency in wireless networks. LearnQueue adapts quickly and intelligently to changes in the wireless environment using a sophisticated reward structure. Testbed results prove that LearnQueue can guarantee low latency while preserving throughput.

Current congestion control approaches that attempt to provide fair bandwidth allocation among competing flows primarily consider only data rate when making decisions on which packets to drop. However, responsive flows with high round trip times (RTTs) can still receive significantly less bandwidth than responsive flows with low round trip times. This paper proposes a congestion control scheme called WHITE that addresses router unfairness in handling flows with significantly different RTTs. Using a best-case estimate of a flow's RTT provided in each packet by the flow source or by an edge router, WHITE computes a stabilized average RTT. The average RTT is then compared with the RTT of each incoming packet, dynamically adjusting the drop probability so as to protect the bandwidth of flows with high RTTs while curtailing the bandwidth of flows with low RTTs. We present simulation results and analysis that demonstrate that WHITE provides better fairness than other rate-based congestion control strategies over a wide-range of traffic conditions. The improved fairness of WHITE comes close to the fairness of Fair Queuing without requiring per flow state information at the router.

Improving network performance in terms of efficiency, fairness in the bandwidth, and system stability has been a research issue for decades. Current Internet traffic control maintains sophistication in end TCPs but simplicity in routers. In each router, incoming packets queue up in a buffer for transmission until the buffer is full, and then the packets are dropped. This router queue management strategy is referred to as Drop Tail. End TCPs eventually detect packet losses and slow down their sending rates to ease congestion in the network. This way, the aggregate sending rate converges to the network capacity. In the past, Drop Tail has been adopted in most routers in the Internet due to its simplicity of implementation and practicability with light traffic loads. However Drop Tail, with heavy-loaded traffic, causes not only high loss rate and low network throughput, but also long packet delay and lengthy congestion conditions. To address these problems, active queue management (AQM) has been proposed with the idea of proactively and selectively dropping packets before an output buffer is full. The essence of AQM is to drop packets in such a way that the congestion avoidance strategy of TCP works most effectively. Significant efforts in developing AQM have been made since random early detection (RED), the first prominent AQM other than Drop Tail, was introduced in 1993. Although various AQMs also tend to improve fairness in bandwidth among flows, the vulnerability of short-lived flows persists due to the conservative nature of TCP. It has been revealed that short-lived flows take up traffic with a relatively small percentage of bytes but in a large number of flows. From the user’s point of view, there is an expectation of timely delivery of short-lived flows. Our approach is to apply artificial intelligence technologies, particularly fuzzy logic (FL), to address these two issues: an effective AQM scheme, and preferential treatment for short-lived flows. Inspired by the success of FL in the robust control of nonlinear complex systems, our hypothesis is that the Internet is one of the most complex systems and FL can be applied to it. First of all, state of the art AQM schemes outperform Drop Tail, but their performance is not consistent under different network scenarios. Research reveals that this inconsistency is due to the selection of congestion indicators. Most existing AQM schemes are reliant on queue length, input rate, and extreme events occurring in the routers, such as a full queue and an empty queue. This drawback might be overcome by introducing an indicator which takes account of not only input traffic but also queue occupancy for early congestion notification. The congestion indicator chosen in this research is traffic load factor. Traffic load factor is in fact dimensionless and thus independent of link capacity, and also it is easy to use in more complex networks where different traffic classes coexist. The traffic load indicator is a descriptive measure of the complex communication network, and is well suited for use in FL control theory. Based on the traffic load indicator, AQM using FL – or FLAQM – is explored and two FLAQM algorithms are proposed. Secondly, a mice and elephants (ME) strategy is proposed for addressing the problem of the vulnerability of short-lived flows. The idea behind ME is to treat short-lived flows preferably over bulk flows. ME’s operational location is chosen at user premise gateways, where surplus processing resources are available compared to other places. By giving absolute priority to short-lived flows, both short and long-lived flows can benefit. One problem with ME is starvation of elephants or long-lived flows. This issue is addressed by dynamically adjusting the threshold distinguishing between mice and elephants with the guarantee that minimum capacity is maintained for elephants. The method used to dynamically adjust the threshold is to apply FL. FLAQM is deployed to control the elephant queue with consideration of capacity usage of mice packets. In addition, flow states in a ME router are periodically updated to maintain the data storage. The application of the traffic load factor for early congestion notification and the ME strategy have been evaluated via extensive experimental simulations with a range of traffic load conditions. The results show that the proposed two FLAQM algorithms outperform some well-known AQM schemes in all the investigated network circumstances in terms of both user-centric measures and network-centric measures. The ME strategy, with the use of FLAQM to control long-lived flow queues, improves not only the performance of short-lived flows but also the overall performance of the network without disadvantaging long-lived flows.

Ce mémoire ne contient pas de résumé.

No description available.

Access to the large web content hosted all over the world by users of the Internet engage many hosts, routers/switches and faster links. They challenge the internet backbone to operate at its capacity to assure e±cient content access. This may result in congestion and raises concerns over various Quality of Service (QoS) issues like high delays, high packet loss and low throughput of the system for various Internet applications. Thus, there is a need to develop effective congestion control mechanisms in order to meet various Quality of Service (QoS) related performance parameters. In this paper, our emphasis is on the Active Queue Management (AQM) mechanisms, particularly Random Early Detection (RED). We propose a threshold based novel analytical model based on standard RED mechanism. Various numerical examples are presented for Internet traffic scenarios containing both the burstiness and correlation properties of the network traffic.

Depuis quelques années le trafic de l'internet mobile ne cesse d'augmenter. Cette croissance soutenue est liée à plusieurs facteurs, parmi lesquels l'évolution des terminaux, la grande diversité des services et des applications disponibles et le déploiement des nouvelles technologies d'accès radio mobile (3G/4G). À cet égard, le standard 3GPP pour les réseaux LTE propose une architecture offrant une gestion fine de la QoS (par flux). Ce modèle, hérité des réseaux mobiles traditionnels orientés connexion, soulève des problèmes en termes de scalabilité, efficacité et performances. Les travaux entrepris dans cette thèse ont pour objectif principal de proposer des solutions plus simples et moins coûteuses pour la gestion de la QoS dans les réseaux mobiles. À cette fin, à l'issue d'une étude et de l'évaluation de l'impact de la signalisation associée au modèle de QoS standard, deux modèles alternatifs ont été proposés. Nous proposons tout d'abord un modèle basée sur les mécanismes IP inspiré de l'approche DiffServ (par agrégat) largement étudié dans les réseaux IP fixes. Ce modèle fournit une gestion de la QoS simple, efficiente et rentable, tout en garantissant des performances équivalentes au modèle standard. Cependant, elle nécessite une remise à niveau de tous les eNB, et donc une longue phase de transition. En conséquence, nous proposons SloMo qui vise à améliorer l'expérience des clients mobiles, mais avec un objectif de déploiement plus rapide. SloMo est une solution de gestion implicite de la QoS depuis un point unique situé sur le chemin des communications. SloMo exploite la dynamique instaurée par le mécanisme de contrôle de flux de TCP. Il vise à recréer un goulot d'étranglement dynamique dans un équipement contrôlé par l'opérateur lorsque les points de congestion réels ne sont pas accessibles. Une fois ce goulot d'étranglement déporté, il est alors aisé d'effectuer une gestion de la qualité IP classique dans l'équipement supportant Slo-Mo. / The mobile data landscape is changing rapidly and mobile operators are today facing the daunting challenge of providing cheap and valuable services to ever more demanding customers. As a consequence, cost reduction is actively sought by operators as well as Quality of Service (QoS) preservation. Current 3GPP standards for LTE/EPC networks offer a fine tuning QoS (per-flow level), which inherits many characteristics of legacy telco networks. In spite of its good performance, such a QoS model reveals costly and cumbersome and finally, it remains very rarely deployed, thereby giving way to basic best-effort hegemony. This thesis aims at improving QoS in mobile networks through cost-effective solutions; To this end, after an evaluation of the impact and cost of signaling associated with the standard QoS model, alternative schemes are proposed, such as the IP-centric QoS model (per aggregate) inspired from the DiffServ approach widely used in fixed IP networks. This model provides a simple, efficient and cost-effective IP level QoS management with a performance level similar to standardized solutions. However, as it requires enhancements in the eNB, this scheme cannot be expected in mobile networks before a rather long time.Thus, we introduce Slo-Mo, which is a lightweight implicit mechanism for managing QoS from a distant point when the congestion point (e.g. eNB) is not able to do it. Slo-Mo creates a self-adaptive bottleneck which adjusts dynamically to the available resources taking advantage of TCP native flow control. Straightforward QoS management at IP level is then performed in the Slo-Mo node, leading to enhanced customer experience at a marginal cost and short term.

Traditionally the Internet is used for the following applications: FTP, e-mail and Web traffic. However in the recent years the Internet is increasingly supporting emerging applications such as IP telephony, video conferencing and online games. These new applications have different requirements in terms of throughput and delay than traditional applications. For example, interactive multimedia applications, unlike traditional applications, have more strict delay constraints and less strict loss constraints. Unfortunately, the current Internet offers only a best-effort service to all applications without any consideration to the applications specific requirements. In this thesis three existing Active Queue Management (AQM) mechanisms are modified by incorporating into these a control function to condition routers for better Quality of Service (QoS). Specifically, delay is considered as the key QoS metric as it is the most important metric for real-time multimedia applications. The first modified mechanism is Drop Tail (DT), which is a simple mechanism in comparison with most AQM schemes. A dynamic threshold has been added to DT in order to maintain packet queueing delay at a specified value. The modified mechanism is referred to as Adaptive Drop Tail (ADT). The second mechanism considered is Early Random Drop (ERD) and, iii in a similar way to ADT, a dynamic threshold has been used to keep the delay at a required value, the main difference being that packets are now dropped probabilistically before the queue reaches full capacity. This mechanism is referred to as Adaptive Early Random Drop (AERD). The final mechanism considered is motivated by the well known Random Early Detection AQM mechanism and is effectively a multi-threshold version of AERD in which packets are dropped with a linear function between the two thresholds and the second threshold is moveable in order to change the slope of the dropping function. This mechanism is called Multi Threshold Adaptive Early Random Drop (MTAERD) and is used in a similar way to the other mechanisms to maintain delay around a specified level. The main focus with all the mechanisms is on queueing delay, which is a significant component of end-to-end delay, and also on reducing the jitter (delay variation) A control algorithm is developed using an analytical model that specifies the delay as a function of the queue threshold position and this function has been used in a simulation to adjust the threshold to an effective value to maintain the delay around a specified value as the packet arrival rate changes over time. iv A two state Markov Modulated Poisson Process is used as the arrival process to each of the three systems to introduce burstiness and correlation of the packet inter-arrival times and to present sudden changes in the arrival process as might be encountered when TCP is used as the transport protocol and step changes the size of its congestion window. In the investigations it is assumed the traffic source is a mixture of TCP and UDP traffic and that the mechanisms conserved apply to the TCP based data. It is also assumed that this consists of the majority proportion of the total traffic so that the control mechanisms have a significant effect on controlling the overall delay. The three mechanisms are evaluated using a Java framework and results are presented showing the amount of improvement in QoS that can be achieved by the mechanisms over their non-adaptive counterparts. The mechanisms are also compared with each other and conclusions drawn.

Cette thèse s'inscrit dans une thématique pluridisciplinaire explorant les liens existants entre la théorie de la commande et les réseaux informatiques. L'idée consiste à appliquer les outils de l'Automatique pour la stabilisation du trafic dans les réseaux de communication. Premièrement, nous nous sommes intéressés à l'analyse de stabilité des systèmes à retards variables au travers de deux approches temporelles. D'une part, nous avons considéré la méthode de Lyapunov-Krasovskii dans laquelle nous avons élaboré des fonctionnelles en adéquation avec de nouvelles modélisations du système (segmentation du retard, dérivée temporelle). D'autre part, la stabilité a également été abordée avec une approche entrée-sortie, empruntant alors les outils de l'analyse robuste. Le système à retard est alors réécrit comme l'interconnexion d'une application linéaire avec une matrice constituée d'opérateurs définissant le système original. Après avoir revisité le principe de séparation quadratique, nous développons des opérateurs auxiliaires afin de caractériser au mieux la dynamique retardée et proposer des critères moins pessimistes. Deuxièmement, la méthodologie développée est ensuite utilisée pour le problème de contrôle de congestion d'un routeur lors de communications TCP. Ce protocole de bout en bout est sensible à la perte de paquet et modifie en conséquence son taux d'émission selon l'algorithme du AIMD. Il s'agit alors de commander le taux de perte par l'intermédiaire d'un mécanisme d'Active Queue Management situé au niveau du routeur afin de réguler le trafic. Les résultats théoriques sont ensuite évalués à l'aide du simulateur de réseaux NS-2.