Bernd E. Wolfinger

Joint Adoption of QoS Schemes for MPEG Streams

Abstract.Indiscriminated packet discards strongly degrade the quality perceived by end users of MPEG video transmissions. This paper investigates different Quality of Service (QoS) schemes and the tradeoffs of jointly adopting such schemes to improve the delivery quality of an MPEG stream. From an analytical model, we evaluate the impact of frame losses on the quality of MPEG streams and on the waste of network resources. Our assessment considers issues such as the use of redundancy by applying a Forward Error Correction (FEC) scheme to tolerate losses, the changing of the compression factor in MPEG encoding, the unequal protection of MPEG frames in a Differentiated Services environment, and how to evaluate the impact of network losses onto application quality. Results provide predicted bounds on the quality to be expected by end users as well as guidelines on how to take the best advantage from the joint adoption of the investigated QoS schemes.

A learning automata-based fault-tolerant routing algorithm for mobile ad hoc networks

Reliable routing of packets in a Mobile Ad Hoc Network (MANET) has always been a major concern. The open medium and the susceptibility of the nodes of being fault-prone make the design of protocols for these networks a challenging task. The faults in these networks, which occur either due to the failure of nodes or due to reorganization, can eventuate to packet loss. Such losses degrade the performance of the routing protocols running on them. In this paper, we propose a routing algorithm, named as learning automata based fault-tolerant routing algorithm (LAFTRA), which is capable of routing in the presence of faulty nodes in MANETs using multipath routing. We have used the theory of Learning Automata (LA) for optimizing the selection of paths, reducing the overhead in the network, and for learning about the faulty nodes present in the network. The proposed algorithm can be juxtaposed to any existing routing protocol in a MANET. The results of simulation of our protocol using network simulator 2 (ns-2) shows the increase in packet delivery ratio and decrease in overhead compared to the existing protocols. The proposed protocol gains an edge over FTAR, E2FT by nearly 2% and by more than 10% when compared with AODV in terms of packet delivery ratio with nearly 30% faulty nodes in the network. The overhead generated by our protocol is lesser by 1% as compared to FTAR and by nearly 17% as compared to E2FT when there are nearly 30% faulty nodes.

Validated Model-Based Performance Prediction of Multi-Core Software Routers

The possibility of using flexible and cost-efficient commodity hardware instead of expensive custom hardware has triggered wide interest in software routers. Performance measurement and simulation are important approaches for identifying bottlenecks of such systems to predict and improve the performance. We measure the performance of software routers using current multi-core hardware architectures. We introduce an innovative and validated node model for intra-node resource contention and realized a resource management extension for the widely-used network simulator ns-3 which allows to evaluate and predict the performance of current and future software router architectures.

A study of networking software induced latency

For long time, high-speed packet processing has been reserved for specialized hardware devices since software based solutions were not able to achieve the required performance. However, off-the-shelf packet processing hardware and software improved over the last years, which is why software based solutions cope with high-speed traffic nowadays. Due to the flexibility of software there is a trend towards doing packet processing in software, e.g. using OpenFlow or virtual switches. Although packet processing in software offers many capabilities, the complexity of such software bases solutions makes it hard to evaluate, optimize, or predict the networking performance of servers, end user hosts, or routers. We present a study that investigates the packet latency caused by the packet processing in the Linux network stack. We develop a simulation model in ns-3 for packet processing via the Linux network stack that helps understanding of its performance implications. We validate our simulation model based on measurements with nanosecond accuracy and software profiling.

Decreasing call blocking probability of Broadband TV services by a channel access control scheme

The current trend of network convergence has been best exemplified by the recent emergence of new services such as Broadband TV (BTV) offered by Telcos around the world. A very relevant research topic is to improve the quality of experience (QoE). This paper focuses on decreasing the user call blocking probability (CBP), which is one of the most important BTV QoE metrics. At first, we analyze a currently typical BTV network architecture. And then, a state-vector-based simulation model is proposed to evaluate the CBP (on a single link). After that, we present a class-based distribution model of channel watching probabilities, and based on this distribution model, we elaborate a general TV channel access control (TCAC) scheme in order to reduce the CBP. This scheme is flexible and can be applicable in many different scenarios. Finally, a case study is carried out, and it demonstrates the potential of this TCAC scheme to improve the BTV QoE by decreasing the CBP on a single link.

On the potential of FEC algorithms in building fault-tolerant distributed applications to support high QoS video communications

Studies for distributed applications have demonstrated that application quality in distributed computing is strongly dependent on the qualityof the underlyingcommunicationsystem.Besidea performance characteristics the reliability properties of a communication system are playing a predominant role with regard to the qurdky achievable from applications’ point of view. In this paper we study distributed applications with video communication via 10SSYcommunication systems. We develop analytical models reflecting transmission of MPEG coded video streams both without and with forward error correction (FEC). These models allow us to determine the optimum amountof redundancyto be used for PEC when communicating via constant bit rate connections. We conclude that FEC algorithms offer the potentiat for high qrsalityof-service even in the case of using a communication system with low quality.

A unified load generator based on formal load specification and load transformation

Optimizing the performance of existing networks often demands realistic and valid load generation tools. As the network loads observed may vary significantly over time, there is a strong need to develop flexible and extensible load generators which can reproduce these loads accurately. This paper presents a unified methodology for artificial load generating at different interfaces within the protocol hierarchy of computer networks. A unified load generator which is based on a formal load specification technique and a new approach to load modeling combined with load transformation is established. The methodology of constructing the unified load generator is validated by means of implementing a prototype of such a load generator (UniLoG) which is capable of producing network traffic with different characteristics in a very simple and flexible manner. We present the architectural design, some implementation aspects and investigate the limits of applicability of our prototype. Recent performance evaluation studies with UniLoG indicate that this load generator is a valid tool for QoS studies in computer networks, and can indeed produce some predefined traffic accurately.

Unicast versus multicast for live TV delivery in networks with tree topology

Multicasting is an essential technology in applications like IPTV or online multiplayer gaming and has been studied extensively in different networks. In this article, we introduce a new performance criterion, called multicast gain, which can be used for computing multicast efficiency. We analyze multicast transmission for delivery of live TV channels in tree-based networks representing, e.g., multicast trees, and determine the gain of subtrees and links in order to quantify the benefits of multicasting. We also define a threshold in order to identify cases in which unicasting might be preferable. We show that although multicasting can usually outperform multiple unicasting in terms of bandwidth usage, due to the inherent overhead in establishing and maintaining multicast connections, in special cases, using multiple unicast flows can be a better choice.

A learning automata-based uplink scheduler for supporting real-time multimedia interactive traffic in IEEE 802.16 WiMAX networks

Worldwide interoperability for microwave access (WiMAX) is one of the most challenging forthcoming technologies providing broadband wireless access (BWA) which promises its users high Quality of Service (QoS) for supporting interactive real-time multimedia services. One of the most important enablers for this is the MAC layer scheduling. For WiMAX, no scheduling algorithm has been standardized and has been left open for the vendors to implement as per their own needs, thereby facilitating the differentiation between their products. Some of the most important metrics for an efficient scheduling algorithm are throughput, delay, jitter (time variation of delay), and packet loss. In this paper, we propose a learning automata (LA)-based scheduling algorithm for WiMAX uplink real-time multimedia interactive traffic in a point to multipoint (PMP) architecture. The main objective of the work presented in the paper is to present the design of an efficient algorithm for providing QoS to real-time uplink traffic. The results of simulation based performance evaluation studies show that the proposed algorithm results in an increase of throughput and decrease in delay and packet loss for real-time service flows as compared to other non-real-time traffic.

Towards an automatic modeling tool for observed system behavior

Current computer systems and communication networks tend to be highly complex, and they typically hide their internal structure from their users. Thus, for selected aspects of capacity planning, overload control and related applications, it is useful to have a method allowing one to find good and relatively simple approximations for the observed system behavior. This paper investigates one such approach where we attempt to represent the latter by adequately selecting the parameters of a set of queueing models. We identify a limited number of queueing models that we use as Building Blocks in our procedure. The selected Building Blocks allow us to accurately approximate the measured behavior of a range of different systems. We propose an approach for selecting and combining suitable Building Blocks, as well as for their calibration. We are able to successfully validate our methodology for a number of case studies. Finally, we discuss the potential and the limitations of the proposed approach.