QoS Measures Research Articles

WiMax quality-of-service estimations and measurement

Nowadays WiMax is considered as one of the main technologies for next generation high speed wireless access network. It provides a larger coverage compared to WiFi while supporting string QoS and security mechanism. However, this technology needs fine configurations tuning to achieve its performance, which implementation in different environments is very difficult for operators willing to deploy it. This paper presents QoS estimation methods and measurements over WiMax networks and highlights the complexity of this process. It presents performance measurements in term of achievable rate of an operational WiMax network to evaluate its capability to support different types of media transmission. Finally, it highlights the importance of an accurate time synchronization in the measurement process which is difficult to achieve and therefore proposes a time synchronization method to derive end-to-end delay measurements with appropriate accuracy.

Integrated network experimentation for QoS measurements in opaque MANETs

AbstractIntegrated network experimentation often combines real nodes with simulated ones, each modeling different portions of the topology, in the same experimental run. They enable new validation techniques and larger experiments than obtainable using real elements alone. Integrated experiments can be particularly useful in testing and validating QoS mechanisms for mobile ad hoc networks (MANETs), when the characteristics of the intermediate wireless network segments are not observable from the end segments, and as a consequence these intermediate segments have to be treated as opaque networks; and also when there is a need to conduct experiments in a real MANET environment at a scale larger than just a few nodes. End‐to‐end QoS assurance for such opaque networks, consisting of admission control and quality adjustment, can be based on techniques for dynamically measuring throughput representing the state of these networks. In this paper, we describe a distributed and hybrid testbed that has been deployed for running large‐scale integrated experiments to demonstrate the efficacy of a measurement‐based QoS solution. The infrastructure for the testbed provides an integrated platform consisting of real nodes running the actual software under test, augmented with a simulated network environment. We define a set of metrics and run experiments to evaluate the effectiveness of the QoS solution as well the performance of the deployed testbed. We propose an alternative architecture that employs a Xen‐based virtualization of the real nodes from the deployed testbed. We compare the performances of the virtualized architecture with the deployed architecture vis‐à‐vis latency and resource utilization. Our goal is to establish benchmarks for running large‐scale experiments on performance and QoS measurements in virtualized environments. Copyright © 2009 John Wiley & Sons, Ltd.

On the Quality of Service of Crash-Recovery Failure Detectors

We model the probabilistic behavior of a system comprising a failure detector and a monitored crash-recovery target. We extend failure detectors to take account of failure recovery in the target system. This involves extending QoS measures to include the recovery detection speed and proportion of failures detected. We also extend estimating the parameters of the failure detector to achieve a required QoS to configuring the crash-recovery failure detector. We investigate the impact of the dependability of the monitored process on the QoS of our failure detector. Our analysis indicates that variation in the MTTF and MTTR of the monitored process can have a significant impact on the QoS of our failure detector. Our analysis is supported by simulations that validate our theoretical results.

Design of location areas with QoS considerations in a cellular network

In this paper, we present a location area (LA) planning model which explicitly has more QoS-conscious constraints than any other models. In order to specify an acceptable level of QoS, paging delay time (PDT) and paging blocking rate (PBR) are introduced. The paging process is described by the M/M/c/K queuing system and queuing analysis is carried out to obtain PDT and PBR. Also, in the proposed model, the time-varying call arrival rates are used to define the paging load on each cell, which helps to reflect the pattern of incoming calls more precisely and subsequently to improve the QoS level. To evaluate the proposed model, some examples are solved along with the average rate model and the peak rate model. The results show that the proposed model has the capability to maintain the acceptable level of QoS with the reasonable location management cost, compared to other models. Also, the sensitivity analysis is performed to see the effect of PDT and PBR in the formulation and it turns out that both PBR and PDT are important elements of the QoS measure and influential to the location management cost.

Agile bandwidth management techniques in slotted all-optical packet switched networks

The next generation metro network is most likely to be based on high-capacity agile all-optical networks. We consider a metro network architecture that consists of a number of buffer-less all-optical core switches, where a number of edge switches are connected to each core switch in an overlaid star topology. We would like to design potential candidates of resource-sharing schemes in such a network and evaluate their performances, where they guarantee to deliver optical packets to their destinations. The first scheme is reservation-based, in which decisions are made at each core switch so that collisions (and therefore loss) can be avoided through reservations. For this scheme, we have re-designed the BvN (Birkhoff von Neumann) scheduling algorithm to accommodate the asymmetric multi-fiber/multi-wavelength feature in star OPS. In the second scheme, distributed and independent decisions are made at edge switches, but we need to retransmit the traffic dropped at core switches in the optical domain. We also implement a hybrid case as the third scheme which combines the good attributes of both previous schemes. We have characterized and evaluated the performance of these resource-sharing schemes by various QoS measures such as delay and loss probabilities in edge switches.

Multi-QoS Criteria Based Selection of Web Services by Using AHP Approach

QoS-based Web service ranking problem is complex due to the diverse characteristic of QoS properties and the trade-off issue between them. A ranking mechanism needs to take into account different ways to effectively handle different types of QoS properties measured by different methods with different units and value types, as well as to unify those various QoS measurements in a fair comparison method. In this paper, we present a novel method for ranking and selecting Web services by introducing an extensible QoS model and a corresponding ranking algorithm. The QoS model takes into account the following considerations: different roles and preferences of users and system brokers in specifying QoS demand, different important levels between required QoS constraints and optional QoS preferences, and different QoS value comparison rules with respect to value types and comparative effects of QoS properties. AHP, a multiple criteria decision making method, is applied as an underlying mechanism for developing a flexible and effective ranking algorithm. The experimental results verify that our proposal produces significant achievement with our extensible QoS model and the new service ranking algorithm.

Scheduling algorithms for conducting conflict-free measurements in overlay networks

Network monitoring is essential to the correct and efficient operation of ISP networks and the kind of applications they support, and active measurement is a key design problem in network monitoring. Unfortunately, almost all active probing algorithms ignore the measurement conflict problem: Active measurements conflict with each other - due to the nature of these measurements, the associated overhead, and the network topology - which leads to reporting incorrect results. In this paper, we consider the problem of scheduling periodic QoS measurement tasks in overlay networks. We first show that this problem is NP-hard, and then propose two conflict-aware scheduling algorithms for uniform and non-uniform tasks whose goal is to maximize the number of measurement tasks that can run concurrently, based on a well-known approximation algorithm. We incorporate topological information to devise a topology-aware scheduling algorithm that reduces the number of time slots required to produce a feasible measurement schedule. Also, we study the conflict-causing overlap among overlay paths using various real-life Internet topologies of the two major service carriers in the US. Experiments conducted using the PlanetLab testbed confirm that measurement conflict is a real issue that needs to be addressed. Simulation results show that our algorithms achieve at least 25% better schedulability over an existing algorithm. Finally, we discuss various practical considerations, and identify several interesting research problems in this context.

Multihop Relaying for Broadband Wireless Mesh Networks: From Theory to Practice

We summarize capacity results to show merits of multihop relaying in broadband cellular mesh networks. Under the guidance of these results, we provide design perspectives on relay deployment, spectrum allocation, and end-to-end optimization of certain QoS measures such as throughput, coverage, reliability, and robustness. We conclude with an overview of recent standardization activities and remarks on remaining open problems and design challenges.

Optimum Practical Design of Distributed and Asynchronous Power Control for Wireless Networks with Shared Bands

The present paper proposes two novel and practical schemes for distributed and asynchronous power control in wireless ad hoc networks, in which users dynamically share several frequency bands as in “cognitive radio” networks. These schemes iteratively adjust transmit powers of individual network transmitters with respect to mutually caused interference in the shared bands. Their most attractive feature is that they find network-wide acceptable trade-offs to diverse signal-to-noise and interference (SINR) requirements and efficiently use techniques of stochastic approximation and time-averaging to guarantee a robust performance in random channels. Advantageously, both proposed algorithms do not assume any particular modulation, coding, QoS measure definition or network architecture, which assures their high applicability in the industry and research. Moreover, the broad definition and non-linear nature of these schemes mathematically generalize and thus encompass as a special case many widely deployed power control schemes such as e.g. those for achieving fixed SINR targets or using game-theoretic utility maximization. Simulations are provided to illustrate our approach and its better performance compared to standard algorithms.

A Generalized Random Mobility Model for Wireless Ad Hoc Networks and Its Analysis: One-Dimensional Case

In wireless ad hoc networks, the ability to analytically characterize the spatial distribution of terminals plays a key role in understanding fundamental network QoS measures such as throughput per source to destination pair, probability of successful transmission, and connectivity. Consequently, mobility models that are general enough to capture the major characteristics of a realistic movement profile, and yet are simple enough to formulate its long-run behavior, are highly desirable. We propose a generalized random mobility model capable of capturing several mobility scenarios and give a mathematical framework for its exact analysis over one-dimensional mobility terrains. The model provides the flexibility to capture hotspots where mobiles accumulate with higher probability and spend more time. The selection process of hotspots is random and correlations between the consecutive hotspot decisions are successfully modeled. Furthermore, the times spent at the destinations can be dependent on the location of destination point, the speed of movement can be a function of distance that is being traveled, and the acceleration characteristics of vehicles can be incorporated into the model. Our solution framework formulates the model as a semi-Markov process using a special discretization technique. We provide long-run location and speed distributions by closed-form expressions for one-dimensional regions (e.g., a highway).

MOQ: Web services ontologies for QoS and general quality evaluations

When describing web services, one of the obvious aspects that need representing is 'Quality of Service' (QoS), the capability of a web service to meet an acceptable level of service as per factors such as availability and accessibility. However, too much of a focus on developing functional QoS ontologies has led to an over-emphasis on representing solely QoS metrics and units of measurement. For instance, what does round trip time actually mean? Is the round trip time of every data item measured? Is it an average, or is every nth item measured? Is it the actual measure that is important or just the percentage of items that are beyond a certain range? Arguably existing QoS ontologies cannot readily answer many of these design questions because these questions have less to do with evaluating QoS and more to do with representing 'what is quality?' Therefore, there is an unmet need for web services ontologies that are designed at a higher level encompassing domain independent concepts, and generally applicable beyond QoS evaluations. The MOQ set of ontologies, designed from the premise that quality is 'conformance to requirements', aims to fill this need. Comprised of ontologies of requirement, measurement, traceability, and quality management systems, MOQ can be extended to encompass QoS metrics and measurement units or be designed to interoperate with existing QoS ontology. The use of MOQ minimises the ambiguity in QoS evaluations.

QoS-aware bandwidth provisioning for IP network links

Current bandwidth provisioning procedures for IP network links are mostly based on simple rules of thumb, using coarse traffic measurements made on a time scale of e.g., 5 or 15 min. A crucial question, however, is whether such coarse measurements give any useful insight into the capacity actually needed: QoS degradation experienced by the users is strongly affected by traffic rate fluctuations on a much smaller time scale. The present paper addresses this question. The goal is to develop provisioning procedures that require a minimal measurement effort. The bandwidth provisioning formula that we propose (and which we justify under minimal model assumptions) is of the form ρ + α ρ . Here ρ (in Mbit/s) is the load of the system, which can evidently be estimated by coarse traffic measurements (e.g., 5 or 15 min measurements). The α depends on the characteristics of the individual flows and the QoS requirements. The QoS measure used is the probability that the traffic supply exceeds the available bandwidth, over some predefined (small) interval T, is below some small fixed number ε. The impact of changing the ‘QoS parameters’, i.e., T and ε, on the coefficient α is explicitly given. The validity of the bandwidth provisioning rule is assessed through extensive measurements performed in several operational network environments.

QoS Management Through Adaptive Reservations

Reservation based (RB) scheduling is a class of scheduling algorithms that is well-suited for a large class of soft real-time applications. They are based on a abstraction, meaning that a task is given the illusion of executing on a dedicated slower processor. In this context, a crucial design issue is deciding the bandwidth that each task should receive. The point we advocate is that, in presence of large fluctuations on the computation requirements of the tasks, it can be a beneficial choice to dynamically adapt the bandwidth based on QoS measurements and on the subsequent application of feedback control (adaptive reservations). In this paper, we present two novel contributions to this research area. First, we propose three new control algorithms inspired to the ideas of stochastic control. Second, we present a flexible and modular software architecture for adaptive reservations. An important feature of this architecture is that it is realised by means of a minimally invasive set of modifications to the Linux kernel.

Performance of a new bus assignment algorithm for an ATM switch fabric

The measurement of the quality of service in ATM* networks involves the activation of the ATM layer management OAM functions. Once an ATM connection is set up, in-service performance monitoring based on OAM cells can be used to measure and estimate the ATM layer QoS parameters. In-service measurement of ATM layer QoS parameters such as cell delay variation, cell transfer delay, cell loss ratio, etc., is necessary to verify on one hand, if the network is meeting the requested QoS and on the other hand, if the users are receiving their promised QoS. Congestion conditions in ATM switches may cause gradual deteriorations in the quality of virtual paths or virtual channel connections in an ATM network. QoS measurement using OAM cells can accurately measure the overall quality of monitored ATM layer QoS parameters in order to detect such deteriorations before the service quality has dropped bellow an acceptable level. Several factors such as network load, congestion, bottlenecks, and incremenTS in network load may affect the cell delay variation of video transmissions in an ATM network. In this work, several ATM layer QoS parameters have been measured on a VBR video source using OAM cells. The simulation resulTS obtained show that with low network loads, the contribution of each individual ATM switch to the local and to the end-to-end CDV is not significant. In the presence of a bottleneck switch, the CDV presented by this switch becomes the dominant one at both local and end-to-end sides. When the network load is between 0.5 and 0.6, small incremenTS in network load do not alter the buffer performance of the non-congested ATM switches, and low values of CDV can be obtained. In contrast, it was found that there is a critical load between 0.65 and 0.7 where small incremenTS in network load may force the CDV to grow abruptly.

Probing limitations for packet loss probability measurement on buffered access links

Performance measurement of packet (e.g. managed IP) networks is a vital element in the commercial viability of broadband networking. Active measurement by injection of probing packets is becoming the standard technique for network monitoring and QoS measurement. Results from simulation and queueing analysis were applied to show that probing for packet loss probability through a buffered access link is limited by the mean time until a probe is lost, which can be very long. Hence it is imperative that network operators and service providers are aware of the amount of time required for data collection when monitoring the Service Level Agreements for different network loads.

Predictive and measurement-based dynamic resource management and QoS control for videos

In this paper, we propose a new dynamic resource allocation and call admission control method for the VBR video sources to meet the user's quality of service requirements and at the same time to achieve an efficient resource management in networks. Without any prior knowledge of the user's traffic statistics, the proposed mechanism dynamically adjusts the necessary bandwidth by the networks based on the provided quality of service satisfaction degree of each connection in respect to the user's requirements in terms of loss ratio and average delay. Using traffic prediction and simple QoS measurements, the next required bandwidth for each video connection is computed. To avoid quality decreasing by new incoming calls, we present a call admission control based on the provided QoS for existing connections. Simulation results show that our proposed dynamic method is able to provide the desired level of quality of service and high network utilization.

Non-uniform traffic issues in DCA wireless multimedia networks

Wireless networks that utilize dynamic channel allocation (DCA) are known to perform better than those with fixed channel allocation, in terms of the call level QoS measures such as the handoff dropping probability. On account of this, the DCA networks are usually designed without the call admission control (CAC). However, given the decrease of cell sizes, together with ever increasing mobile phone and terminal population, dynamic channel allocation policies (such as channel borrowing) may not be sufficient to cope with the hot-spot area size and its traffic intensity. This paper analyses the performance of the DCA networks, both with and without the call admission control, under the hot-spot traffic regime. In such cases, the pure DCA approach fails to ensure sufficiently low level of QoS in both the hot-spot area and the surrounding cells. We propose a CAC policy that can stabilize the QoS under non-uniform traffic, whilst being easy to integrate in the distributed DCA policies.

QOS measurement and evaluation of telecommunications quality of service [Book Review

QOS measurement and evaluation of telecommunications quality of service [Book Review

Admission Control for VoIP Networks using Edge-to-edge QoS measurements

Admission Control for VoIP Networks using Edge-to-edge QoS measurements

Fair resource sharing scheduling for cellular data services with QoS provisioning

To provide cellular data services with QoS provisions, a shared resource scheme, based on optimisation theory and LaGrange λ-calculus, has been developed. This scheme can generate a fair schedule for a diverse mix of traffic with diverse QoS requirements in a limited radio spectrum. The authors define the acceptance indication (AI) as the QoS measurement of the shared resource scheme. The experimental results show that as much as 2.5% improvement in the mean acceptance rate is obtained relative to other existing schemes.