Measurement-based Admission Control Research Articles

Scalable qDPMsub 2 mechanism: admission control for differentiated service networks

This paper examines the problem of achieving the QoS goals in differentiated service (DiffServ) networks, studies a dynamic provisioning mechanism for admission control in the DiffServ network, and proposes the scalable qDPM/sup 2/ mechanism based on the centralized bandwidth broker, distributed measurement based admission control and movable boundary bandwidth management.

A time-scale decomposition approach to measurement-based admission control

We propose a time-scale decomposition approach to measurement-based admission control (MBAC). We identify a critical time scale, T/spl tilde//sub h/, such that: 1) aggregate traffic fluctuations slower than T/spl tilde//sub h/ can be tracked by the admission controller and compensated for by flow admissions and departures; 2) fluctuations faster than T/spl tilde//sub h/ have to be absorbed by reserving spare bandwidth on the link. The critical time scale is shown to scale as T/sub h///spl radic/n, where T/sub h/ is the average flow duration and n is the size of the link in terms of the number of flows it can carry. An MBAC design is presented which filters aggregate measurements into low- and high-frequency components separated at the cutoff frequency, 1/T/spl tilde//sub h/, using the low-frequency component to track slow time-scale traffic fluctuations and the high-frequency component to estimate the spare bandwidth needed. Our analysis shows that the scheme achieves high utilization and is robust to traffic heterogeneity, multiple time-scale fluctuations and measurement errors. The scheme uses only measurements of aggregate bandwidth and does not need to keep track of per-flow information.

Measurement‐based admission control scheme with priority and service classes for application in wireless IP networks

AbstractWireless IP networks will provide voice and data services using IP protocols over the wireless channel. But current IP is unsuitable to provide delay or loss bounds and insufficient to support diverse quality of service, both required by real‐time applications. In order to support real‐time applications in wireless IP networks, in this paper a measurement‐based admission control (MBAC) with priority criteria and service classes is considered. First we have shown the suitability of MBAC in wireless IP networks by comparing its performance with a parameter‐based scheme. Next, we have investigated the performance of strictly policy‐based MBAC and policy plus traffic characteristic‐based MBAC schemes in terms of (1) increasing the user mobility, (2) changing traffic parameters and (3) the presence of greedy users. The efficiency and fairness of each scheme are measured in terms of lower class new and handoff traffic performance. Copyright © 2003 John Wiley & Sons, Ltd.

A service model for guaranteeing packet loss bounds in differentiated services architectures

Differentiated Services (DiffServ) is a proposed architecture for the Internet in which various applications are supported using a simple classification scheme. Packets entering the DiffServ domain are marked depending on the packets' class. Premium service and Assured service are the first two types of services proposed within the DiffServ architecture other than the best effort service. Premium service provides a strict guarantee on users' peak rates, hence delivering the highest Quality of Service (QoS). However, it expected to charge at high rates and also to have low bandwidth utilization. The Assured service provides high priority packets with preferential treatment over low priority packets but without any quantitative QoS guarantees. In this paper, we propose a new service, which we call the Loss bound Guaranteed (LG) service for DiffServ architectures. This service can provide a quantitative QoS guarantee in terms of loss rate. A measurement-based admission control scheme and a signaling protocol are designed to implement the LG service. An extensive simulation model has been developed to study the performance and viability of the LG service model. We have tested a variety of traffic conditions and measurement parameter settings in our simulation. The results show that the LG service can achieve a high level of utilization while still reliably keeping the traffic within the maximum loss rate requirement. Indeed, we show that the DiffServ architecture can provide packet-loss guarantees without the need for explicit resource reservation.

A measurement-based call admission control scheme for ATM networks based on the diffusion approximation

We present a dynamic call admission control method based on the diffusion approximation method. The diffusion approximation method is traditionally used for calculating the effective bandwidth for a number of connections having a certain cell loss requirement. The proposed method has a low implementation overhead since it performs measurements on the aggregate traffic stream at a particular ATM switch as opposed to other methods that need to perform per-source measurements. We provide a study of the performance of the method for specific traffic types using simulation and compare with the performance of the static diffusion approximation effective bandwidth method. We also study the influence of variation in traffic parameters such as the source traffic descriptors, the buffer capacity, and the required cell loss probability on the performance of the method.

Measurement-based admission control using maximum burstiness

We introduce a new measurement-based admission control using the maximum burstiness (MB) of the aggregate traffic. In the MB algorithm, admission control is performed by testing the schedulability of a new flow request using the target loss probability to get a feasible value of a performance tuning parameter. The MB algorithm obtains robustness by applying the quality-of-service requirements to the admission control test and achieves high bandwidth utilization by adopting statistical multiplexing.

A new admission control approach based on prediction

Admission control plays an important role in providing QoS to network users. Motivated by the measurement-based admission control algorithm, this letter proposed a new admission control approach for integrated service packet network based on traffic prediction. In the letter, FARIMA(p,d,q) models in the admission control algorithm is deployed. A method to simplify the FARIMA model fitting procedure and hence to reduce the time of traffic modeling and prediction is suggested. The feasibility-study experiments show that FARIMA models which have less number of parameters can be used to model and predict actual traffic on quite a large time scale. Simulation results validate the promising approach.

Connection Admission Control for PCS-to-Internet Protocol Internetworking

This paper studies the connection admission control for PCS-to-Internet protocol internetworking. In order to provide QoS to the Internet and avoid scalability problems, several recent papers proposed endpoint measurement-based admission control (EMAC) schemes, in which admission decisions are based on measurement results from probe packets exchanged between end hosts. Although EMAC has many desirable features in the wireline networks as shown in previous work, we show in this paper that several distinct characteristics in mobile wireless networks, such as low bandwidth, user mobility, and high bit error rate, make EMAC difficult to implement. Observing that networks offering 3rd-generation (3G) personal communication service (PCS), such as UMTS, already incorporate their own connection admission control functions, we propose a mobile-EMAC (M-EMAC) scheme for PCS-to-IP internetworking. Simulation results show that M-EMAC outperforms EMAC in the wireless domain by improving channel utilization while reducing packet losses.

Dynamic resizing of utilization target in measurement-based admission control

A major issue in design of measurement-based admission control (MBAC) algorithms relies on how to maximize link utilization while meeting applications’ QoS constraints. Studies have claimed that many proposed MBAC algorithms would give the same performance if the parameters were properly tuned. In this paper, we study the effects of tuning two of the most important parameters in MBAC algorithms: Measurement Window and Utilization Target. We show that the measurement window is not a good tuning knob for enforcing rigid loss probability constraints. We propose a new parameter called Soft Utilization Target combined with a simple flow admittance policy called the Latch algorithm. The simulation results show that this scheme can achieve better enforcement of loss probability constraint while maintaining higher link utilization.

Measurement-based admission control with aggregate traffic envelopes

The goal of admission control is to support the quality-of-service demands of real-time applications via resource reservation. We introduce a new approach to measurement-based admission control for multiclass networks with link sharing. We employ adaptive and measurement-based maximal rate envelopes of the aggregate traffic flow to provide a general and accurate traffic characterization that captures its temporal correlation as well as the available statistical multiplexing gain. In estimating the applications' future performance, we introduce the notion of a schedulability confidence level which describes the uncertainty of the measurement-based "prediction" and reflects temporal variations in the measured envelope. We then devise techniques to control loss probability for a buffered multiplexer servicing heterogeneous and bursty traffic flows, even in the regime of a moderate number of traffic flows, which is important in link-sharing environments. Finally, we have developed an implementation of the scheme on a prototype router and performed a testbed measurement study, which together with extensive trace-driven simulations illustrates the effectiveness of the approach in practical scenarios.

Traffic management and QoS provisioning in future wireless IP networks

The huge commercial success of mobile telephony, phenomenal growth of mobile and Internet users, and emerging popularity of IP-based multimedia applications are the major driving forces behind third-generation (3G) evolution. 3G will bring will bring applications, both data and multimedia, into wireless environments. It will operate on IP-based infrastructures to provide wider service access capability. However, current IP is designed for data applications with a single service class, best-effort. Therefore, it is insufficient to support real-time applications that require diverse quality of service (QoS). Although IP can provide greater service flexibility in terms of spectrum efficiency and QoS, it is not the most appropriate choice in wireless environments. In order to support real-time applications in wireless IP networks, measurement-based admission control (MBAC) is considered. We consider two admission criteria, reservation and priority. The system performance is measured in terms of new call blocking and hand-off call dropping probabilities with three different levels of user mobility: low, moderate and high level of mobility. Considering the delay-tolerant nature of real-time applications in wireless IP networks, the effect of delay tolerance on the overall system performance is also investigated. We discuss the suitability of MBAC in wireless IP networks by comparing its pros and cons with traditional parameter based schemes, and then describe its procedures. Next, we discuss simulation details, and present simulation results with discussions. We compare new and hand-off traffic performance with strictly policy-based and policy and traffic characteristic-based admission control schemes. We discuss the effectiveness of MBAC and its role in 3G and forthcoming 4G networks.

An adaptive algorithm for measurement-based admission control in integrated services packet networks

Call admission control is typically devised so as to employ the declared worst-case traffic descriptors for incoming calls, a solution resulting in poor bandwidth utilization. A measurement-based admission scheme is an appealing alternative, offering adaptivity to the changing traffic conditions, while allowing statistical multiplexing gains to be exploited. In this paper, we examine the problem of determining which traffic characterization a measurement-based admission control algorithm should require of sources requesting access. We also propose an adaptive measurement-based admission control algorithm that simplifies the estimation process. We test the new algorithm under different traffic scenarios and show that a high utilization level is achieved without violating delay-based QoS guarantees.

An adaptive call admission control in ATM networks using optimized measurements windows

We propose an adaptive measurement-based admission control algorithm that uses optimization procedure to choose an appropriate measurement window size dynamically. Previous designs on measurement-based admission control mainly focused on strategies that consider the worst case traffic source model to guarantee QoS bounds for all connections. In this paper, we develop a simple mechanism in which both statistical multiplexing gain and QoS are considered An accurate formula for the cell loss probability which is combined with measurements, is presented for refined on/off traffic model. Furthermore, more accurate measurements of traffic would be given to admission controller through optimized window size. The proposed adaptive control is an appealing alternative: not only does it offer adaptivity to changing traffic conditions, but supports a guarantee that loss QoS bound is not violated due to the admission of new calls into ATM network. Simulations were performed to test the performance of the proposed algorithm.

Transient analysis of traffic generated by bursty sources, and its application to measurement-based admission control

The first part of the paper is devoted to a transient analysis of traffic generated by bursty sources. These sources are governed by a modulating process, whose state determines the traffic rate at which the source transmits. The class of modulating processes contains, e.g., on/off traffic sources with general on and off times (but is considerably broader). We focus on the probability of extreme fluctuations of the resulting traffic rate, or more precisely, we determine the probability of the number of sources being in the on state reaching a certain threshold, given a measurement of the number of sources in the on state t units of time ago. In particular, we derive large deviations asymptotics of this probability when the number of sources is large. These asymptotics are numerically manageable, and it is empirically verified that they lead to an overestimation of the probability of our interest. The analysis is extended to alternative measurement procedures. These procedures allow to take into account, for instance, more historic measurements than just one, possibly combined with an exponential weighting of these measurements. In the second part of the paper, we apply the asymptotic calculation methods to gain insight into the feasibility of measurement?based admission control (MBAC) algorithms for ATM or IP networks. These algorithms attempt to regulate the network's load (to provide the customers with a sufficient Quality of Service), and at the same time achieve an acceptable utilization of the resources. An MBAC algorithm may base acceptance or rejection of a new request on the measured momentary load imposed on the switch or router; if this load is below a given threshold, the source can be admitted. We investigate whether such a scheme is robust under the possible stochastic properties of the traffic offered. Both the burst level (i.e., the distribution of the on and off times of the sources) and the call level (particularly the distribution of the call duration) are taken into account. Special attention is paid to the influence of the bursts, silences, or call durations having a distribution with a "heavy tail".

A large deviation analysis of errors in measurement based admission control to buffered and bufferless resources

In measurement based admission control, measured traffic parameters are used to determine the maximum number of connections that can be admitted to a resource within a given quality constraint. The assumption that the measured parameters are the true ones can compromise admission controls measured parameters are random quantities, causing additional variability. This paper analyzes the impact of measurement error within the framework of Large Deviation theory. For a class of admission controls, large deviation principles are established for the number of admitted connections, and for the attained overflow rates. These are applied to admission to bufferless resources, and buffered resources in both the many sources and large buffer asymptotic. The sampling properties of effective bandwidths are presented, together with a discussion the impact of the temporal extent of individual samples on estimator variability. Sample correlations are shown to increase estimator variances procedures to make admission control robust with respect to these are described.

An integrated admission control based on measurements in ATM networks

We developed a novel scheme to maintain QoS (Quality of Service) requirement for different traffic classes in the dynamic environment through full utilization of statistical multiplexing. Previous designs on measurement-based admission control mainly focused on strategies that consider the worst case traffic source model to guarantee QoS bounds for all connections. This can result in overallocation of resources. The proposed mechanism not only allows statistical multiplexing gain to be exploited, it also simplifies the estimation process. An accurate formula for the cell loss probability which is combined with measurements, is presented for a new traffic model. Furthermore we enhance the performance of this mechanism through real-time measurements of traffic and monitoring of QoS.

A framework for robust measurement-based admission control

Measurement-based admission control (MBAC) is an attractive mechanism to concurrently offer quality of service (QoS) to users, without requiring a priori traffic specification and on-line policing. However, several aspects of such a system need to be dearly understood in order to devise robust MBAC schemes, i.e., schemes that can match a given QoS target despite the inherent measurement uncertainty, and without the tuning of external system parameters. We study the impact of measurement uncertainty, flow arrival, departure dynamics, and of estimation memory on the performance of a generic MBAC system in a common analytical framework. We show that a certainty equivalence assumption, i.e., assuming that the measured parameters are the real ones, can grossly compromise the target performance of the system. We quantify the improvement in performance as a function of the length of the estimation window and an adjustment of the target QoS. We demonstrate the existence of a critical time scale over which the impact of admission decisions persists. Our results yield new insights into the performance of MBAC schemes, and represent quantitative and qualitative guidelines for the design of robust schemes.

A measurement-based admission control algorithm using variable-sized window in ATM networks

The decisions of Connection Admission in ATM networks should be made in real time through the use of fast algorithm. Since it is difficult to construct an accurate model for the multiplexed traffic, approximation of the multiplexed load is necessary. In this paper, we focus on a dynamic CAC (Connection Admission Control) algorithm as a different approach, in which admission control decisions are made based on network measurements. The algorithm observes the traffic through a moving window and the window size is recomputed from the measured cell loss amount. Unlike analytic modeling, the proposed method is based on the measurements of the network load over a period that is long enough to reflect the current traffic behavior. This approach also makes is possible to reallocate network resources (bandwidth and buffers) for multiple traffic classes. The performance of the proposed method is analyzed by means of simulated tests. As will be seen in the simulation results, this approach enables us to achieve a high level of bandwidth utilization and can be easily implemented for real-time admission control while still reliably keeping the cell loss ratio below a certain level.

A framework for robust measurement-based admission control

Measurement-based Admission Control (MBAC) is an attractive mechanism to concurrently offer Quality of Service (QoS) to users, without requiring a-priori traffic specification and on-line policing. However, several aspects of such a system need to be clearly understood in order to devise robust MBAC schemes. Through a sequence of increasingly sophisticated stochastic models, we study the impact of parameter estimation errors, of flow arrival and departure dynamics, and of estimation memory on the performance of an MBAC system.We show that a certainty equivalence assumption, i.e., assuming that the measured parameters are the real ones, can grossly compromise the target performance of the system. We quantify the improvement in performance as a function of the memory size of the estimator and a more conservative choice of the certainty-equivalent parameters. Our results yield valuable new insight into the performance of MBAC schemes, and represent quantitative guidelines for the design of robust schemes.

Advance reservations for predictive service in the Internet

We extend a measurement-based admission control algorithm suggested for predictive service to provideadvance reservations for guaranteed and predictive service.while retaining the attractive features of predictive service.The admission decision for advance reservations is based oninformation about flows that overlap in time. For flows thathave not yet started, the requested values are used, and forthose that have already started measurements are used. Thisallows us to estimate the network load accurately for thenear future. To provide advance reservations we ask usersto include durations in their requests. We present simulationresults to show that predictive service with advance reservations provides utilization levels significantly higher thanthose for guaranteed service, and comparable to those forpredicted service without advance reservations. Those utilization levels are reached without any preemption of otheradmitted flows. Finally, we discuss how to setup advancereservations over multiple hops in the Internet using resourcereservation setup protocols.