Bandwidth Reservation Policy Research Articles

Performance evaluation of a mobile hotspot supporting quasi‐random traffic

AbstractThe authors study and evaluate a mobility‐aware call admission control algorithm in a mobile hotspot. More specifically, a vehicle which has an access point of a fixed capacity and may alternate between stop and moving phases is considered. In the stop phase, the vehicle services new and handover calls. To prioritise handover calls a probabilistic bandwidth reservation policy is considered where a fraction of the capacity is reserved for handover calls. Based on this policy, new calls may enter the reservation space with a predefined probability. In addition, handover calls have the option to wait in a queue of finite size if there are no available resources at the time of their arrival. In the moving phase, the vehicle services only new calls under the classical complete sharing policy. In both phases, calls arrive in the system according to a quasi‐random process, require a single bandwidth unit for their acceptance in the system and have an exponentially distributed service time. To analytically determine the various performance measures, such as time congestion probabilities, call blocking probabilities and link utilisation, an accurate analytical method is presented based on three‐dimensional Markov chains.

Call Blocking Probabilities under a Probabilistic Bandwidth Reservation Policy in Mobile Hotspots

In this paper we study a mobility-aware call admission control algorithm in a mobile hotspot. To this end, a vehicle is considered which has an access point with a fixed capacity. The vehicle alternates between stop and moving phases. When the vehicle is in the stop phase, it services new and handover calls by prioritizing them via a probabilistic bandwidth reservation (BR) policy. Based on this policy, new handover calls may enter the reservation space with a predefined probability. When the vehicle is in the moving phase, it services new calls only. In that phase, two different policies are considered: (a) the classical complete sharing (CS) policy, where new calls are accepted in the system whenever there exists available bandwidth, and (b) the probabilistic BR policy. Depending on the selected policy in the moving phase, we propose the probabilistic BR loss model (if the CS policy is selected) and the generalized probabilistic BR loss model (if the probabilistic BR policy is selected). In both stop and moving phases, where the call arrival process is Poisson, calls require a single bandwidth unit in order to be accepted in the system, while the service time is exponentially distributed. To analytically determine call blocking probabilities and the system’s utilization, we propose efficient iterative algorithms based on two-dimensional Markov chains. The accuracy of the proposed algorithms is verified via simulation.

Congestion probabilities in OFDM wireless networks with compound Poisson arrivals

The authors study the downlink of an orthogonal frequency division multiplexing (OFDM) based cell that services calls from different service-classes with various resource requirements. They assume that calls arrive in the cell as batches according to a compound Poisson process. They consider that the batch size is generally distributed while each call of a batch is treated separately from the other calls of the same batch, according to the complete sharing policy. To determine the most important performance metrics, i.e. congestion probabilities and resource utilisation in this OFDM-based cell, they model it as a multirate loss model, show that the steady-state probabilities can be determined via a product form solution (PFS) and propose recursive formulas which reduce the complexity of the calculations. In addition, they study the bandwidth reservation (BR) policy which can be used for the reservation of subcarriers in order to favour service-classes whose calls have high subcarrier requirements. The existence of the BR policy destroys the PFS of the steady-state probabilities. However, they show that there exist recursive formulas for the determination of the various performance measures. Simulation verifies the accuracy of the proposed formulas.

User Mobility in a 5G Cell with Quasi-Random Traffic under the Complete Sharing and Bandwidth Reservation Policies

5G cellular mobile networks aspire to accommodate large numbers of users and devices most of which are expected to be mobile. They will also provide much higher data rates than previous generations’ networks. Analyzing how a cell’s performance is impacted by user mobility becomes paramount in an effort to provide the best possible quality of services. In this paper we propose a model of a generic cell where each user is moving based on the random waypoint model. We then adapt this model to a 5G cell to extract meaningful quality of service metrics. To this end and to provide a better perspective of how the model can be adapted to different bandwidth allocation policies, we extend the Engset multirate loss model to obtain recursive but efficient formulas for various performance measures, including call blocking and time congestion probabilities for both the complete sharing and bandwidth reservation policies.

Congestion probabilities in the X2 link of LTE networks

In this paper, first we review two multirate loss models, whereby we can assess the call-level QoS of the Long Term Evolution X2 link supporting calls of different service-classes with fixed bandwidth requirements. The X2 interface directly connects two neighboring evolved NodeBs and is mainly responsible for the transfer of user-plane and control-plane data during a handover. In both models, the X2 interface is modelled as a link of fixed capacity. Handover calls are accepted in the X2 link whenever there exists available bandwidth, i.e., no QoS guarantee is achieved for high-speed calls. Secondly, we propose three multirate loss models where calls arrive in the X2 link according to a quasi-random process and compete for the available bandwidth under the Multiple Fractional Channel Reservation (MFCR) policy, the Bandwidth Reservation (BR) policy and the Complete Sharing (CS) policy. The MFCR/BR policies allow the reservation of real/integer number of channels, respectively, in order to benefit high-speed calls. The CS policy allows calls to enter the system when there exists available bandwidth (no reservation is allowed). We propose approximate but recursive formulas for the calculation of time and call congestion probabilities as well as link utilization for all three policies. The accuracy of the proposed formulas is verified through simulation and found to be highly satisfactory.

An efficient bandwidth reservation policy for data-intensive applications in cloud

Recently, more and more data-intensive applications have been deployed in cloud platforms due to its cost-effective and elastic capability. Although computing resources (i.e., CPU and RAM) can be provisioned in on-demand manner, the network bandwidth has enforce certain admission controlling for providing desirable quality-of-service (QoS). As a result, data-intensive applications often suffer from high delay during the phase of bulk data transferring between computing nodes. In this paper, we present an integrated bandwidth reservation algorithm, which combines different admission controlling policies and minimises the finishing time for all jobs in an online manner. Performance evaluation on a real-world cloud platform shows that the proposed reservation algorithm can ensure efficient bulk data transferring even when the system is facing heavy workloads.

An efficient bandwidth reservation policy for data-intensive applications in cloud

Recently, more and more data-intensive applications have been deployed in cloud platforms due to its cost-effective and elastic capability. Although computing resources (i.e., CPU and RAM) can be provisioned in on-demand manner, the network bandwidth has enforce certain admission controlling for providing desirable quality-of-service (QoS). As a result, data-intensive applications often suffer from high delay during the phase of bulk data transferring between computing nodes. In this paper, we present an integrated bandwidth reservation algorithm, which combines different admission controlling policies and minimises the finishing time for all jobs in an online manner. Performance evaluation on a real-world cloud platform shows that the proposed reservation algorithm can ensure efficient bulk data transferring even when the system is facing heavy workloads.

Call blocking probabilities in a two‐link multirate loss system for Poisson traffic

The authors propose two multirate teletraffic loss models in a two-link system that accommodates Poisson arriving calls from different service-classes with different bandwidth-per-call requirements. Each link has two thresholds which refer to the number of in-service calls in the link. The lowest threshold named support threshold, defines up to which point the link can support calls offloaded from the other link. The highest threshold, named offloading threshold, defines the point where the link starts offloading calls to the other link. Two different bandwidth sharing policies are considered: (i) the complete sharing policy, in which a call can be accepted in a link if there exist enough available bandwidth units and (ii) the bandwidth reservation policy, in which an integer number of bandwidth units is reserved to benefit calls of high bandwidth requirements. The two models do not have a product form solution for the steady state probabilities. However, they propose approximate formulas for the calculation of call blocking probabilities. The accuracy of the formulas is verified through simulation and found to be quite satisfactory.

Performance modelling of a multirate loss system with batched Poisson arrivals under a probabilistic threshold policy

The authors consider a link accommodating batched Poisson arriving calls of different service classes. A batch of generally distributed number of calls arrive to the link at exponentially distributed time-points. Each call is treated separately from the rest, and its acceptance is decided, according to the available link bandwidth (partial batch blocking discipline). For congestion control, if the number of in-service calls of a service class exceeds a threshold (dedicated to the service class), a new call of the same service class is accepted with a probability, dependent on the system state (probabilistic threshold policy). The link is analysed as a multirate loss system, via a reversible Markov chain. The latter leads to a product form solution (PFS) for the steady-state distribution. Based on the PFS, the authors propose models for the accurate determination of time and call congestion probabilities and link utilisation. Comparison against other existing models under the complete sharing or the bandwidth reservation policy reveals the necessity and consistency of the proposed models.

An Integrated Bandwidth Adaptation Scheme for Multimedia Wireless Networks and its Connection-Level Performance Analysis

This paper presents an integrated bandwidth adaptation scheme for multimedia wireless networks using application utility functions. With the proposed scheme, each call in the network is assigned a utility function according to itsadaptive characteristics. Depending on the network load theallocated bandwidth of ongoing calls are upgraded or degraded dynamically so that the achieved utility of the network is maximized. Appropriate call admission control and bandwidth reservation policies are also incorporated into the scheme to provide QoS guarantees to the new and handoff calls, respectively. Extensive simulation experiments have been conducted to evaluate the connection-level performance of the proposed scheme. Results show that our bandwidth adaptation scheme is effective in both increasing the utility and bandwidthutilization of wireless networks while keeping the call blocking and handoff dropping probabilities substantially low.

Call Blocking Probabilities of Multirate Elastic and Adaptive Traffic under the Threshold and Bandwidth Reservation Policies

This paper proposes multirate teletraffic loss models of a link that accommodates different service-classes of elastic and adaptive calls. Calls follow a Poisson process, can tolerate bandwidth compression and have an exponentially distributed service time. When bandwidth compression occurs, the service time of new and in-service elastic calls increases. Adaptive calls do not alter their service time. All calls compete for the available link bandwidth under the combination of the Threshold (TH) and the Bandwidth Reservation (BR) policies. The TH policy can provide different QoS among service-classes by limiting the number of calls of a service-class up to a predefined threshold, which can be different for each service-class. The BR policy reserves part of the available link bandwidth to benefit calls of high bandwidth requirements. The analysis of the proposed models is based on approximate but recursive formulas, whereby authors determine call blocking probabilities and link utilization. The accuracy of the proposed formulas is verified through simulation and found to be very satisfactory.

English

English

Congestion probabilities of elastic and adaptive calls in Erlang-Engset multirate loss models under the threshold and bandwidth reservation policies

In this paper, we consider a single link of fixed capacity that accommodates calls of different service-classes with different bandwidth-per-call requirements. The link behaves as a multirate loss system. Calls of each service-class arrive in the link according to a Poisson (random) or a quasi-random process and have an exponentially distributed service time. Poisson or quasi-random arriving calls are generated by an infinite or finite number of traffic sources, respectively. Service-classes are also distinguished according to the behavior of in-service calls, in elastic and adaptive service-classes. Elastic calls can compress their bandwidth by simultaneously increasing their service time. Adaptive calls tolerate bandwidth compression without affecting their service time. All calls compete for the available link bandwidth under the combination of the Threshold (TH) and the Bandwidth Reservation (BR) policies. The TH policy can provide different QoS among service-classes by limiting the number of calls of a service-class up to a predefined threshold, which can be different for each service-class. The BR policy reserves part of the available link bandwidth to benefit calls of high bandwidth requirements. The proposed models, for random or quasi-random traffic, do not have a product form solution for the determination of the steady state probabilities. However, we approximate both models by reversible Markov chains, and prove recursive formulas for the efficient calculation of the call-level performance metrics, such as time and call congestion probabilities as well as link utilization. The accuracy of the proposed formulas is verified through simulation and found to be quite satisfactory.

Blocking probabilities of elastic and adaptive calls in the Erlang multirate loss model under the threshold policy

In this paper, we propose a new multirate teletraffic loss model of a single link with certain bandwidth capacity that accommodates Poisson arriving calls, which can tolerate bandwidth compression, under the threshold policy of admission control. When compression occurs, the service time may increase (elastic calls) or not (adaptive calls). The threshold policy can provide different QoS among service-classes by limiting the number of calls of a service-class up to a predefined threshold, which can be different for each service-class. The proposed model does not have a product form solution for the determination of the steady state probabilities. However, we approximate the model by a reversible Markov chain, and provide recursive formulas for the efficient calculation of the call-level performance metrics, such as call blocking probabilities and link utilization. In addition, we provide similar formulas when the threshold policy co-exists with the bandwidth reservation policy. The latter reserves part of the available link bandwidth to benefit calls of high bandwidth requirements. The accuracy of the proposed formulas is verified through simulation and found to be quite satisfactory. We also show the necessity and consistency of the proposed models.

Performance metrics of a multirate resource sharing teletraffic model with finite sources under the threshold and bandwidth reservation policies

The authors propose a new multirate teletraffic loss model of a single link with certain capacity that accommodates different service-classes whose calls come from finite traffic sources. Calls compete for the available link bandwidth under the combination of the threshold (TH) and the bandwidth reservation (BR) policies. The TH policy can provide different quality of service among service-classes by limiting calls of each service-class up to a certain number, which is a predefined TH, which can be different for each service-class. The BR policy reserves part of the available link bandwidth to benefit calls of high bandwidth requirements. They show that the proposed model, without the BR policy, has a product form solution (PFS) and prove recursive formulas for the efficient calculation of the call-level performance metrics, such as time and call congestion probabilities as well as link utilisation. The combination of the TH and BR policies destroys the PFS of the model. However, they show that approximate but recursive formulas still exist for the efficient calculation of the call-level performance metrics. The accuracy of the proposed formulas is verified through simulation and found to be very satisfactory.

QoS Equalization in a W-CDMA Cell Supporting Calls of Infinite or Finite Sources with Interference Cancelation

In this paper, a multirate loss model for the calculation of time and call congestion probabilities in a Wideband Code Division Multiple Access (W-CDMA) cell is considered. It utilizes the Bandwidth Reservation (BR) policy and supports calls generated by an infinite or finite number of users. The BR policy achieves QoS equalization by equalizing congestion probabilities among calls of different service-classes. In the proposed models a multiple access interference is considered, and the notion of local blocking, user's activity and interference cancelation. Although the analysis of the proposed models reveals that the steady state probabilities do not have a product form solution, the authors show that the calculation of time and call congestion probabilities can be based on approximate but recursive formulas, whose accuracy is verified through simulation and found to be quite satisfactory.

The extended connection‐dependent threshold model for call‐level performance analysis of multi‐rate loss systems under the bandwidth reservation policy

SUMMARYFirstly, we reviewed two extensions of the Erlang multi‐rate loss model, whereby we can assess the call‐level QoS of telecom networks supporting elastic traffic: (i) the extended Erlang multi‐rate loss model, where random arriving calls of certain bandwidth requirements at call setup can tolerate bandwidth compression while in service; and (ii) the connection‐dependent threshold model, where arriving calls may have several contingency bandwidth requirements, whereas in‐service calls cannot tolerate bandwidth compression. Secondly, we proposed a new model, the extended connection‐dependent threshold model. Calls may have alternative bandwidth requirements at call setup and can tolerate bandwidth compression while in service. We proposed a recurrent formula for the efficient calculation of link occupancy distribution and consequently call blocking probabilities, link utilization, and throughput per service class. Furthermore, in the proposed model, we incorporated the bandwidth reservation policy, whereby we can (i) equalize the call blocking probabilities of different service classes, (ii) guarantee specific QoS per service class, and (iii) implement different maximum bandwidth compression/expansion rate per service class so that the network supports both elastic and stream traffic. The accuracy of the new model is verified by simulation. Moreover, the proposed model performs better than the existing models. Finally, we generalize the proposed model by incorporating service classes with either random or quasi‐random arrivals. Copyright © 2011 John Wiley & Sons, Ltd.

The Erlang multirate loss model with Batched Poisson arrival processes under the bandwidth reservation policy

The Erlang Multirate Loss Model (EMLM) has been widely used as a springboard in the study of multirate loss systems. In this paper, we study an extension of the EMLM in order to ensure QoS guarantee per service-class in the heterogeneous environment of telecom networks. The proposed model is named Batched Poisson EMLM under Bandwidth Reservation (BR) policy (BP-EMLM/BR), since its input process is a Batched Poisson arrival process; that is, calls of each service-class arrive to a fixed-capacity link as batches. The distribution of the batch size can be general. The batch blocking discipline is the partial batch blocking, i.e. depending on the available link bandwidth, a part of an arriving batch can be accepted while the rest of it is discarded. The Call Admission Control is based on the BR policy, whereby we can achieve our goal to guarantee specific QoS for each service-class at call-level. In order to apply the BR policy in the BP-EMLM/BR, we examine two methods also used in the EMLM: (a) the Roberts’ method and (b) the Stasiak–Glabowski (S&G) method. The proposed model does not have a product form solution and therefore we propose approximate but recursive formulas for the calculation of various performance measures, such as time and call congestion probabilities as well as link utilization. A comparison between the analytical and simulation results shows that both methods provide satisfactory results when equalization between time congestion probabilities of all service-classes is desired. The S&G method, however, performs much better than the Roberts’ method in the case where more than two service-classes are accommodated in the link.

On utility-fair bandwidth adaptation for multi-class traffic QoS provisioning in wireless networks

The ability to adjust the allocated bandwidth of ongoing calls to cope with wireless network resource fluctuations is becoming increasingly important. In this paper, we describe a utility-based bandwidth adaptation scheme for multi-class traffic quality-of-service (QoS) provisioning in wireless networks. With the proposed scheme, each call is assigned a utility function according to its adaptive characteristics. Depending on the network load the allocated bandwidth of ongoing calls are upgraded or degraded dynamically such that each call receives fair utility. The quantization of utility function by dividing the utility range into a fixed number of equal intervals is a key feature of our proposed utility-fair algorithm. Appropriate call admission control and bandwidth reservation policies are also integrated into the proposed scheme to provide QoS guarantees to the new and handoff calls. Extensive simulation experiments have been conducted to evaluate the performance of the proposed scheme compared with two other existing ones. Results show that our bandwidth adaptation scheme is effective in achieving utility fairness while keeping the call blocking and handoff dropping probabilities substantially low.

Engset multi-rate state-dependent loss models with QoS guarantee

We consider a single link loss system of quasi-random input, described by the Engset multirate loss model (EnMLM). Blocked calls may once reattempt to be connected to the system requiring less bandwidth; then the system is described by the single retry model for finite sources (f-SRM). The EnMLM and the f-SRM are extended with the single threshold finite source model (f-STM), where calls may attempt to be connected to the system with less bandwidth requirements, according to the link occupancy, before blocking occurs. We focus on CBP equalization in the EnMLM, f-SRM and f-STM, under the bandwidth reservation (BR) policy. For this analysis, we apply two approximate methods, the Roberts' method and the method of reverse transition rates (RTR), which lead to a recursive CBP calculation. We evaluate the accuracy of the above models under the BR policy by comparing the analytical with simulation CBP results, based on the relative approximation errors (RAE). The results are highly satisfactory because they show that the proposed models (formulas) lead to small RAE. Copyright © 2005 John Wiley & Sons, Ltd.