Probabilistic framework and performance evaluation for prioritized call admission control in next generation networks
Introduction
To support various integrated services, such as voice, data, world wide web (WWW) browsing, e-mail and multimedia via mobile computing devices over the wireless infrastructure of next generation networks (NGN), radio resource management (RRM) has evolved as a critical issue [1], [2]. One of the main RRM functions is call admission control (CAC) which, following a certain policy, determines whether a new call request should be accepted [3], [4]. Thus, due to its critical role with regard to quality of service (QoS), CAC in next generation multi service class (SC) networks is given a great attention. An efficient CAC scheme for present and next generation multi-SC wireless cellular networks, which remains a reliable method of telecommunication infrastructure, should satisfy the following main objectives: (a) establish a mechanism assigning appropriate priorities to different types of traffic load, (b) achieve low call blocking probabilities (CBPs) for all SCs, especially for those of high priority, (c) maximize resource utilization with fairness for all SCs, (d) achieve a high network throughput, (e) prevent congestion and (f) meet the specific QoS requirements of each SC [5].
Calls belonging to different SCs may have different bandwidth requirements, call holding times and cell residence times, which results in different QoS levels [6], [7], [8]. Combined with the limited resources of network cells allowing for only a certain number of calls from every SC, this feature leads to the adoption of prioritization models. In class based CAC, assigning priorities to certain SCs is critical in an attempt to minimize the CBP of high priority SCs. Thus, higher priorities are assigned to delay-non-tolerant (DNT) SC calls compared to delay-tolerant (DT) ones.
The design and implementation of these schemes seem challenging since some important issues come up. The main problem of CAC schemes basing their admission decision on SC priority is that SC calls of high priority often monopolize network resources. This leads to unacceptably high CBP levels for SC calls of low priority. Thus, fairness in call admission among users of the same SC (under different channel conditions and mobility characteristics) as well as among users of different SCs is an important issue in class based CAC schemes [3], [6]. Usually in NGNs, unfairness is a side effect of resource optimization. Furthermore, to provide the required QoS level to high priority SCs, many CAC schemes reserve network resources resulting in low average throughput levels. Therefore, apart from the CBP achieved, CAC schemes should also take into account the effectiveness in assigning priority as well as fairness and average throughput.
To the authors best knowledge, however, little attention has been given to the implementation of probabilistic models in NGNs. In this paper, a probabilistic CAC framework is proposed aiming at fairly and efficiently controlling the admission of different SC calls in NGNs. One of the key points is that the CAC scheme proposed is more general than the existing ones, allowing the extension of the proposed method to multiple SCs, by assigning different call holding times for every SC. The analysis is based on multidimensional Markov chains yielding CBP expressions of closed form. Moreover, the performance of the proposed probabilistic scheme is examined under certain efficiency metrics through appropriate numerical results and simulations.
The rest of the paper is organized as follows. Related work on prioritizing different SCs along with relevant ideas proposed so far are presented in Section 2. Section 3 introduces the system model and the assumptions for the NGN. In Section 4, the proposed probabilistic CAC scheme under consideration is analyzed and relevant CBP expressions are derived. In Section 5, new metrics for efficiency evaluation are introduced. Section 6 presents the simulations to evaluate and validate the performance of the proposed CAC scheme under different load conditions. Conclusions are deduced in Section 7.
Section snippets
Related work
Prioritization among new and handoff calls in wireless networks has been extensively investigated in the literature [9], [10]. In modern multimedia wireless networks research as to resource management has been focused on CAC schemes that assign priority to DNT SC calls over DT ones. Assigning priority to certain SCs is accomplished by adopting criteria that relegate the admission of low priority calls. In [11] three schemes are proposed to assign priorities among calls of different SCs. In this
System model and network assumptions
A single cell of a next generation cellular mobile communication network is considered. The assumptions made and notations used are stated below.
- (1)
The cellular network is assumed homogeneous; therefore it suffices to examine a single cell isolated from the others. Each network cell may serve U SCs of calls e.g. telephone, data, etc.
- (2)
C is the bandwidth capacity of the cell expressed in terms of bandwidth units (BUs), whereas M = C/c is the total number of calls where c is the BUs required by an
Probabilistic call admission control
The probabilistic CAC scheme for multi-SC wireless NGNs proposed in the present work under the name probabilistic bandwidth reservation scheme (PBRS) is based on the total number of BUs occupied in a cell. The multiple SCs supported by the network necessitate the use of multidimensional Markov chains, where each dimension of the chain corresponds to a different SC call stream. The rationale behind PBRS is to treat new SC calls of low priority by adjusting their admission rate according to
Criteria for performance evaluation
In the following, performance criteria for the proposed CAC scheme are defined. The efficiency of the priority assigning mechanism is evaluated through an appropriate priority index. Also, a fairness metric based on CBP is defined. This metric will next be used in the following to evaluate the performance of the CAC scheme with regard to fairness among calls of different SCs. Furthermore, a method for the estimation of the network throughput is also presented based on CBP.
Numerical results and performance evaluation
The performance of the proposed schemes has been evaluated under different traffic load conditions. The probabilistic framework adopted in the present work has been proven capable to provide priority to DNT calls in NGN. The numerical results intend to mark out how the probabilities imposed through the priority assigning mechanism enhance the performance of the proposed scheme under variable traffic load. Moreover, extensive simulations were performed employing a discrete event simulation
Conclusions and future work
A CAC scheme for NGNs was introduced based on a probabilistic framework to regulate the admission of low priority SC calls. Through this probabilistic framework, a novel priority assigning mechanism is introduced based on the selection of the admission probability of DT SC calls. Assuming independent call holding times for each SC supported, analytical expressions estimating CBP were provided based on Markov chain theory. Apart from the CBPs achieved, the proposed CAC scheme was evaluated under
References (46)
- et al.
A general call admission policy for next generation wireless networks
Comput. Commun.
(2005) - et al.
Refined approximations for performance analysis and optimization of queuing model with guard channels for handovers in cellular networks
Comput. Commun.
(2006) - et al.
Call admission control for wireless personal communications
Comput. Commun.
(2003) - et al.
Dual-threshold admission control for non-real-time traffic in wireless data networks
Comput. Commun.
(2008) - et al.
Traffic model and performance analysis for cellular mobile radio telephone systems with prioritized and non-prioritized handoff procedures
IEEE Trans. Veh. Technol.
(1986) - et al.
A framework for bandwidth degradation and call admission control schemes for multiclass traffic in next-generation wireless networks
J. Select. Areas Commun.
(2003) Call admission control in wireless networks: a comprehensive survey
IEEE Commun. Survey Tutorials
(2005)- et al.
Admission control for QoS support in heterogeneous 4G wireless networks
IEEE Netw.
(2008) - et al.
Call admission control for an adaptive heterogeneous multimedia mobile network
IEEE Trans. Wireless Commun.
(2007) - et al.
Call admission control for multiservice wireless networks with bandwidth asymmetry between uplink and downlink
IEEE Trans. Veh. Technol.
(2006)
Analysis of adaptive bandwidth allocation in wireless networks with multilevel degradable quality of service
IEEE Trans. Mobile Comput.
Realistic cell-oriented adaptive admission control for QoS support in wireless multimedia networks
IEEE Trans. Veh. Technol.
A dynamic call admission policy with precision QoS guarantee using stochastic control for mobile wireless networks
IEEE Trans. Netw.
QoS-based admission and load control algorithm for integrated voice/data services over wideband CDMA
Wireless Pers. Commun.
Virtual partitioning resource allocation for multiclass traffic in cellular systems with QoS constraints
IEEE Trans. Veh. Technol.
Efficient call admission control for heterogeneous services in wireless mobile ATM networks
IEEE Commun. Mag.
Analysis of a hybrid cutoff priority scheme for multiple classes of traffic in multimedia wireless networks
Wireless Netw.
Threshold-type call admission control in wireless/mobile multimedia networks using prioritized adaptive framework
IEE Elect. Lett.
A fair resource allocation protocol for multimedia wireless networks
IEEE Trans. Parallel Distrib. Syst.
On optimal call admission control in cellular networks
Wireless Netw.
Cited by (7)
Analysis and comparison of two optimized call admission control schemes in mobile computing environment
2016, Zhongnan Daxue Xuebao (Ziran Kexue Ban)/Journal of Central South University (Science and Technology)Mobility-Prediction-Aware Bandwidth Reservation Scheme for Mobile Networks
2015, IEEE Transactions on Vehicular TechnologyAn integrated predictive mobile-oriented bandwidth-reservation framework to support mobile multimedia streaming
2014, IEEE Transactions on Wireless CommunicationsMultidimensional queueing models in telecommunication networks
2014, Multidimensional Queueing Models in Telecommunication NetworksDynamic channel allocation to interfered cellular networks
2014, Wireless Personal Communications