A Hierarchical Approach to Joint Call Admission Control in ... - CiteSeerX

IST-Africa 2006 Conference Proceedings Paul Cunningham and Miriam Cunningham (Eds) IIMC International Information Management Corporation, 2006 ISBN: 1-905824-01-7

A Hierarchical Approach to Joint Call Admission Control in Heterogeneous Wireless Networks Kamil H. SULEIMAN, H. Anthony CHAN, and Mqhele E. DLODLO University of Cape Town, Rondebosch, Cape Town, 7700, South Africa Emails: [email protected]; [email protected]; [email protected] Abstract: The heterogeneous nature of the next generation wireless systems (NGWS) poses many challenges to the call admission control algorithm. In this paper, a hierarchical approach to joint call admission control is proposed for the heterogeneous wireless networks. In the proposed scheme, the joint call admission control algorithm will be divided into two; namely, horizontal call admission control (HCAC) and vertical call admission control (VCAC). This approach will avoid unnecessary signalling and processing for certain stages of the joint call admission control that are specific to an access network technology. We modify the dual threshold bandwidth reservation scheme (DTBR) to triple threshold bandwidth reservation scheme (TTBR) to support vertical handoff calls. The strong feature of this scheme is that it is based on complete sharing (CS) of the overall bandwidth, which is believed to yield high radio resource utilization while meeting guaranteed quality of service (QoS). Keywords: CS, QoS, vertical handoff.

1. Introduction The next generation wireless systems (NGWS) will integrate the current different wireless technologies and possibly the emerging ones to address the increasing demands of users [1]. In this heterogeneous environment the global roaming of mobile terminals will be supported by the IP backbone [2]. The end user is expected to be able to connect to any of the different available access networks. The end user can also roam seamlessly within these access networks through vertical handoffs. Support for priority system based on users’ service type and agreement is one of the issues that the NGWS aim to accomplish. Each access network technology offers different levels of coverage, QoS as well as cost to the end user. Admission of users into the different networks of the NGWS is controlled by the joint call admission control (JCAC) protocol [3]. Upon the arrival of a new call with a specified priority, the JCAC protocol will choose the best access network to admit the service request [3]. The JCAC policy will also determine the appropriate times for vertical handoffs. The JCAC protocol is desired to improve the utilization of the network resources by minimizing the blocking probability of new calls. At the same time, the JCAC protocol should avoid dropping of handoff calls of users who are sensitive to service interruption [4]. The protocol design must also be careful to avoid violation of guaranteed QoS, since improvement of blocking probability is generally done at the cost of QoS. Thus, an intelligent JCAC protocol in NGWS should find the optimum tradeoff between the blocking probability of new calls and QoS provisioning while avoiding dropping of handoff calls [5], Copyright © 2006 The authors

www.IST-Africa.org/Conference2006

Page 1 of 9

[6]. There should also be a means for load balancing between the different networks in situations where some of the networks serve so many users while others serve fewer. The extra signalling for vertical handoffs function should be kept as low as possible so that it does not consume a lot of network resources as well as mobile terminal’s power. Further, in [7] it was shown that frequent bandwidth switching among different bandwidth levels may be worse than a large QoS degradation ratio. These latter concepts lead to the objective of avoiding unnecessary vertical handoffs and giving priority to horizontal handoffs to a certain extent. Though vast research has been done on the call admission control (CAC) of homogeneous networks, the CAC in NGWS (i. e. JCAC) has not been well discussed yet. Because of the heterogeneous nature of the NGWS, the JCAC has to be treated differently. In this paper a hierarchical approach to the JCAC in the NGWS is proposed. We will extend the DTBR scheme to TTBR scheme to support vertical handoff calls in the heterogeneous environment. The rest of the paper is organized as follows. In section 2 we describe some work related to our discussion. Section 3 discusses the NGWS architecture and our assumptions. The proposed JCAC is presented in Section 4. Section 5 is intended to show how the proposed entities work together during a call arrival. Finally, in section 6 we make a conclusion and recommendation for future research.

2. Previous Work 2.1 CAC in Cellular Networks First come first served is the simplest admission control algorithm. However, it offers no priority of service. In [8], the authors proposed an algorithm based on complete partition (CP). In this scheme, the bandwidth is divided into segments where each type of call can be admitted only in a segment assigned to it. The major problem associated with this algorithm is the waste of bandwidth. A call request to a full segment is rejected even if there is enough bandwidth in a segment assigned for lower priority service calls. In [9], a “movable” boundary system was used based on CP to support voice and data calls. In this algorithm, the boundary for the partition is adjustable to deal with traffic changes in the system. The above scheme was extended in [10] using the guarded channel policy to differentiate between new and handoff voice calls. In this scheme, called dynamic partition (DP) scheme, K1 out of the C channels of a cell are reserved for voice calls (new/handoff) and K2 channels are reserved for data calls (Fig. 1). The remaining (C - K1 - K2) channels are shared in a fair manner by both voice and data calls. New voice calls can only use K3 out of the K1 channels. When a new voice call arrives, a channel is sought in K3. If a channel is available there, it will be admitted. Otherwise, a channel is sought in the shared area. It will be blocked if there is no channel in both K3 and the shared area. When a handoff voice call arrives, a similar search is done in the voice only area and then the shared area. It will be dropped only if there is no channel available both in the voice only and shared area. A similar decision is made for a data call arrival by first searching in the data only area and then the shared area. Still in the DP algorithm, there is an efficiency problem in the cases where there is no bandwidth available for a voice call (new/handoff) while there is enough bandwidth in the data only area. This is a waste of radio resource since data calls should get lower priority than voice calls.

Copyright © 2006 The authors


Page 2 of 9

Data calls

K2 Shared Channels

C

Handoff voice calls K3

New voice calls

K1

Figure 1. DP algorithm

In [11] and [12] the authors proposed the DTBR scheme for voice/elastic data cellular networks. This scheme is based on CS and the authors showed that it offers higher network utilization than the DP and meets guaranteed QoS. In DTBR, the C channels of the cell are divided into three regions by two thresholds K1 and K2 (K1 > K2). When the network occupancy level L is less than the threshold K2, both voice and data calls can be admitted into the system. When L is greater than K2, no data call can be admitted into the system. When L is greater than K1, no data call or new voice call can be admitted to the system. A handoff voice call will be dropped only if there is no channel available. Handoff voice calls New voice calls C K1 K2 Data calls

Figure. 2. DTBR algorithm.

2.2 JCAC in NGWS Some research has been done to solve the resource allocation challenges of the JCAC. The work in [13] and [14] presents an adaptive policy based access management system to support heterogeneous networks. The authors in [3] presented a JCAC scheme for integrated UMTS-WLAN network. Because of the fact that WLAN technologies are short-range networks and offer lower costs to users, the vertical handoff from UMTS to WLAN was defined as a desirable handoff. On the other hand, since the mobile terminal’s connection has to remain seamless as a user connected to the WLAN roams out of the WLAN domain, the handoff from the WLAN to the UMTS was defined in [3] to be a necessary handoff. This work gave more priority to necessary handoffs than desirable handoffs based on threshold values.

3. Network Architecture and Assumptions As shown in Figure 3, different access networks in the NGWS are connected to the IP backbone [15], [16]. More than one access network technology will be available in several Copyright © 2006 The authors


Page 3 of 9

places. Devices like the radio resource manager (RRM) will serve to integrate these access network technologies. IP backbone

RRM RNC AR

BSC AR

WLAN

WLAN UMTS cell

GPRS cell

Figure. 3. IP backbone based NGWS architecture. RNC = radio network controller, AR = access router, BSC = base station controller.

To simplify our discussion, we assume that the NGWS categorize calls into only two different types in terms of QoS guarantees; namely, voice and data calls. Voice calls are delay intolerant and a constant amount of bandwidth has to be reserved for each voice call throughout the call duration. Because data calls are delay insensitive, the bandwidth reserved for data calls can be flexible. Intuitively, more priority should be given to voice calls than data calls based on the above assumptions. The priority system in an access network identifies four different types of calls: horizontal handoff voice, vertical handoff voice, new voice and data calls, respectively from highest to lowest priority of admission. We don’t differentiate between new and handoff data calls since data applications are less sensitive to dropping of calls. In this paper we assume the amount of bandwidth in an access network under consideration (a cell, a WLAN domain…etc) is C unit channels.

4. Hierarchical Joint Call Admission Control Algorithm 4.1 A High Level Description In this paper, we propose a hierarchical approach to the JCAC. Here, the task of the JCAC is divided into two hierarchies; namely, Horizontal Call Admission Control (HCAC) and Vertical Call Admission Control (VCAC). The aim of this approach is to avoid processing unnecessary steps due to a unified JCAC that involves a lot of signalling. The HCAC controls intra domain admission policy in an access network while the VCAC controls inter domain admission policy to distribute calls among the different access networks that are integrated at a certain area. The HCAC and VCAC communicate with each other at necessary times. 4.2 TTBR Scheme We modify the dual threshold bandwidth reservation (DTBR) scheme to the triple threshold bandwidth reservation (TTBR) scheme as follows. The C channels of an access network are divided into four regions by three thresholds K1, K2, and K3 (K1 > K2 > K3) as shown in Fig. 4. When the network occupancy level L is less than the threshold K3, then both voice and data (whether new or handoff) calls can be admitted into the network. When L is greater Copyright © 2006 The authors


Page 4 of 9

than K3, no data call is admitted into the network. When L is greater than K2, then no new voice call can be admitted. Vertical handoff voice calls are dropped if L is greater than K1. Horizontal handoff voice calls are dropped only if L is equal to C. Horizontal handoff voice calls Vertical handoff voice calls New voice calls C K3

K2

K1

Data calls

Figure 4. TTBR scheme.

4.3 HCAC Each access network is managed by an HCAC protocol that is assigned to it. The function of the HCAC is similar to the CAC that before-4G wireless networks employ. The HCAC decides whether or not to admit a certain call to the access network. In the case of access technologies where there are multiple cells, the HCAC in a cell periodically shares traffic information with HCACs in neighbouring cells to predict future traffic conditions [17]. In such cases, it dynamically adjusts threshold values using the knowledge of the current traffic condition in its domain and neighbouring cells. However, the proposed HCAC needs to be able to carry out some additional functions in order to comply with the heterogeneous environment. It replies to a query by the VCAC as to whether it is possible to admit a new call and what the current status of the access network is (to be explained later). It also periodically reports on its own time to the VCAC to inform it about the network status. The information that the HCAC gives to the VCAC is determined according to the TTBR algorithm. The thresholds of the TTBR scheme are determined based on the preference of service providers as to which type of calls they want to serve more. The HCAC can be deployed in either of two modes: centralized or decentralized. In the centralized mode, for each access network an HCAC protocol that manages its admission will reside in the central RRM. For example, if there are 3 access networks that an RRM controls, then there will be three HCACs in it. In the decentralized mode, there will be an HCAC in each access network. There are many advantages that the latter alternative offers over the first one. For example, it prevents waste of resources. Individual access networks sometimes cover parts of the domains of multiple RRM’s (Figure 5). An access network in such a situation will be managed by all the neighbouring RRM's that it overlaps. That means, the number of HCAC entities in the centralized mode is generally greater than the number of access networks if a wide network coverage area involving multiple RRM’s is considered. But the number of HCACs in the decentralized mode is always the same as the number of access networks. Another advantage is that it is easier to modify the current CAC protocols of individual access networks and fit them into the NGWS. Thus in the proposed scheme, we choose the decentralized mode.



Page 5 of 9

IP backbone

RRM1

RNC AR

RRM2

BSC AR WLAN

WLAN UMTS cell

GPRS cell

Figure. 5. A scenario in which two neighboring RRM’s and overlapping UMTS and GPRS cells exist. The rectangles indicate the domains of the two RRM’s.

4.4 VCAC Three conditions may trigger the VCAC algorithm. We will discuss them in details in the following subsections. 1. Call Arrival When a new call arrives, the VCAC requests traffic information from all HCACs in access networks that are in the vicinity of the user. From the replies, the VCAC learns which access networks are able to accept the call and what the preference of each network is for the type of the service request (voice/data). It then compares the replies from the HCACs to find the most suitable network in the current situation. This will reduce load imbalance among the different access networks. If there are multiple access networks with the highest preference to accept the call, then the VCAC will randomly choose one of them to admit the call. The flow chart in Fig. 6 shows how the VCAC operates when a new call arrives. 2. Necessary Handoff Not all access networks are necessarily available to a service type at a certain location. Therefore, when the end user roams through the heterogeneous network, there is a chance that the access network to which the mobile terminal (MT) is connected is not available at certain locations. In this case, the HCAC will know that a vertical handoff is necessary and it informs the VCAC about this situation. Then the VCAC will choose the best access network to admit this call. 3. Desirable Handoff Even though the VCAC considers load balance when it admits new calls, because of departures of calls, load imbalance occurs. The VCAC periodically examines the traffic conditions of the different access networks using the report that it receives from individual HCACs. When the load imbalance approaches a certain threshold, the VCAC will decide to choose some MTs from the heavily loaded networks and connect them to lightly loaded ones. However, the VCAC should avoid unnecessary vertical handoffs because they cause a lot of signalling which consumes network resources and mobile terminals’ power. Thus, if network load imbalance occurs at an acceptable level, vertical handoffs will not be done. The JCAC instead puts more emphasis on attempting to create balance during the arrival of new calls.



Page 6 of 9

Start VCAC Get type of service

Get network availability and preference

Any network available for type i?

No

Reject call

Yes Compare network preferences

Multiple networks with highest preference?

End

Yes

Randomly choose one from highest preference networks

No Choose the network with highest preference

Admit call in chosen network

Figure 6. VCAC operation. i = an integer representing the type of service request.

5. How Things Fit Together Let us define the virtual load level (VLL) as follows. The VLL of an access network is a measure of how much of the bandwidth is occupied by calls of a certain type out of the optimum amount preferred as compared to other access networks. If two access networks have the same VLL at a certain time for a certain type of call, then both will have the same probability of accepting a call with this service type. Note that VLL is not an actual measure of network load; it is a virtual measure of load that reflects the preference of the admission policy for the service type proportion. An access network with a certain actual load level can have different VLLs for the different types of service request. The VLL is set with respect to the bandwidth in the access network. For simplicity let us assume that there are only four VLL thresholds with values 0, 1, 2 and 3. The VLL values 0, 1, 2 and 3 correspond to the 0, K3, K2 and K1 levels when using the TTBR algorithm. Let us see an example of how the VLL is used. Fig. 7 illustrates the admission policy in a heterogeneous environment consisting of WLAN, GPRS, and UMTS. The WLAN is preferred to be always filled with data calls (no reservation) while the UMTS is preferred to serve more calls (voice/data) than the GPRS in traffic conditions where not all of the networks are full. In this figure, the WLAN is chosen to admit a new voice call only if there is no other network that is able to admit it.



Page 7 of 9

VLL (v) = 2 VLL (d) = 0

VLL (v, d) = 3 VLL (v, d) = 3

L

VLL (v, d) = 2 L

VLL (v, d) = 2 C

C

VLL (v) = 1

WLAN

L VLL (v, d) = 1

C

VLL (v, d) = 0

GPRS

VLL (v, d) = 1

VLL (v, d) = 0

UMTS

Figure 7. . Examples of traffic conditions for three access networks. v = voice calls, d = data calls.

HCAC and VCAC work together as follows. When a new call request arrives to a network coverage area, the mobile terminal communicates to the RRM. The first entity in the RRM that is responsible to react for the admission request is the VCAC. The VCAC issues a query to each individual HCAC. Let the type of service requested be i. The query looks like this: “Can you accommodate the traffic type i? What is the current VLL?” The formal algorithm looks like: q (i); where q is a function that requests traffic information from the HCAC with an argument of the service type i. The reply from the HCACs can be explained as follows. Let b be a binary digit representing a Boolean value (1 for yes and 0 for no). Then the format of the reply will be (b, VLL). Here, the HCACs in the different access networks refer to the value of VLL (d) for a data call arrival and VLL (v) for a voice call arrival. For example, for a new data call arrival, the reply from the HCAC of the WLAN in Fig. 7 will be (1, 0) meaning yes I can accommodate it and the virtual load level in the network is 0 (“I am desperate to accept this call”). This same network with the same network condition would reply with the same value of b but with a different VLL for a new voice call arrival. This is the reply: (1, 1) with a meaning yes I can accommodate it and my virtual load level is 1 (“I don’t mind accepting this call if there is no other network that accepts it”). After receiving all the replies from the individual HCACs, the VCAC will select all the access networks that reply with b=1 for the call request and compares their VLL. If there is only one access network with the least VLL, the VCAC will decide to admit the call in the access network with the least VLL. If there are multiple networks with the least VLL, then it will randomly choose one of them to admit the call.

6. Conclusion In this paper we proposed a hierarchical approach to JCAC by dividing its work into the algorithms HCAC and VCAC that manage intra-network and inter-network admission. This hierarchy will simplify the work of a centralized JCAC by distributing its tasks to individual access networks. It also avoids unnecessary steps that follow from employing a uniform algorithm for every access network and service type. We modified the DTBR scheme to TTBR scheme to use it for the HCAC. Future work includes developing analytical models to determine optimal values for the thresholds in the TTBR. We also would like to find performance parameters to test the proposed system.



Page 8 of 9

References [1] J. Sachs, “A generic link layer for future generation wireless networking,” Proc. IEEE ICC, pp. 834-838, May 2003. [2] C. Perkins, “IP mobility support for IPV4,” IETF RFC 3344, 2002. [3] O. E. Falowo and H. A. Chan, “Joint call admission control for integrated UMTS-WLAN network,” SATNAC, 2005. [4] C. H. Choi, M. I. Kim, and S. J. Kim, “Call admission control using moving pattern of mobile users for mobile multimedia networks,” IEEE Conference on Local Computer Networks, pp. 431-440, 2002. [5] D. Karabudak, C. -C. Hung, and B. Bing, “A call admission control scheme using genetic algorithms,” ACM Symposium on Applied Computing, Nicosia, Cyprus, pp. 1151-1158, Mar. 2004. [6] G. H. S. Carvalho, C. R. L. Frances, J. C. W. A. Costa, R. C. M. Rodrigues, and S. V. Carvalho, “Optimal call admission control and vandwidth adaptation in multimedia cellular mobile networks,” Proc. ASMTA, Gunter Bolch, ISBN 1-84233-112-4, 2005. [7] C. T. Chou and K. G. Shin, “Analysis of combined adaptive bandwidth allocation and admission control in wireless networks,” Proc. IEEE INFOCOM, pp. 676-684, June 2002. [8] Y. C. Lai, and S. F. Tsai, “A fair admission control for large bandwidth multimedia applications,” Proc.GLOBECOM, Sam Antonio, TX, pp 317-322, Nov. 2002. [9] Y. -R. Huang, Y. B. Lin, and J. M. Ho, “Performance analysis for voice/data integration on a finite mobile systems,” IEEE Trans. Veh. Technol., vol. 49, pp. 367-378, Mar. 2000. [10] B. Li, L. Li, B. Li, K. M. Sivalingam, and X. –R. Cao, “Call admission control for voice/data integrated cellular networks: performance analysis and comparative study,” IEEE Journal on Selected Areas in Comm., vol. 22, pp. 706-718, May 2004. [11] L. Yin, B. Li, Z. Zhang, and Y. –B. Lin, “Performance analysis of a dual threshold reservation (DTR) scheme for voice/data integrated mobile wireless networks,” Proc. IEEE WCNC, pp. 258-262, Sept. 2000. [12] L. –Z. Li, B. Li, and X. –R. Cao, “Performance analysis of bandwidth allocations for multi-service mobile wireless cellular networks,” Proc. IEEE WCNC, pp. 1072-1077, Mar. 2003. [13] K. Muray, R. Mathur, and D. Pesch, “Adaptive policy based access management in heterogeous wireless networks,” Proc IEEE WPMC, Vol. 1, pp. 325-329, Oct. 2003. [14] F. Yu, V. Wong, and V. Leung, “A new QoS provisioning method for adaptive multimedia in cellular wireless networks,” Proc IEEE INFOCOM, pp. 2131-2142, Mar. 2004. [15] P. Vidales, G. Mapp, F. Stajano, J. Crowcroft, and C. J. Bernardos, “A practical approach for 4G systems: deployment of overlay networks,” TRIDENTCOM, Trento, Italy, pp. 172-181, Feb. 2005. [16] W. Kellerer and H. Vogel, “A communication gateway for infrastructure-independent 4G wireless access,” IEEE Comm. Mag., pp. 126-131, Mar. 2002. [17] M. Naghshineh and M. Schwartz, “Distributed call admission control in obile/wireless networks,” IEEE J. Select. Areas Commun., vol. 14, no. 4, pp. 711-717, May 1996. [18] M. Ghaderi, R. Boutaba, and G. W. Kenward, “Stochastic admission control for quality of service in wireless packet networks,” in Proc. 4th IFIP Networking Conference, pp. 1309-1320, May 2005. [19] Y. -B. Lin, “Reducing location update cost in PCS network,” IEEE/ACM Trans. Networking, vol. 5, pp. 25-33, Feb. 1997.



Page 9 of 9