طراحی و تجزیه و تحلیل عملکرد از استراتژی های مدیریت مکان سلسله مراتبی برای سیستم های ارتباطی بی سیم تلفن همراه

In this paper, we propose three hierarchical location management strategies for wireless mobile communication systems. The first one is home server first (HSF) scheme in which callee's location is queried at callee's home location server first. Next, we propose least-common-ancestor server first (LSF) scheme in which a search starts at the least-common-ancestor server of caller's current location and callee's home server. Finally, we propose distributed LSF in which the location management is made on 2-layered distributed structure. We evaluate the performance of the proposed schemes and compare it with that of another hierarchical scheme proposed by Wang (J.Z. Wang, A fully distributed location registration strategy for universal personal communication systems, IEEE J. Select. Areas Commun., 11(6) (1993) 850–860) and IS-41-based scheme. It is shown that the proposed schemes have reduced location management cost and that the cost is less affected by the increment of communication cost than the compared schemes.

In wireless mobile communication systems, the exchange of information with the mobile terminals should be facilitated regardless of their locations. To meet this requirement, the network should have well-defined functions to find out the access points of the mobiles. Thus, the design of an efficient location management strategy is one of the important issues in wireless mobile network [1], [2] and [3]. Much research on location management has been conducted in two large classes: centralized and distributed strategies. The centralized location management strategies are based on 2-layered structure having the centralized database (Home Location Register, HLR) with the multiple number of the local database (Visitor Location Register, VLR). Examples of the centralized strategies are Electric/Telecommunications Industry Associations (EIA/TIA), Interim Standard 41 (IS-41) [4] and Global Systems for Mobile Communications (GSM) Mobile Application Part (MAP) [5]. The modified versions of these standards such as per-user caching scheme [6], user profile replication scheme [7], pointer forwarding scheme [8] and local anchoring scheme [9] were proposed to reduce the message traffic and the database processing load at the central HLR. The distributed location management strategies have been proposed to solve the latent problems of IS-41 and GSM MAP standards and their modified versions. Considering the large number of subscribers in future wireless mobile communication systems, the centralized management strategy will become impractical due to the signaling traffic congestion and the high-database processing load at the central database. To solve these latent problems, fully distributed scheme [10] and hierarchically distributed scheme [1] were proposed for personal communication systems. The hierarchical Private Network-to-Network Interface (PNNI)-based location management for wireless Asynchronous Transfer Mode (ATM) networks was proposed and is under discussion in Wireless ATM Forum [11]. In addition, an algorithm to determine the optimal database placement has been proposed in Ref. [12] to reduce the location update and access load on each database on the hierarchical network structure. The fixed transport networks of mobile communication systems such as cellular, personal communication services (PCS), wireless ATM and international mobile telecommunication 2000 (IMT-2000) networks are structured in a hierarchical way. Hence, the distributed location management based on the hierarchical manner is attractive for the next generation mobile networks. In this paper, we propose three location management strategies based on hierarchical network structure. The first one is home server first (HSF) scheme, the second one is least-common-ancestor server first (LSF) scheme and the last one is distributed LSF scheme. We evaluate the expected location registration and call delivery costs with consideration given to the cost of signaling message transmission and database processing. In addition, we compare the location management cost of the proposed strategies with that of a hierarchical scheme proposed in Ref. [1] and an IS-41-based scheme.

In this paper, we propose the new location management schemes based on the hierarchical network structure, i.e. HSF, LSF and distributed LSF schemes. We evaluate the location registration cost and the call delivery cost as well as the expected number of the database accesses, with consideration given to hierarchical network model proposed in Ref. [1] and IS-41-based system. In addition, we examine the effects of the database processing cost and the communication cost on the total location management cost. From the analytical results, we can conclude the following. • IS-41-based scheme shows the relatively low and constant number of database accesses for location registration and call delivery. The number of database entries at central HLR of IS-41-based scheme is, however, larger than that of the proposed strategies when the terminal mobility increases. Considering the signaling traffic congestion around HLR caused by this fact in the real environment, the IS-41-based scheme may show the worse performance than the result shown in this paper. • Distributed LSF scheme is expected to be most attractive among the schemes considered in this paper for the most regions of pout. This scheme can significantly decrease the location management cost with the small increase in the number of the location servers from IS-41-based systems. • The location management cost of the three proposed schemes is less affected by the magnitude of communication cost than that of IS-41-based scheme and Wang's scheme. These results can be utilized not only for evolving the location management of current cellular and PCS systems but also for designing future wireless mobile communication systems such as wireless ATM networks and IMT-2000 services. Acknowledgements The authors would like to thank the anonymous reviewers of this paper for their precious comments and suggestions, which helped to improve the original version of this manuscript.