Movement prediction oriented adaptive location management provides a major role in personal communication service (PCS) system. Generally the GSM system supports two level architecture. Because it supports two kinds of databases-Home Location register and Visitor Location Register. Every time when the user crosses the location area it has to register with the HLR. This creates high cost for registration and location tracking as it involves the use of costly bandwidth between the Mobile Switching Center (MSC) and the HLR. In this paper the technique for reducing the costs during the location tracking and location update is proposed. Taking the movement prediction of the users it creates the block and the user registers with the HLR only after crossing the block instead of crossing the single cell. This movement prediction is generated using one neural network model for all the users. The block register (BR) is introduced between the block and the HLR in two level systems, thus introduces three level architecture. In this architecture some signaling cost values between the MSC-BR, BR-HLR and BR-BR are maintained to get the better performance. In this proposed system the aim is to set the value between the MSC and BR and the two BR as small as possible and the value between the BR and the HLR must be higher to get the better performance.
The enhanced population of mobile users of mobile network needs a reliable and uninterrupted communication. Due to increasing capacity of mobile users the signaling traffic is expanding day by day. In this situation personal communication service (PCS) network provides the efficient and uninterrupted communication to the mobile subscribers. PCS users are carrying mobile stations can communicate with the remote mobile regardless of their current location. There are two commonly used standards for mobility management. In North America IS-41 is used and in Europe Global System for Mobile Communication (GSM) is commonly used. We have used the GSM standard for the overall discussion and we will discuss the performance analysis of our proposed system with the standard GSM system.
Generally PCS network supports cellular architecture. Each cell consists a base station. When a mobile station (MS) resides in a cell it always communicates with the nearest base station. The coverage of the cell varies from few hundred meters to several km. Each cellular base station is allocated a group of radio channels to be used within a cell. The network sends the massage to the mobile station through the Mobile Switching Center (MSC). MSC acts as a bridge between the wired network and the wireless network. Mainly the base stations are connected with the MSC. Several cells constitute the location area (LA). One location area is controlled by one MSC. And each cell contains one base station. The wire line network carries user information and signaling among the MSCs and the location databases. In Figure 1, Public Switched Telephone Network (PSTN) is used as wired network. GSM standard is a two level database architecture because there are two kinds of databases in GSM architecture – Home Location Register (HLR) and Visitor Location Register (VLR). Generally the VLRs are coupled with the MSCs. It stores the location information of the users in its service area. HLR contains the permanent information of the users and pointing to the current serving VLR. HLR acts as a centralized database for storing all the user’s profile. The VLRs are distributed throughout the PCS network. When the MS moves to the new MSC, signaling messages are exchanged among the HLR and the new and old VLR to record the current location of the mobile user. In the same way during the call tracking procedure when the call is going from the calling MS to the called MS the signaling messages are exchanged among the VLRs of calling MS and the VLRs of called MS. We will be discussed the registration and call tracking procedure in the following sections because the registration and the call tracking procedures are the major part of location management.
PCS signaling network architecture
When we are discussing about the location management, first we have to define the concept of mobility management. Mobility management is the technology that supports roaming users with mobile stations to enjoy their services through wireless networks when they are moving into a new service area and location management means locating that station to deliver the calls without regarding the fact that there locations may change from time to time.
Signaling system of the PCS system contains Signal transfer point (STP) and Service Control Point (SCP). STP acts as a switch for routing the signaling message to the destination and SCP contains all the databases and also handles database query and database update initiated by MSC. STPs are installed in pairs .This is shown in Figure 2.
When we are talking about the location management we have to consider two processes: Location Registration and Location Tracking. As our main aim is to reduce the location management cost so first we have to discuss the details of the location registration and location tracking process for standard GSM system.
Key Terms in this Chapter
Movement Pattern: Movement Pattern is the history of movement of a mobile host recorded during a period of time interval.
Mobile Station: The mobile station (MS) refers to the equipments of the user and software needed for the communication with other MS of the network. It consists subscriber identity module (SIM), which stores all the user specific data.
Location Tracking: It is the process of searching the user’s location from registration data.
Neural Network: When the work is performed on artificial neural network then it is known as neural network. The knowledge is acquired by the neural network from the environment through the learning process.
Block Register: Block register is a special kind of register that holds two kinds of pointers and it directly or indirectly points to the MSC of its block. This kind of block is introduced between the HLR and the MSC of our proposed system.
Call to Mobility Ratio: It is referred to as the call to mobility ratio. When the CMR is low the mobility rate is high so that the registration cost will be high. When the CMR is high the mobility is low so that in this situation the registration cost will be low.
Location Prediction: Location Prediction means calculating the user’s future location from historical movement data.
Complete Chapter List
Elhadi Shakshuki, Xinyu Xing, Haroon Malik
Reinhard Kronsteiner, Bettina Thurnher
Goran Gvozden, Mislav Grgic, Sonja Grgic, Miran Gosta
Mamun I. Abu-Tair
Abdulhussain E. Mahdi
Wanji Mai, Chris Tweed, Peter Hung, Seán McLoone, Ronan Farrell
Eduardo Antonio Viruete Navarro
Paolo Barsocchi, Alan A. Bertossi, M. Cristina Pinotti, Francesco Potortì
Do van Thanh, Ivar Jørstad
Yoshio Nakajima, Alireza Goudarzi Nemati, Tomoya Enokido, Makoto Takizawa
Ben Abdallah Abderazek, Arquimedes Canedo, Kenichi Kuroda
Wieland Schwinger, Christoph Grün, Birgit Pröll, Werner Retschitzegger
Daniel C. Doolan, Sabin Tabirca, Laurence T. Yang
Daniel C. Doolan, Sabin Tabirca, Laurence T. Yang
Daniel C. Doolan, Kevin Duggan, Sabin Tabirca, Laurence T. Yang
Christos K. Georgiadis
Hongbo Ni, Xingshe Zhou, Zhiwen Yu, Daqing Zhang
Pavol Podhradský, Eugen Mikóczy, Matejka Juraj, Ondrej Lábaj, Róbert Tomek
Robert Schmohl, Uwe Baumgarten, Lars Köthner
Roman Y. Shtykh, Qun Jin, Shunichi Nakadate, Norihiro Kandou, Takeshi Hayata, Jianhua Ma
Stephan Reiff-Marganiec, Yi Hong, Hong Qing Yu, Schahram Dustdar, Christoph Dorn, Daniel Schall
Baud Haryo Prananto
Diego Moreira Alves
Dietmar G. Wiedemann
Mahieddine Djoudi, Saad Harous
Patrícia Dockhorn Costa, Luís Ferreira Pires, Marten van Sinderen
Frédéric Lassabe, Philippe Canalda, Damien Charlet, Pascal Chatonnay, François Spies
Anastasis A. Sofokleous, Marios C. Angelides, Christos N. Schizas
Wee Hyong Tok, Stéphane Bressan, Panagiotis Kalnis, Baihua Zheng
Ioannis Priggouris, Evangelos Zervas, Stathes Hadjiefthymiades
Ghita Kouadri Mostéfaoui
Do Van Thanh, Ivar Jørstad, Schahram Dustdar
Mohamed Ali Feki
Damien Charlet, Frédéric Lassabe, Philippe Canalda, Pascal Chatonnay, François Spies
Roland Wagner, Franz Gruber, Werner Hartmann