Application of AI techniques for locating mobile terminals in mobile wireless cellular networks: A novel paging scheme

Abstract

This work proposes a novel intelligent terminal paging scheme, which essentially employs a popular concept of AI, namely, rule-based technique. The scheme is so devised that it can design ''rules'' that potentially maps apparently disparate input data items (though having some regular stochastic relations) to an almost exact position of mobile terminals (MTs). The scheme, called RBPS^*, has been developed utilizing the conventional models of call arrival, cell residence time and microscopic mobility. As a mobility model, random-walk mobility and movement-based model have been utilized. Hence, a stochastic model has been developed to analyze the performance of the scheme. RBPS^* requires no additional processing overhead at MTs' end, though it involves a nominal overhead at the MSC end. This scheme not only attempts to resolve the issue of inherent tradeoff between the cost components, comprising paging and location update (LU) cost, it also reduces the total cost for location management, compared to the other schemes referred in this work. Simulation results can successfully demonstrate that the proposed RBPS^* significantly outperforms GSM-adopted blanket paging scheme and the popular shortest distance first (SDF) scheme. Another important point to be noted is that the proposed scheme, RBPS^*, is a generic one and can be deployed on-line. This scheme can potentially be also utilized in next generation wireless cellular networks, irrespective of any standards, by undertaking minor modifications, as required to conform to the respective standards.