Radio Network Planning and Resource Optimization : Mathematical Models and Algorithms for UMTS, WLANs, and Ad Hoc Networks

Abstract: The tremendous popularity of wireless technologies during the last decade has created a considerable expansion of wireless networks both in size and use. This fact, together with a great variety of mobile devices and numerous di®erent services that are becoming increasingly resourcedemanding, have attracted the attention of many researchers into the area of radio resource planning and optimization. Due to network complexity, these tasks require intelligent, automated approaches that are able to deal with many factors in order to enable design of high capacity networks with a high service quality at the lowest possible cost. This is a perfect application of optimization theory.In this thesis, mathematical optimization is considered as the main approach to designing and improving the performance of wireless networks such as Universal Mobile Telecommunications System (UMTS), Wireless Local Area Networks (WLANs) and ad hoc networks. Due to different underlying access technologies, the optimization goals, design parameters and system limitations vary by network type. Therefore, the goals of the presented work are to identify a relevant optimization problem for each type of network, to model the problem and to apply the optimization approach in order to facilitate wireless network planning and improve radio resource utilization.The optimization problems addressed in this thesis, in the context of UMTS networks, focus on minimizing the total amount of pilot power which, from the modeling point of view, is not just an amount of power consumed by a certain type of control signal, but also an indicator of the interference level in the network and means of controlling cell coverage. The presented models and algorithms enable °exible coverage planning and optimization of pilot power and radio base station antenna confiration in large networks.For WLANs, in the First part of the study, the access point placement and the channel assignment problems are considered jointly to maximize net user throughput and minimize co- and adjacent channel interference and contention. The second part of the study addresses the contention issue and involves, among the other decisions, optimization of access point transmit power.Due to the dynamic and infrastructureless nature of ad hoc networks, static resource planning is less suitable for this type of network. Two algorithmic frameworks which enable dynamic topology control for power-efficient broadcasting in stationary and mobile networks are presented. In both frameworks, the performance of the presented algorithms is studied by simulations.