Abstract
With the evolution towards 4G cellular networks, there is a need to develop new approaches for radio network planning (RNP) that can capture the technology enhancements to determine optimized locations and configurations of base station sites. Conventional RNP approaches are normally based on a deterministic link budget analysis that compensates for signal statistical variation via pre-determined power margins; this is normally followed by Monte-Carlo simulations to fine tune the planning outcome. In this work, we present a novel approach for cellular RNP that captures the uncertainty in signals and interference as part of the problem formulation leading directly to an optimized planning solution; the approach is generic and can capture signal uncertainty due to fading and dynamic resource allocation with possible extension to other adaptive system features. The approach is divided into two parts that are addressed separately using chance constrained optimization and then combined within a common framework. The first part, denoted as site selection problem, aims at selecting the minimum cardinality set of eNodeBs from a given large set, that satisfies target performance requirements. The second part, denoted as site placement problem, aims at refining the locations of the selected eNodeBs in order to further enhance the planning quality. Finally, both parts are combined within a common framework to determine the optimized number and locations of eNodeBs over a given geographical area based on a wide range of system and user parameters. A divide and conquer approach is also proposed to deal with the complexity of planning large network scenarios. Performance results are presented to highlight the effectiveness of the proposed framework with detailed analysis and verification via Monte-Carlo simulations.