Optimizing cellular phone coverage in dense urban environments using 3d geo-information: A case study of the Nairobi central business district, Kenya

Abstract

Optimal signal coverage of a target area has always been a fundamental issue for cellular network operators. Other issues related to capacity, quality of service and cost efficiency are also rapidly gaining prominence. Thus, computer-based planning tools to aid wireless network engineers plan the positioning of Base Transceiver Stations (BTSs) such that optimal coverage of the target area is achieved, are increasingly being sought. To achieve optimal mobile signal coverage in an urban area, characterized by numerous high-rise buildings, is particularly challenging due to its complex nature.

In order to determine signal coverage, engineers rely mainly on two dimensional (20) terrain maps and rather simple empirical propagation-prediction models that nevertheless provide inaccurate results. In this research, a framework which provides a more efficient and cost effective network coverage optimization for a dense urban environment was investigated. A Geographic Information System (GIS) of the study area was developed comprising three-dimensional (3~) geo-information databases. The signal propagation-prediction tool based on ray-tracing coupled with the 3D geo-information was used to model the radio signal coverage for the BTSs. In order to verify the accuracy of the propagation model employed, signal strengths obtained from the propagation model were compared with those obtained from field measurements for two of the existing BTSs for one of the mobile phone operators licensed to provide mobile phone services in Kenya, i.e. Zain Kenya. The ray-tracing propagation model was found to be quite accurate.

To determine the best locations of the BTSs for optimal signal coverage of the study area, spatial analysis tools in GIS were used. The ray-tracing propagation model was used to model the signal coverage for the determined BTSs. The results . indicated optimal signal coverage of the study area. This framework can thus be used to optimize mobile signal coverage in any dense urban environment without resorting to lengthy field measurement to enhance the accuracy and thus minimizing on costs of wireless network planning