Multi-objective site selection and analysis for GSM cellular network planning
Although considerable effort has been placed on developing techniques and algo rithms to create feasible cell plans, much less effort has been placed on understanding the relationship between variables and objectives. The purpose of this thesis is to improve the body of knowledge aimed at understanding the trade-offs and tensions in the selection of transmission sites and in the configuration of macro-cells for GSM and related FDMA wireless systems. The work begins by using an abstract 2-dimensional (2D) model for area coverage. A multiple objective optimisation framework is de veloped to optimise the sequential placement and configuration of downlink wireless cells. This is deployed using a range of evolutionary algorithms whose performance is compared. The framework is further tuned via a decoding mechanisms using the best performing evolutionary algorithm. The relationship between primary variables in the 2D model is analysed in detail. To improve realism, the thesis additionally addresses complexities relating to planning in 3-dimensional (3D) environments. A detailed open source static model is developed and the optimisation framework is extended to accommodate the additional model complexities and choices in algorithm design are compared. Finally, sensitivity analysis is performed to determine the relationship between objectives in the 3D model and benchmark solutions are provided.
92 references, page 1 of 10
views in local repository
downloads in local repository
The information is available from the following content providers: