Explicit characterization of tensions in downlink cell planning

Providing downlink received signal strength involves balancing trade-offs between service coverage objectives and cost minimisation, while taking into account radio engineering constraints concerning interference, cell overlap for call handover and traffic capacity. We describe a framework to characterise explicitly the choices available in 2G and 2.5G GSM networks. To evaluate the trade-off between coverage and costs, we apply a self-learning evolutionary approach that mimics the potential operations of a manual planner. At the same time, realistic operational requirements are incorporated.