Service Coverage Bounds Through Efficient Load Approximation in UMTS Network Planning

Cell planning requires transmission site selection and configuration. Because this is an NP-hard problem, approximate solutions have to be found when problem instances are large. As such bounds on optimal solutions have to be determined. This is particularly challenging for CDMA-based systems such as 3G because coverage, load and interference are dependent variables. In this paper we first formulate and derive bounds on maximum achievable service coverage ratio subject to the available resources. We then exploit a load approximation algorithm, which substantially reduce computational complexity for downlink formulations of the problem while ensuring that no cell is overloaded. The simulation results show that our approximation algorithm can achieve good coverage accuracy in comparison to the obtained bounds using much lesser computational time at the cost of slightly higher power requirement.