Multi-parameter optimization in WCDMA radio networks

This paper proposes a novel sensitivity-matrix-based multi-parameter optimization algorithm, aiming to optimize the performance of WCDMA radio access networks (RAN) in an adaptive and automated way. A correlation approach is proposed to examine the sensitivity between key performance indicators (KPIs) and tuned parameters. Simulation results proved that the proposed algorithm is effective to lead the network performance to the optimum point and achieve around 10% improvement in throughput and call success rate (CSR), and 2-5% improvement in link power outage and bad-quality (BQ) call rate in a simulated hexagonal macro-cell network.