Optimising RRM algorithms via simulations

The paper describes two projects for optimising radio resource management (RRM) algorithm parameters using WCDMA dynamic simulations. The first project was conducted to determine the optimal QoS for different services. A series of simulations with different service BLER (block error rate) targets were conducted and network performance monitored. One significant conclusion from this project is that, by reducing the BLER target for packet switched (PS) services from 10% to 5%, the overall network performance criteria is achieved without a significant reduction of network capacity. The second project evaluated the effect of different service mixes on network performance. This has been achieved for the same user (service) profile accessing the network but with changing parameters that refer to the initial load contribution of different bearers (as used by the admission control algorithm). The objective of this study was to maximise the network throughput/capacity without compromising either CS or PS data call grade of service. One of the conclusions from this analysis is that the network throughput is improved by "favouring" allocation of 64 kbps and 128 kbps compared to allocation of 384 kbps for PS service downlink data rates. A large number of RRM parameters are configurable by either operator or equipment vendor and they are largely interdependent. Therefore, RRM parameter optimisation using this method is ongoing and performed in parallel with 3G network development.