Efficient performance evaluation of spread spectrum transmission systems

Performance evaluation of complex transmission systems is generally accomplished by extensive Monte Carlo simulations. Especially for spread spectrum systems with highly complex transmission waveforms and sophisticated receiver structures, the simulation run-times tend to become extremely high even on powerful hardware platforms. A run-time efficient, but less well-known and less easy-to-use alternative to conventional Monte Carlo simulation is importance sampling (IS). Importance sampling potentially decreases simulation run-times by up to some orders of magnitude. However, application of this simulation technique so far has been limited to isolated and rather theoretical problems. Importance sampling is applied to evaluate the synchronization properties of a complex real-world spread spectrum system. The various steps required for designing the simulation experiment are derived, and simulation run-times as well as simulation results are compared to those obtained by the conventional approach. In the course of the case study, a novel system decomposition approach is presented which enables application of importance sampling to a wide range of spread spectrum systems