Importance Sampling

the long version

This monograph on fast stochastic simulation deals with methods of adaptive importance sampling (IS). The concept of IS is introduced and described in detail with several numerical examples in the context of rare event simulation. Adaptive simulation and system parameter optimization to achieve specified performance criteria are described. The techniques are applied to the analysis and design of radar CFAR (constant false alarm rate) detectors. Development of robust detection algorithms using ensemble - or E-CFAR processing is described. A second application treats the performance evaluation and parameter optimization of digital communication systems that cannot be handled analytically or even by using standard numerical techniques.