Importance sampling. Applications in communications and detection
This research monograph deals with fast stochastic simulation based on importance sampling (IS) principles and some of its applications. Chapter 1 introduces the concept of IS with emphasis on estimation of rare event probabilities. Different biasing methods are described in Chapter 2 and evaluated in terms of the variances of the estimates that they provide. The concept of adaptive biasing is introduced in this chapter and optimization algorithms are developed. The IS simulation problem is posed as one of variance minimization using stochastic algorithms. Chapter 3 is devoted to sums of random variables. Tail probability estimation is discussed and a method based on conditional probabilities is developed. A simulation methodology for determining a number that is exceeded by sums of random variables with a given small probability is formulated. It is referred to as the inverse IS problem and forms the basis for parameter optimization in systems to achieve specified performance levels. This has several practical applications, as demonstrated by numerical examples. In the same chapter, an approximation for tail probabilities is derived. The derivation of the Srinivasan density, an approximation for densities of sums of independent and identically distributed random variables, is given. Several simulation examples are given in these chapters to illustrate the nature of results that can be expected from well designed IS algorithms. In Chapter 4, approximations developed in Chapter 3 are used to obtain analytical expressions for false alarm and detection probabilities. The remaining chapters, 5 to 8, are on applications of the IS techniques discussed previously. Chapter 5 presents an effective solution for applying IS to constant false alarm rate detection algorithms that are used in radar and sonar receivers. It is shown how adaptive techniques can be applied to their analysis and design. Several detection situations are described and numerical results provided. Results are provided in Chapter 6 on ensemble detection, a technique that combines the outputs of several processors to achieve robustness properties. In Chapter 7, blind simulation is discussed, where the statistical distributions of the underlying processes may be unknown or partially known. It is applied to a detection algorithm to demonstrate its capabilities. Chapter 8 deals with performance evaluation of digital communication systems that cannot be handled analytically or even using standard numerical techniques. Parameter optimization is also addressed. Several examples are given that serve to illustrate how adaptive IS can be used for such systems. Some of the application examples in these last chapters are treated briefly in regard to their setting and mathematical background. This monograph is intended for researchers, analysts, designers, and advanced students, who are interested in fast simulation based on adaptive IS methods. Note that exercises are not provided in this book.
- Some results in importance sampling and an application to detection
- The sample size required in importance sampling
- Combining multiple surrogate models to accelerate failure probability estimation with expensive high-fidelity models
- Bayesian learning of stochastic dynamical models
- \texttt{CAMERA}: a method for cost-aware, adaptive, multifidelity, efficient reliability analysis
- A continuous coordinate system for the plane by triangular symmetry
- A population balance-Monte Carlo method for particle coagulation in spatially inhomogeneous systems
- Performance evaluation via Monte Carlo importance sampling in single user digital communication systems
- Multifidelity importance sampling
- Optimizing multi-objective decision making having qualitative evaluation
- A randomized bias technique for the importance sampling simulation of Bayesian equalizers
- Importance sampling for analysis of direct detection optical communication systems
- A statistical method for verifying mesh convergence in Monte Carlo simulations with application to fragmentation
- Multifidelity preconditioning of the cross-entropy method for rare event simulation and failure probability estimation
- Computation of credit portfolio loss distribution by a cross entropy method
- A statistical approach to the inverse problem in magnetoencephalography
- An efficient surrogate-based method for computing rare failure probability
- A note on estimating false alarm rates via importance sampling
- Importance sampling for the random phase Gaussian channel
- Importance Sampling Evaluation of Digital Phase Detectors With Carrier Phase Tracking: A Large Deviations Approach
- Stochastic finite element methods for partial differential equations with random input data
- scientific article; zbMATH DE number 1928750 (Why is no real title available?)
- Importance sampling techniques in neural detector training
- Efficient Computation of Extreme Excursion Probabilities for Dynamical Systems through Rice's Formula
- Ensemble approximate control variate estimators: applications to multifidelity importance sampling
- Importance Sampling for Generalized Likelihood Ratio Procedures in Sequential Analysis
- Geometrically nonlinear behavior of structural systems with random material property: an asymptotic spectral stochastic approach
- Counting the number of group orbits by marrying the Burnside process with importance sampling
- Importance sampling via a simulacrum
- Data-based importance sampling estimates for extreme events
This page was built for publication: Importance sampling. Applications in communications and detection
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4778523)
