Parameter estimation for systems with binary subsystems

Consider a stochastic system of multiple subsystems, each subsystem having binary (“0 or 1”) output. The full system may have general binary or non-binary (e.g., Gaussian) output. Such systems are widely encountered in practice, and include engineering systems for reliability, communications and sensor networks, the collection of patients in a clinical trial, and Internet-based control systems. This paper considers the identification of parameters for such systems for general structural relationships between the subsystems and the full system. Maximum likelihood estimation (MLE) is used to estimate the mean output for the full system and the “success” probabilities for the subsystems. The MLE approach is well suited to providing asymptotic or finite-sample confidence bounds through the use of Fisher information or bootstrap Monte Carlo-based sampling. Three examples are presented to illustrate the method.

[1]  Ori Davidov,et al.  Order-Restricted Inference for Multivariate Binary Data With Application to Toxicology , 2011, Journal of the American Statistical Association.

[2]  Donghua Zhou,et al.  Leakage Fault Diagnosis for an Internet-Based Three-Tank System: An Experimental Study , 2012, IEEE Transactions on Control Systems Technology.

[3]  James C. Spall.  Convergence analysis for maximum likelihood-based reliability estimation from subsystem and full system tests , 2010, 49th IEEE Conference on Decision and Control (CDC).

[4]  James C. Spall,et al.  Asymptotic normality and uncertainty bounds for reliability estimates from subsystem and full system tests , 2012, 2012 American Control Conference (ACC).

[5]  R. Serfling.  Approximation Theorems of Mathematical Statistics , 1980 .

[6]  R. A. Waller,et al.  Bayesian reliability analysis of complex series/parallel systems of binomial subsystems and components , 1990 .

[7]  James C. Spall,et al.  System reliability estimation and confidence regions from subsystem and full system tests , 2009, 2009 American Control Conference.

[8]  Z. Birnbaum,et al.  Modules of Coherent Binary Systems , 1965 .

[9]  Han La Poutré,et al.  A Versatile Approach to Combining Trust Values for Making Binary Decisions , 2006, iTrust.

[10]  Nick G. Duffield,et al.  Network Tomography of Binary Network Performance Characteristics , 2006, IEEE Transactions on Information Theory.

[11]  Coire J. Maranzano,et al.  Robust test design for reliability estimation with modeling error when combining full system and subsystem tests , 2010, Proceedings of the 2010 American Control Conference.

[12]  Harry F. Martz,et al.  Bayesian reliability analysis of series systems of binomial subsystems and components , 1988 .

[13]  James C. Spall,et al.  Introduction to stochastic search and optimization - estimation, simulation, and control , 2003, Wiley-Interscience series in discrete mathematics and optimization.

[14]  James C. Spall,et al.  Framework for estimating system reliability from full system and subsystem tests with dependence on dynamic inputs , 2011, IEEE Conference on Decision and Control and European Control Conference.

[15]  B. Hoadley.  Asymptotic Properties of Maximum Likelihood Estimators for the Independent Not Identically Distributed Case , 1971 .