-
Book
-
Chapter
System Settings
This chapter presents basic system structures, sensor representations, input types and characterizations, system configurations, and uncertainty types for the entire book. This chapter provides a problem formu...
-
Chapter
Worst-Case Identification Using Quantized Observations
In this chapter, the parameter identification problem under unknown-butbounded disturbances and quantized output sensors is discussed. In Chapter 9, an input sequence in (9.5) was used to generate observation ...
-
Chapter
Estimation Error Bounds: Including Unmodeled Dynamics
In Chapter 3, we derived convergent estimators of the system parameters using binary-valued observations. Our aim here is to obtain further bounds on estimation errors from unmodeled dynamics. In this book, un...
-
Chapter
Identification of Hammerstein Systems with Quantized Observations
This chapter concerns the identification of Hammerstein systems whose outputs are measured by quantized sensors. The system consists of a memoryless nonlinearity that is polynomial and possibly noninvertible, ...
-
Chapter
Quantized Identification and Asymptotic Efficiency
Up to this point, we have been treating binary-valued observations. The fundamental principles and basic algorithms for binary-valued observations can be modified to handle quantized observations as well. One ...
-
Chapter
Space and Time Complexities, Threshold Selection, Adaptation
The number m0 of thresholds is a measure of space complexity, whereas the observation length N is a measure of time complexity that quantifies how fast uncertainty can be reduced. The significance of understandin...
-
Chapter
Identification of Sensor Thresholds and Noise Distribution Functions
The developments in the early chapters rely on the knowledge of the distribution function F· or its inverse, as well as the threshold C. However, in many applications, the noise distributions are not known, or on...
-
Chapter
Introduction
This book studies the identification of systems in which only quantized output observations are available. The corresponding problem is termed quantized identification.
-
Chapter
Identification of Wiener Systems with Binary-Valued Observations
This chapter studies the identification of Wiener systems whose outputs are measured by binary-valued sensors. The system consists of a linear FIR (finite impulse response) subsystem of known order, followed b...
-
Chapter
Rational Systems
The systems in Chapters 3 and 4 are finite impulse-response models. Due to nonlinearity in output observations, switching or nonsmooth nonlinearity enters the regressor for rational models. A common technique ...
-
Chapter
Input Design for Identification in Connected Systems
Input design is of essential importance in system identification to provide sufficient probing capabilities to guarantee the convergence of parameter estimators to their true values; namely, the estimators are...
-
Chapter
Impact of Communication Channels on System Identification
This chapter deals with the identification of systems whose outputs must be quantized, transmitted through a communication channel, and observed afterwards. Communication errors introduce additional uncertaint...
-
Chapter
Worst-Case Identification under Binary-Valued Observations
This chapter focuses on the identification of systems where the disturbances are formulated in a deterministic framework as unknown but bounded. Different from the previous chapters, here the identification er...
-
Chapter
Empirical-Measure-Based Identification: Binary-Valued Observations
This chapter presents a stochastic framework for systems identification based on empirical measures that are derived from binary-valued observations. This scenario serves as a fundamental building block for su...
-
Chapter
Systems with Markovian Parameters
This chapter concerns the identification of systems with time-varying parameters. The parameters are vector-valued and take values in a finite set. As in the previous chapters, only binary-valued observations ...
-
Book
System Identification Using Regular and Quantized Observations
Applications of Large Deviations Principles
-
Chapter
Introduction and Overview
Traditional system identification taking noise measurement into consideration concentrates on convergence in suitable senses (such as in mean square, in distribution, or with probability one) and rates of conv...
-
Chapter
Large Deviations: An Introduction
The theory of large deviations characterizes probabilities and moments of certain sequences that are associated with “rare” events. In a typical application, consider the sum of N independent and identically dist...
-
Chapter
LDP of System Identification under Mixing Observation Noises
Up to this point, the observation noises are assumed to be uncorrelated. In this chapter, we demonstrate that a much larger class of noise processes can be treated.