System Identification Using Regular and Quantized Observations: Applications of Large Deviations Principles

​This brief presents characterizations of identification errors under a probabilistic framework when output sensors are binary, quantized, or regular. By considering both space complexity in terms of signal quantization and time complexity with respect to data window sizes, this study provides a new perspective to understand the fundamental relationship between probabilistic errors and resources, which may represent data sizes in computer usage, computational complexity in algorithms, sample sizes in statistical analysis and channel bandwidths in communications. Table of Contents Cover System Identification Using Regular and Quantized Observations - Applications of Large Deviations Principles ISBN 9781461462910 ISBN 9781461462927 Preface Contents Notation and Abbreviations 1 Introduction and Overview 2 System Identification Formulation 3 Large Deviations: An Introduction 4 LDP of System Identification under Independent and Identically Distributed Observation Noises 4.1 LDP of System Identification with Regular Sensors 4.2 LDP of System Identification with Binary Sensors 4.3 LDP of System Identification with Quantized Data 4.4 Examples and Discussion 4.4.1 Space Complexity: Monotonicity of Rate Functions with Respect to Numbers of Sensor Thresholds 5 LDP of System Identification under Mixing Observation Noises 5.1 LDP for Empirical Means under f-Mixing Conditions 5.2 LDP for System Identification with Regular Sensors under Mixing Noises 5.3 LDP for Identification with Binary Sensors under Mixing Conditions 6 Applications to Battery Diagnosis 6.1 Battery Models 6.2 Joint Estimation of Model Parameters and SOC 6.3 Convergence 6.4 Probabilistic Description of Estimation Errors and Diagnosis Reliability 6.5 Computation of Diagnosis Reliability 6.6 Diagnosis Reliability via the Large Deviations Principle 7 Applications to Medical Signal Processing 7.1 Signal Separation and Noise Cancellation Problems 7.2 Cyclic System Reconfiguratio for Source Separation and Noise Cancellation 7.2.1 Cyclic Adaptive Source Separation 7.2.2 Cyclic Adaptive Signal Separation and Noise Cancellation 7.3 Identification Algorithms 7.3.1 Recursive Time-Split Channel Identification 7.3.2 Inversion Problem and Optimal Model Matching 7.4 Quality of Channel Identification 7.4.1 Estimation Error Analysis for ANC 7.4.2 Signal/Noise Correlation and the Large Deviations Principle 8 Applications to Electric Machines 8.1 Identification of PMDC-Motor Models 8.2 Binary System Identification of PMDC Motor Parameters 8.3 Convergence Analysis 8.4 Quantized Identification 8.5 Large Deviations Characterization of Speed Estimation 9 Remarks and Conclusion 9.1 Discussion of Aperiodic Inputs 9.2 Escape from a Domain 9.3 Randomly Varying Parameters 9.4 Further Remarks and Conclusions References Index