Typescript (photocopy). / Digitized by Kansas Correctional Industries

Thesis (M.Sc.)--York University, 2008. Graduate Programme in Science. / Typescript. Includes bibliographical references (leaves 223-232). Also available on the Internet. MODE OF ACCESS via web browser by entering the following URL: http://gateway.proquest.com/openurl?url_ver=Z39.88-2004&res_dat=xri:pqdiss&rft_val_fmt=info:ofi/fmt:kev:mtx:dissertation&rft_dat=xri:pqdiss:MR45965

Optimal input design concerns the design of an input sequence to maximize the information retrieved from an experiment. The design of the input sequence is performed by optimizing a cost function related to the intended model application. Several approaches to input design have been proposed, with results mainly on linear models. Under the linear assumption of the model structure, the input design problem can be solved in the frequency domain, where the corresponding spectrum is optimized subject to power constraints. However, the optimization of the input spectrum using frequency domain techniques cannot include time-domain amplitude constraints, which could arise due to practical or safety reasons. In this thesis, a new input design method for nonlinear models is introduced. The method considers the optimization of an input sequence as a realization of the stationary Markov process with finite memory. Assuming a finite set of possible values for the input, the feasible set of stationary processes can be described using graph theory, where de Bruijn graphs can be employed to describe the process. By using de Bruijn graphs, we can express any element in the set of stationary processes as a convex combination of the measures associated with the extreme points of the set. Therefore, by a suitable choice of the cost function, the resulting optimization problem is convex even for nonlinear models. In addition, since the input is restricted to a finite set of values, the proposed input design method can naturally handle amplitude constraints. The thesis considers a theoretical discussion of the proposed input design method for identification of nonlinear output error and nonlinear state space models. In addition, this thesis includes practical applications of the method to solve problems arising in wireless communications, where an estimate of the communication channel with quantized data is required, and application oriented closed-loop experiment design, where quality constraints on the identified parameters must be satisfied when performing the identification step. / <p>QC 20141110</p>

Sparse parameter estimation is an important aspect of system identification, as it allows for reducing the order of a model, and also some models in system identification inherently exhibit sparsity in their parameters. The accuracy of the estimated sparse model depends directly on the performance of the sparse estimation methods. It is well known that the accuracy of a sparse estimation method relies on the correlations between the regressors of the model being estimated. Mutual coherence represents the maximum of these correlations. When the parameter vector is known to be sparse, accurate estimation requires a low mutual coherence. However, in system identification, a major challenge arises from the construction of the regressor based on time series data, which often leads to a high mutual coherence. This conflict hinders accurate sparse estimation. To address this issue, the first part of this thesis introduces novel methods that reduce mutual coherence through linear coordinate transformations. These methods can be integrated with any sparse estimation techniques. Our numerical studies demonstrate significant improvements in performance compared to state-of-the-art sparse estimation algorithms. In the second part of the thesis, we shift our focus to optimal input design in system identification, which aims to achieve maximum accuracy in a model based on specific criteria. The original optimal input design techniques lack coherence constraints between the input sequences, often resulting in high mutual coherence and, consequently, increased sparse estimation errors for sparse models. Therefore, the second part of the thesis concentrates on designing optimal input for sparse models. We formulate the proposed methods and propose numerical algorithms using alternating minimization. Additionally, we compare the performance of our proposed methods with state-of-the-art input design algorithms, and we provide theoretical analysis of the proposed methods in both parts of the thesis. / Gles parameterestimering är viktigt inom systemidentifiering eftersom vissa modeller har naturligt förekommande gleshet i dess parametrar, men även för att det kan låta en minska ordningen av icke-glesa modeller. Noggrannheten av en skattad gles modell beror direkt på prestandan av de glesa estimeringsmetoderna. Det ¨ar välkänt att noggrannheten av en gles estimeringsmetod beror på korrelationer mellan regressorerna av den skattade modellen. Ömsesidig koherens (eng: mutual coherence) representerar maximum av dessa korrelationer. Noggrann estimering kräver låg ömsesidig koherens i de fallen då det är känt att parametervektorn är gles. En stor utmaning inom systemidentifiering orsakas av att, när en regressor konstrueras av tidsserie-data, så leder detta ofta till hög ömsesidig koherens. Denna konflikt hindrar noggrann gles estimering. För att åtgärda detta problem så introducerar avhandlingens första del nya metoder som minskar den ömsesidiga koherensen genom linjära koordinattransformationer. Dessa metoder är möjliga att kombinera med godtyckliga glesa estimeringsmetoder. Våra numeriska studier visar märkvärdig förbättring av prestanda jämfört med de bästa tillgängliga algoritmerna för gles parameterestimering. I avhandlingens andra del så ändrar vi vårt fokus till design utav optimala insignaler i systemidentifiering, där målet är att uppnå maximal noggrannhet i en modell, baserat på specifika kriterier. De ursprungliga metoderna för design av insignaler saknar bivillkor för ömsesidig koherens mellan insignalssekvenserna, vilket ofta resulterar i hög ömsesidig koherens och därmed också högre estimeringsfel för glesa modeller. Det är därför avhandlingens andra del fokuserar på att designa optimala insignaler för glesa modeller. Vi formulerar de föreslagna metoderna och erbjuder numeriska algoritmer som använder sig utav alternerande minimering. Vi jämför dessutom prestandan av vår metod med de bästa tillgängliga metoderna för design av insignaler, och vi presenterar även teoretisk analys av de föreslagna metoderna i avhandlingens båda delar. / <p>QC 20230911</p>

Call number: LD2668 .T4 1984 C67 / Master of Science

by Tham Yiu-Man. / Thesis (M.Phil.)--Chinese University of Hong Kong, 1993. / Includes bibliographical references. / Abstract --- p.i / Acknowledgements --- p.iii / Table of Contents --- p.iv / Chapter Chapter I --- Introduction / Chapter 1.1 --- Introduction --- p.1-1 / Chapter 1.2 --- Survey on Chinese Character Recognition --- p.1-4 / Chapter 1.3 --- Methodology Adopts in Our System --- p.1-7 / Chapter 1.4 --- Contributions and Organization of the Thesis --- p.1-11 / Chapter Chapter II --- Pre-processing and Stroke Extraction / Chapter 2.1 --- Introduction --- p.2-1 / Chapter 2.2 --- Thinning --- p.2-1 / Chapter 2.2.1 --- Introduction to Thinning --- p.2-1 / Chapter 2.2.2 --- Proposed Thinning Algorithm Cater for Stroke Extraction --- p.2-6 / Chapter 2.2.3 --- Thinning Results --- p.2-9 / Chapter 2.3 --- Stroke Extraction --- p.2-13 / Chapter 2.3.1 --- Introduction to Stroke Extraction --- p.2-13 / Chapter 2.3.2 --- Proposed Stroke Extraction Method --- p.2-14 / Chapter 2.3.2.1 --- Fork point detection --- p.2-16 / Chapter 2.3.2.2 --- 8-connected fork point merging --- p.2-18 / Chapter 2.3.2.3 --- Sub-stroke extraction --- p.2-18 / Chapter 2.3.2.4 --- Fork point merging --- p.2-19 / Chapter 2.3.2.5 --- Sub-stroke connection --- p.2-24 / Chapter 2.3.3 --- Stroke Extraction Accuracy --- p.2-27 / Chapter 2.3.4 --- Corner Detection --- p.2-29 / Chapter 2.3.4.1 --- Introduction to Corner Detection --- p.2-29 / Chapter 2.3.4.2 --- Proposed Corner Detection Formulation --- p.2-30 / Chapter 2.4 --- Concluding Remarks --- p.2-33 / Chapter Chapter III --- Four Corner Code / Chapter 3.1 --- Introduction --- p.3-1 / Chapter 3.2 --- Deletion of Hook Strokes --- p.3-3 / Chapter 3.3 --- Stroke Types Selection --- p.3-5 / Chapter 3.4 --- Probability Formulations of Stroke Types --- p.3-7 / Chapter 3.4.1 --- Simple Strokes --- p.3-7 / Chapter 3.4.2 --- Square --- p.3-8 / Chapter 3.4.3 --- Cross --- p.3-10 / Chapter 3.4.4 --- Upper Right Corner --- p.3-12 / Chapter 3.4.5 --- Lower Left Corner --- p.3-12 / Chapter 3.5 --- Corner Segments Extraction Procedure --- p.3-14 / Chapter 3.5.1 --- Corner Segment Probability --- p.3-21 / Chapter 3.5.2 --- Corner Segment Extraction --- p.3-23 / Chapter 3.6 4 --- C Codes Generation --- p.3-26 / Chapter 3.7 --- Parameters Determination --- p.3-29 / Chapter 3.8 --- Sensitivity Test --- p.3-31 / Chapter 3.9 --- Classification Rate --- p.3-32 / Chapter 3.10 --- Feedback by Corner Segments --- p.3-34 / Chapter 3.11 --- Classification Rate with Feedback by Corner Segment --- p.3-37 / Chapter 3.12 --- Reasons for Mis-classification --- p.3-38 / Chapter 3.13 --- Suggested Solution to the Mis-interpretation of Stroke Type --- p.3-41 / Chapter 3.14 --- Reduce Size of Candidate Set by No.of Input Segments --- p.3-43 / Chapter 3.15 --- Extension to Higher Order Code --- p.3-45 / Chapter 3.16 --- Concluding Remarks --- p.3-46 / Chapter Chapter IV --- Relaxation / Chapter 4.1 --- Introduction --- p.4-1 / Chapter 4.1.1 --- Introduction to Relaxation --- p.4-1 / Chapter 4.1.2 --- Formulation of Relaxation --- p.4-2 / Chapter 4.1.3 --- Survey on Chinese Character Recognition by using Relaxation --- p.4-5 / Chapter 4.2 --- Relaxation Formulations --- p.4-9 / Chapter 4.2.1 --- Definition of Neighbour Segments --- p.4-9 / Chapter 4.2.2 --- Formulation of Initial Probability Assignment --- p.4-12 / Chapter 4.2.3 --- Formulation of Compatibility Function --- p.4-14 / Chapter 4.2.4 --- Formulation of Support from Neighbours --- p.4-16 / Chapter 4.2.5 --- Stopping Criteria --- p.4-17 / Chapter 4.2.6 --- Distance Measures --- p.4-17 / Chapter 4.2.7 --- Parameters Determination --- p.4-21 / Chapter 4.3 --- Recognition Rate --- p.4-23 / Chapter 4.4 --- Reasons for Mis-recognition in Relaxation --- p.4-27 / Chapter 4.5 --- Introduction of No-label Class --- p.4-31 / Chapter 4.5.1 --- No-label Initial Probability --- p.4-31 / Chapter 4.5.2 --- No-label Compatibility Function --- p.4-32 / Chapter 4.5.3 --- Improvement by No-label Class --- p.4-33 / Chapter 4.6 --- Rate of Convergence --- p.4-35 / Chapter 4.6.1 --- Updating Formulae in Exponential Form --- p.4-38 / Chapter 4.7 --- Comparison with Yamamoto et al's Relaxation Method --- p.4-40 / Chapter 4.7.1 --- Formulations in Yamamoto et al's Relaxation Method --- p.4-40 / Chapter 4.7.2 --- Modifications in [YAMAM82] --- p.4-42 / Chapter 4.7.3 --- Performance Comparison with [YAMAM82] --- p.4-43 / Chapter 4.8 --- System Overall Recognition Rate --- p.4-45 / Chapter 4.9 --- Concluding Remarks --- p.4-48 / Chapter Chapter V --- Concluding Remarks / Chapter 5.1 --- Recapitulation and Conclusions --- p.5-1 / Chapter 5.2 --- Limitations in the System --- p.5-4 / Chapter 5.3 --- Suggestions for Further Developments --- p.5-6 / References --- p.R-1 / Appendix User's Guide / Chapter A .l --- System Functions --- p.A-1 / Chapter A.2 --- Platform and Compiler --- p.A-1 / Chapter A.3 --- File List --- p.A-2 / Chapter A.4 --- Directory --- p.A-3 / Chapter A.5 --- Description of Sub-routines --- p.A-3 / Chapter A.6 --- Data Structures and Header Files --- p.A-12 / Chapter A.7 --- Character File charfile Structure --- p.A-15 / Chapter A.8 --- Suggested Program to Implement the System --- p.A-17

by Lau Chi Ching, Donny. / Thesis (M.Sc.)--Chinese University of Hong Kong, 1993. / Includes bibliographical references (leaves [71]). / Chapter 1. --- Introduction / Chapter 1.1 --- Introduction --- p.1 / Chapter 1.2 --- Comparison of Phonetic and Written Character Input --- p.2 / Chapter 1.3 --- Significance of Phonetic Input --- p.4 / Chapter 1.4 --- Drawbacks of Current Phonetic Input Methods --- p.4 / Chapter 2. --- Objectives of the Research / Chapter 2.1 --- Main Objectives --- p.6 / Chapter 2.2 --- User Background Pre-requisite --- p.8 / Chapter 2.3 --- Roman-Spelling (Recommended Phonetic Scheme) --- p.9 / Chapter 2.4 --- User Input and the Output Scenario --- p.10 / Chapter 2.5 --- Outline of Free-Style Phonetic Input Processing --- p.15 / Chapter 3. --- Lexical Analyser / Chapter 3.1 --- Overview of Lexical Analyser --- p.17 / Chapter 3.2 --- Identification of Character Boundary --- p.19 / Chapter 3.3 --- Lexical Tree --- p.20 / Chapter 4. --- Selection Module / Chapter 4.1 --- Overview of Selection Module --- p.23 / Chapter 4.2 --- Fault-tolerance Capability --- p.24 / Chapter 4.3 --- Group Table (Groups of Similar Sounds) --- p.26 / Chapter 4.4 --- Distance Calculation Algorithm --- p.30 / Chapter 4.4.1 --- Character Dictionary --- p.31 / Chapter 4.4.2 --- Phrase Dictionary --- p.33 / Chapter 4.4.3 --- Hashing Key of the Dictionaries --- p.35 / Chapter 4.4.4 --- Maintenance of Dictionaries --- p.36 / Chapter 4.4.5 --- Distance Calculation of Character Input --- p.37 / Chapter 4.4.5.1 --- Examples of Character Output --- p.39 / Chapter 4.4.6 --- Distance Calculation of Phrase Input --- p.40 / Chapter 4.4.6.1 --- Examples of Phrase Output --- p.44 / Chapter 4.4.7 --- Explanation of Algorithm --- p.45 / Chapter 5. --- Syntax Analyser / Chapter 5.1 --- Overview of Syntax Analyser --- p.46 / Chapter 5.2 --- Overview of a Chinese Simple Sentence --- p.47 / Chapter 5.3 --- Testing Simple Sentence Rules --- p.48 / Chapter 5.3.1 --- NDFA for Chinese Grammar Rules --- p.49 / Chapter 5.4 --- Syntax Analysis Algorithm --- p.51 / Chapter 5.4.1 --- Explanation of Algorithm --- p.52 / Chapter 5.4.2 --- Justification of Algorithm --- p.54 / Chapter 5.4.3 --- Examples of Syntax Analysis --- p.55 / Chapter 5.5 --- Parse Tree for Semantic Analysis --- p.59 / Chapter 6. --- Division of Technical Work --- p.61 / Chapter 7. --- Applied Areas of the Research / Chapter 7.1 --- Chinese User Interface with Operating System --- p.63 / Chapter 7.2 --- Bilingual Programming Language Editor --- p.64 / Chapter 7.3 --- Development of a Chinese Programming Language --- p.66 / Chapter 7.4 --- Putonghua Training --- p.67 / Chapter 8. --- Conclusions and Future Improvements / Chapter 8.1 --- Conclusions --- p.68 / Chapter 8.2 --- Future Improvements --- p.69 / References / Appendix A

by Chan Shing Chi, Michael. / Thesis (M.Ph.)--Chinese University of Hong Kong, 1987. / Bibliography: leaves [112]-[114]

by Yeung Chuen-sang. / Thesis (M.Ph.)--Chinese University of Hong Kong, 1987. / Bibliography: leaves 190-194.

by Wai-Yip Tung. / Thesis (M.Phil.)--Chinese University of Hong Kong, 1993. / Includes bibliographical references (leaves 55-56). / Chapter 1. --- Introduction --- p.1 / Chapter 1.1. --- Previous work --- p.2 / Chapter 1.1.1. --- A Chinese METAFONT --- p.2 / Chapter 1.1.2. --- Chinese character generator --- p.2 / Chapter 1.1.3. --- Chinese Character Design System CCDS --- p.2 / Chapter 1.2. --- Goals of the thesis --- p.3 / Chapter 1.3. --- Overview of the thesis --- p.3 / Chapter 2. --- Construction of Chinese Characters --- p.5 / Chapter 2.1 --- Introduction --- p.5 / Chapter 2.2. --- liu shu(六書）Six Principles of Chinese Character Construction --- p.5 / Chapter 2.3. --- Structural Analysis of Chinese Characters --- p.7 / Chapter 2.3.1. --- Left-Right Structure --- p.8 / Chapter 2.3.2. --- Top-Bottom Structure --- p.9 / Chapter 2.3.3. --- Inside-Outside Structure --- p.10 / Chapter 2.3.4. --- Singleton Structure --- p.10 / Chapter 2.4. --- Usage frequency of radicals --- p.11 / Chapter 2.5. --- Usage frequency of Bushou --- p.11 / Chapter 2.6. --- Usage frequency of Shengpang --- p.13 / Chapter 2.7. --- Summary --- p.15 / Chapter 3. --- Composition by Radicals --- p.17 / Chapter 3.1. --- Introduction --- p.17 / Chapter 3.2. --- Transforming radicals --- p.18 / Chapter 3.3. --- Quality of transformed radicals --- p.19 / Chapter 3.4. --- Lower level components --- p.20 / Chapter 3.5. --- Summary --- p.23 / Chapter 4. --- Automatic Hinting for Chinese Font --- p.24 / Chapter 4.1 --- Introduction --- p.24 / Chapter 4.2. --- Automatic hinting for Chinese font --- p.26 / Chapter 4.3. --- Stroke recognition --- p.30 / Chapter 4.3.1. --- Identify horizontal lines --- p.31 / Chapter 4.3.2. --- Identify stroke segments --- p.31 / Chapter 4.3.3. --- Stroke recognition --- p.32 / Chapter 4.4. --- Regularize stroke width --- p.33 / Chapter 4.5. --- Grid-fitting horizontal and vertical strokes --- p.33 / Chapter 4.6. --- Grid-fitting radicals --- p.37 / Chapter 4.7. --- Summary --- p.39 / Chapter 5. --- RADIT - A Chinese Font Editor --- p.41 / Chapter 5.1. --- Introduction --- p.41 / Chapter 5.2. --- RADIT basics --- p.41 / Chapter 5.2.1. --- Character selection window --- p.42 / Chapter 5.2.2. --- Character window --- p.42 / Chapter 5.2.3. --- Tools Palette --- p.43 / Chapter 5.2.4. --- Toolbar --- p.43 / Chapter 5.2.5. --- Zooming the character window --- p.44 / Chapter 5.3. --- Editing a character --- p.44 / Chapter 5.3.1. --- Selecting handles --- p.44 / Chapter 5.3.2. --- Adding lines and curves --- p.45 / Chapter 5.3.3. --- Delete control points --- p.45 / Chapter 5.3.4. --- Moving control points --- p.45 / Chapter 5.3.5. --- Cut and paste --- p.46 / Chapter 5.3.6. --- Undo --- p.46 / Chapter 5.4. --- Adding radicals to a character --- p.46 / Chapter 5.5. --- Rasterizing and grid-fitting a character --- p.47 / Chapter 5.5.1. --- Rasterizing a character --- p.48 / Chapter 5.5.2. --- Stroke detection and regularization --- p.48 / Chapter 5.5.3. --- Grid-fitting and rasterizing a character --- p.49 / Chapter 6. --- Conclusions --- p.50 / Appendix A: Sample Fonts --- p.52 / References --- p.55