Integrated optics as a platform for signal processing offers significant benefits such as large bandwidth, low loss, and a potentially high degree of reconfigurability. Silicon (Si) has unique advantages as a material platform for integration, as well as properties such as a strong thermo-optic mechanism that allows for the realization of highly reconfigurable photonic systems. Chapter 1 is devoted to the discussion of these advantages, and Chapter 2 provides the theoretical background for the analysis of integrated Si-photonic devices. The thermo-optic property of Si, while proving extremely useful in facilitating reconfiguration, can turn into a nuisance when there is a need for thermally stable devices on the photonic chip. Chapter 3 presents a technique for resolving this issue without relying on a dynamic temperature stabilization process. Temperature-insensitive (or “athermal”) Si microdisk resonators with low optical loss are realized by using a polymer overlayer whose thermo-optic property is opposite to that of Si, and TiO2 is introduced as an alternative to polymer to deal with potential CMOS-compatibility issues. Chapter 4 demonstrates an ultra-compact, low-loss, fully reconfigurable, and high-finesse integrated photonic filter implemented on a Si chip, which can be used for RF-photonic as well as purely optical signal processing purposes. A novel, thermally reconfigurable reflection suppressor is presented in Chapter 5 for on-chip feedback elimination which can be critical for mitigating spurious interferences and protecting lasers from disturbance. Chapter 6 demonstrates a novel device for on-chip control of optical fiber polarization. Chapter 7 deals with select issues in the implementation of Si integrated photonic circuits. Chapter 8 concludes the dissertation.

Thesis (M.S.)--University of Texas at El Paso, 2008. / Title from title screen. Vita. CD-ROM. Includes bibliographical references. Also available online.

Thesis (Ph.D.)--University of Melbourne, Dept. of Electrical and Electronic Engineering, 2010. / Typescript. Includes bibliographical references (p. 177-187)

Thesis (M.S.)--Ohio University, March, 2006. / Title from PDF t.p. Includes bibliographical references (p. 80-82)

Thesis (M.Eng.)--Memorial University of Newfoundland, 2003. / Bibliography: leaves 99-106. Also available online.

Thesis (M. Phil.)--University of Hong Kong, 2005. / Title proper from title frame. Also available in printed format.

D.Phil. (Electrical and Electronic Engineering) / The complexity of multi-objective functions and diverse variables involved with optimization of parallel manipulator or parallel kinematic machine design has inspired the research conducted in this thesis to investigate techniques that are suitable to tackle this problem efficiently. Further the parallel manipulator dimensional synthesis problem is multimodal and has no explicit analytical expressions. This process requires optimization techniques which offer high level of accuracy and robustness. The goal of this work is to present method(s) based on Artificial Intelligence (AI) that may be applied in addressing the challenge stated above. The performance criteria considered include; stiffness, dexterity and workspace. The case studied in this work is a 6 degrees of freedom (DOF) parallel manipulator, particularly because it is considered much more complicated than the lesser DOF mechanisms, owing to the number of independent parameters or inputs needed to specify its configuration (i.e. the higher the DOFs, the more the number of independent variables to be considered). The first contribution in this thesis is a comparative study of several hybrid Multi- Objective Optimization (MOO) AI algorithms, in application of a parallel manipulator dimensional synthesis. Artificial neural networks are utilized to approximate a multiple function for the analytical solution of the 6 DOF parallel manipulator’s performance indices, followed by implementation of Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) as search algorithms. Further two hybrid techniques are proposed which implement Simulated Annealing and Random Forest in searching for optimum solutions in the Multi-objective Optimization problem. The final contribution in this thesis is ensemble machine learning algorithms for approximation of a multiple objective function for the 6 DOF parallel manipulator analytical solution. The results from the experiments demonstrated not only neural network (NN) but also other machine learning algorithms namely K- Nearest Neighbour (k-NN), M5 Prime (M5’), Zero R (ZR) and Decision Stump (DS) can effectively be implemented for the application of function approximation.

The major motivation for self-adaptive computing systems is the self-adjustment of the software according to a changing environment. Adaptive computing systems can add, remove, and replace their own components in response to changes in the system itself and in the operating environment of a software system. Although these systems may provide a certain degree of confidence against new environments, their structural and behavioral changes should be validated after adaptation occurs at runtime. Testing dynamically adaptive systems is extremely challenging because both the structure and behavior of the system may change during its execution. After self adaptation occurs in autonomic software, new components may be integrated to the software system. When new components are incorporated, testing them becomes vital phase for ensuring that they will interact and behave as expected. When self adaptation is about removing existing components, a predefined test set may no longer be applicable due to changes in the program structure. Investigating techniques for dynamically updating regression tests after adaptation is therefore necessary to ensure such approaches can be applied in practice. We propose a model-driven approach that is based on change propagation for synchronizing a runtime test model for a software system with the model of its component structure after dynamic adaptation. A workflow and meta-model to support the approach was provided, referred to as Test Information Propagation (TIP). To demonstrate TIP, a prototype was developed that simulates a reductive and additive change to an autonomic, service-oriented healthcare application. To demonstrate the generalization of our TIP approach to be instantiated into the domain of up-to-date runtime testing for self-adaptive software systems, the TIP approach was applied to the self-adaptive JPacman 3.0 system. To measure the accuracy of the TIP engine, we consider and compare the work of a developer who manually identifyied changes that should be performed to update the test model after self-adaptation occurs in self-adaptive systems in our study. The experiments show how TIP is highly accurate for reductive change propagation across self-adaptive systems. Promising results have been achieved in simulating the additive changes as well.

No description available.

Approved for public release, distribution is unlimited / A hardware interface is designed, developed, constructed, and tested to interface a naval radar to the SRC 6E reconfigurable computer. The U.S. Navy AN/SPS 65 radar provides in-phase (I) and quadrature (Q) channels along with the AGC voltage to the hardware interface in analog form. The hardware interface receives a sampling clock from the SRC 6E and in turn performs the requisite attenuation and digital conversion before presenting the signals to the SRC 6E through its CHAIN port. The results show that the SRC 6E can effectively generate a sampling clock to drive the analog-to-digital converters and that real- time radar data can be brought into the SRC 6E via its high speed CHAIN port for performing high speed digital signal processing. / Lieutenant, United States Naval Reserve