<p>The multiple hypotheses tracker (MHT) is an optimal tracking method due to the enumeration of all possible measurement-to-track associations. However, its practical implementation is limited by the NP-hard nature of this enumeration. To bound the computational complexity, some means of limiting the number of possible associations is required. Typical solutions include the interposition of rules to guide the pruning and merging of tracks. Other proposals have shown that the performance of a tracker, MHT or not, can be improved using feature information (e.g., signal strength, size, type) in addition to kinematic data. The inclusion of feature information allows for the discrimination to further gate the data associations. However, in most tracking systems, the schemes to manage the data association problem are extraneous to the Bayesian framework of the MHT. Further, the extraction of features from the raw sensor data is typically independent of the subsequent association and filtering stages. The features are then used in either an ad hoc way or are they are fused with the MHT tracker; they are not used intrinsically within MHT framework. In this thesis, a new approach whereby there is an intrinsic interaction between feature extraction and the MHT is presented. The measure of the quality of feature extraction is input into measurement-to-track association while the prediction step feeds back information to be used in the next round of feature extraction to increase the information available a priori. The motivation for this forward and backward interaction between feature extraction and tracking is to improve the performance in both steps. This approach allows for a more rational partitioning of the feature space, removing unlikely features from. the assignment problem. In addition, a track-specific detection probability becomes available to the prior. This probability significantly improves the coasting behavior when measurements are not available for track continuation. Simulation results demonstrate the benefits of the proposed approach.</p> / Master of Applied Science (MASc)

<p>An improved Perfectly Matched Layer (PML) boundary condition is introduced that addresses the previously reported instabilities. A PML for acoustic waves is derived by closely following Bérenger's derivation of a PML for electromagnetic waves. A new matching condition is developed to relate the velocity and stress loss-coefficients similar to the matching condition defined for electromagnetic waves.<br /> Whereas, the spatial and temporal derivatives are related by simple scalars in Maxwell's equations, elastodynamic equations describing waves in piezoelectric materials require the use of tensor quantities whose general forms are material dependent. <br /> In practice, SAW generation is often also accompanied by some small creation of SAWs, which act as parasitic waves. It is thus desired to remove the reflection of these waves from the bottom, in both physical SAW devices and in modeling.</p> / Master of Applied Science (MASc)

<p>This thesis is concerned with interpolation along the Z-axis for application in medical imaging and increasing out-of plane resolution of 3-D medical image sets. Interpolation along the Z-axis is an essential task in clinical studies for better diagnosis and analysis of body organs and their functions. It is also necessary when sets of images with different out-of plane resolutions should be analyzed together.</p> <p>The first part of the thesis discusses a 3-D interpolation method based on a piece-wise autoregressive model that has been already proven to be efficient for 2-D image interpolation. The 3-D image set is modeled as a 3-D piece-wise autoregressive model and the model parameters are estimated within a cube that slides through the low resolution image set.</p> <p>The major part of this thesis is devoted to a new interpolation algorithm, called contextbased 3-D interpolation. The proposed method represents a new approach of aiding 3-D interpolation and improving its performance by efficient use of domain knowledge about the anatomy, orientation and imaging modalities. In the new approach a family of adaptive 3-D interpolation filters are designed and conditioned on different spatial contexts (classes of feature vectors). Training is used to incorporate the domain knowledge into the design of these interpolators. Experimental results show significant improvement of the new approach over some existing 3D interpolation techniques.</p> / Master of Applied Science (MASc)

<p>In many radar tracking systems ·with a certain pulse-repetition frequency (PRF), Doppler (or range rate) measurements are available in addition to position measurements. The extra range rate information Doppler measurements provide is able to improve the tracking performance. However, a fundamental problem associated with Pulse-Doppler radars is, especially at low PRFs, range rate ambiguity. This is because Doppler shifts in the frequency spectrum are aliased by a difference of an integer times of PRF. In this case, the observed Doppler measurement shifts from the true range rate by an unknown difference.</p> <p>In previous works, algorithms to eliminate the Doppler ambiguity based on the Chinese Remainder Theorem have been proposed for radars with multiple PRFs. Time- Frequency Analysis (TFA) using Fast Fourier Transform (FFT) can reduce the ambiguity in the frequency domain.</p> <p>In this paper, a new approach for multitarget detection and tracking with Doppler ambiguity is presented. Ambiguous Doppler measurements in addition to the position measurements are directly used in data association and tracking. To solve the Doppler ambiguity for single target tracking, three methods are proposed based on UKF, MHT and PDA, respectively. In addition, modifications to Multiple Hypothesis Tracking (MHT) and Joint Integrated Probabilistic Data Association (JIPDA) algorithms to resolve Doppler ambiguity in multitarget tracking are considered. Simulation results are preformed to demonstrate the new algorithms with tracking results analysis.</p> / Master of Applied Science (MASc)

<p>p.p1 {margin: 0.0px 0.0px 0.0px 0.0px; font: 11.5px Times; color: #565656} p.p2 {margin: 0.0px 0.0px 0.0px 0.0px; font: 11.5px Times; color: #6a6a6a} span.s1 {color: #7b7b7b} span.s2 {color: #949494} span.s3 {color: #414141} span.s4 {color: #6a6a6a} span.s5 {color: #565656}</p> <p>Traditionally, robots are used to handle rigid objects, such as in automotive and manufacturing industries. Emerging robotic fields like food industry and medical robotics has inspired research into robots capable of accurately manipulating deformable objects.</p> <p>This thesis fo cuses on a task known as the indirect simultaneous positioning problem. The task requires accurate deformation control of a nonrigid obj ect. The desired configuration is prescribed by control points defined within the object body. This deformation is achieved by applying forces to to manipulation points located elsewhere on the body.</p> <p>This thesis approaches the problem using a linear robust output regulation framework. This framework requires the following steps:</p> <p>p.p1 {margin: 0.0px 0.0px 0.0px 0.0px; font: 11.5px Times; color: #565656} p.p2 {margin: 0.0px 0.0px 0.0px 0.0px; font: 11.5px Times; color: #7b7b7b} p.p3 {margin: 0.0px 0.0px 0.0px 0.0px; font: 11.5px Times; color: #6a6a6a} span.s1 {font: 10.0px Helvetica} span.s2 {color: #6a6a6a} span.s3 {color: #414141} span.s4 {color: #7b7b7b} span.s5 {color: #565656} span.s6 {color: #ababab} span.s7 {color: #949494}</p> <p>1. construct a second-order dynamical model of the object</p> <p>2. linearize about the equilibrium point</p> <p>3. assess the controllability, observability, and solution to the regulator equations</p> <p>4. compute the controller gains that solve the regulation problem</p> <p>p.p1 {margin: 0.0px 0.0px 0.0px 0.0px; font: 11.5px Times; color: #565656} span.s1 {color: #6a6a6a} span.s2 {color: #414141} span.s3 {color: #7b7b7b}</p> <p>This control law approach is tested in both simulations using a nonlinear RKPM model of a deformable object, and in experiments using a robot manipulator to apply the desired forces to the manipulation points of a physical, elastic, deformable object. This thesis proves the validity of t his approach using homogeneous and nonhomogeneous planar objects, as well as multiple control point and manipulation point configurations.</p> / Master of Applied Science (MASc)

<p>An effective mode matching method (MMM) is developed for dealing with three- dimensional (3D) dielectiic waveguide structures with arbitrary transverse index profiles and scattering interfaces along the longitudinal waveguide axis. With the introduction of a perfectly conducting box coated by a perfectly matched layer (PML) as the lateral boundary, the method circumvents the inherent difficulties associated with the continuous radiation modes of open waveguide structures. The semi and full-vectorial finite difference (FD) methods are employed for mode calculation. The guideline for choice of the PML parameters in the mode matching analysis is discussed via a two-dimensional (2D) waveguide structure with a single step discontinuity. We show results for the 3D waveguide air gap, facet and polarization converter. The effectiveness and efficiency of the method are validated.</p> / Master of Applied Science (MASc)

<p>The energy crises, along with the recent global warming trends, demand an immediate cut in the use of fossil fuel. Therefore, renewable sources of energy and especially solar power, have gained tremendous attention from the consumer countries as a possible candidate to replace the carbon-based energy supplies. Among all different types of solar cells, flexibility, cost-effective fabrication processes, combined with low-price materials and most importantly a great potential for improvement make organic solar cells an interesting topic for research. On the other hand, modeling provides a valuable opportunity to study device properties that experimentally are out of reach, expensive or need a long time to measure. These mentioned reasons motivated us to choose modeling of organic solar cells as the subject of our research.<br /><br />In this research, we tried to provide a complete study of the power generating procedure in organic solar cells by modeling all of the following processes: in-coupling of the photons, absorption of the photons, formation of the excitons, diffusion of the excitons, dissociation of the excitons, transportation of the charges and collection of the charges at the electrodes.<br /><br />To get a better understanding and also because of basic physical differences, the modeling is<br />divided into two parts: the optical section and the electrical section. Each section is also divided<br />into two separate segments, analytical and numerical analysis.<br /><br />Using the optical models with different designs to improve the performance of the solar cells, the effect of the layer thickness and two- and three-dimensional light focusing apertures on the intensity of light at the junction of n-type and p-type materials (for bilayer heterojunction organic solar cells) are studied. Results show that for a certain design of the light focusing aperture, a 98% increase in the light absorption in a bilayer heterojunction solar cell can be obtained.<br /><br />The electrical performance of organic solar cell is also studied by using analytical modeling of exciton diffusion for bilayer heterojunction solar cells and numerical models based on driftdiffusion procedures by using COMSOL multiphysics software for bulk heterojunction solar cells. Based on the mismatch between the calculated results and measurement (counterdiode effect), a tunneling current correction is introduced. Finally, using the tunneling current model, the energy diagram of the organic active layer at the metallic contact is characterised.<br /><br />In summary, five different models are described in five separate sections, and at the end of each section, results are reported and compared with the literature that prove that the presented models can be used for a new design of organic solar cell characteristics to improve the performance of the device. Also, by introducing the tunneling current to model the counterdiode effect, we have contributed to the literature.</p> / Master of Applied Science (MASc)

<p>Multiplying by known constants is a common operation in many digital signal processing (DSP) algorithms. High performance DSP systems are implemented in custom hardware, in which the designer has the ability to choose which logic elements will be used to perform the computation. By exploiting the properties of binary multiplication, it is possible to realize constant multiplication with fewer logic resources than required by a generic multiplier. In this thesis, we present several new algorithms for solving the constant multiplication problem - <em>given</em> a set of constants, <em>find</em> a low-cost logic circuit that realizes multiplication by each of the constants.</p> <p>In this thesis, a thorough analysis of the existing algorithms, the underlying frameworks, and the associated properties is provided. We also propose new strategies which are fundamentally different from the existing methods, such as the integration of a heuristic algorithm within an optimal algorithm. In our proposed optimal exhaustive algorithms, we introduce aggressive pruning methods to improve the compute efficiency (compared to existing optimal exhaustive algorithms). Our proposed heuristics attempt to address the weaknesses of the existing heuristics. By extending the analysis of prior work and providing new insight, we are often able to improve both the run time and the performance (in terms of minimizing logic resources).</p> / Master of Applied Science (MASc)

<p>This thesis concerns with recovery of compressive sampled images. Since many natural signals such as images are non-stationary, the sparse space varies in time/spatial domain. Therefore, compressive sensing (CS) recovery should be carried on locally adaptive, signal-dependent spaces to answer the fact that the CS measurements are not dependant to the signal structures. Existing CS reconstruction algorithms use fixed basis such as wavelets and DCT for the signals. To address this problem, we proposed new technique for model guided adaptive recovery of compressive sensing. The proposed algorithms are based on two dimensional piecewise autoregressive model and can adaptively recover compressive sampled images. In addition, proposed algorithms offer a powerful mechanism to characterize structured sparsity of natural images. This mechanism greatly restricts the CS solution space. Simulation results show the preeminent effect of our algorithms in the recovery of wide range of natural images. In average our best algorithm improves the reconstruction quality of existing CS methods by 2dB.</p> / Master of Applied Science (MASc)

<p>Probability Hypothesis Density (PHD) filter is a unified framework for multitarget tracking that provides estimates for a number of targets as well as individual target states. Sequential Monte Carlo (SMC) implementation of a PHD filter can be used for nonlinear non-Gaussian problems. However, the application of PHD based state estimators for a distributed sensor network, where each tracking node runs its own PHD based state estimator, is more challenging compared with single sensor tracking due to communication limitations. A distributed state estimator should use the available communication resources efficiently in order to avoid the degradation of filter performance. In this thesis, a method that communicates encoded measurements between nodes efficiently while maintaining the filter accuracy is proposed. This coding is complicated in the presence of high clutter and instantaneous target births. This problem is mitigated using adaptive quantization and encoding techniques. The performance of the algorithm is quantified using a Posterior Cramér-Rao Lower Bound (PCRLB), which incorporates quantization errors. Simulation studies are performed to demonstrate the effectiveness of the proposed algorithm.</p> / Master of Applied Science (MASc)