Polynomial phase modulated (PPM) signals have been shown to provide improved error rate performance with respect to conventional modulation formats under additive white Gaussian noise and fading channels in single-input single-output (SISO) communication systems. In this dissertation, systems with two and four transmit antennas using PPM signals were presented. In both cases we employed full-rate space-time block codes in order to take advantage of the multipath channel. For two transmit antennas, we used the orthogonal space-time block code (OSTBC) proposed by Alamouti and performed symbol-wise decoding by estimating the phase coefficients of the PPM signal using three different methods: maximum-likelihood (ML), sub-optimal ML (S-ML) and the high-order ambiguity function (HAF). In the case of four transmit antennas, we used the full-rate quasi-OSTBC (QOSTBC) proposed by Jafarkhani. However, in order to ensure the best error rate performance, PPM signals were selected such as to maximize the QOSTBC’s minimum coding gain distance (CGD). Since this method does not always provide a unique solution, an additional criterion known as maximum channel interference coefficient (CIC) was proposed. Through Monte Carlo simulations it was shown that by using QOSTBCs along with the properly selected PPM constellations based on the CGD and CIC criteria, full diversity in flat fading channels and thus, low BER at high signal-to-noise ratios (SNR) can be ensured. Lastly, the performance of symbol-wise decoding for QOSTBCs was evaluated. In this case a quasi zero-forcing method was used to decouple the received signal and it was shown that although this technique reduces the decoding complexity of the system, there is a penalty to be paid in terms of error rate performance at high SNRs.

Modern smartphones contain multiple sensors and long lasting batteries, making them ideal platforms for mobility monitoring. Mobility monitoring can provide rehabilitation professionals with an objective portrait of a patient’s daily mobility habits outside of a clinical setting. The objective of this thesis was to improve the performance of the human activity recognition within a custom Wearable Mobility Measurement System (WMMS). Performance of a current WMMS was evaluated on able-bodied and stroke participants to identify areas in need of improvement and differences between populations. Signal features for the waist-worn smartphone WMMS were selected using classifier-independent methods to identify features that were useful across populations. The newly selected features and a transition state recognition method were then implemented before evaluating the improved WMMS system’s activity recognition performance. This thesis demonstrated: 1) diverse population data is important for WMMS system design; 2) certain signal features are useful for human activity recognition across diverse populations; 3) the use of carefully selected features and transition state identification can provide accurate human activity recognition results without computationally complex methods.

D.Ing. (Electrical And Electronic Engineering) / Please refer to full text to view abstract

The challenge in finding genes in eukaryotic organisms using computational methods is an ongoing problem in the biology. Based on various genomic signals found in eukaryotic genomes, this problem can be divided into many different sub­-problems such as identification of transcription start sites, translation initiation sites, splice sites, poly (A) signals, etc. Each sub-­problem deals with a particular type of genomic signals and various computational methods are used to solve each sub-­problem. Aggregating information from all these individual sub-­problems can lead to a complete annotation of a gene and its component signals. The fundamental principle of most of these computational methods is the mapping principle – building an input-­output model for the prediction of a particular genomic signal based on a set of known input signals and their corresponding output signal. The type of input signals used to build the model is an essential element in most of these computational methods. The common factor of most of these methods is that they are mainly based on the statistical analysis of the basic nucleotide sequence string composition. 4 Our study is based on a novel approach to predict genomic signals in which uniquely generated structural profiles that combine compressed physicochemical properties with topological and compositional properties of DNA sequences are used to develop machine learning predictive models. The compression of the physicochemical properties is made using principal component analysis transformation. Our ideas are evaluated through prediction models of canonical splice sites using support vector machine models. We demonstrate across several species that the proposed methodology has resulted in the most accurate splice site predictors that are publicly available or described. We believe that the approach in this study is quite general and has various applications in other biological modeling problems.

This dissertation examines whether simultaneously conditioning on the multidimensional characteristics of information signals can help predict returns that are of economic significance. We use large price changes, public announcements, and large volume increases to proxy for the magnitude, dissemination, and precision of information signals. Abnormal returns following large price change events are found to be unimportant. As we condition on other characteristics of information signals, the abnormal returns become large. Large price change events accompanied by both a public announcement and an increase in volume have a 20-day abnormal return of almost 2% for positive events and -1.68% for negative events. The type of news provides further refinement. If the news relates to earnings announcements, management earnings forecasts, or analyst recommendations then the 20-day abnormal returns becomes much larger: ranging from 3% to 4% for positive events and about -2.25% for negative events. For these news events, we also find that the underreaction is greater for positive (negative) event firms that underperformed (overperformed) the market in the prior period, earning 20-day post-event abnormal returns of 4.85% (-3.50%). This evidence is consistent with the Barberis, Shleifer, and Vishny (1998) model of investor sentiment that suggests that investors are slow to change their beliefs. The evidence from our sample does not provide much support for strategic trading models under information asymmetry. Finally, an out-of-sample trading strategy generates 20-day post-event statistically significant abnormal return of 2.18% for positive events and -2.40% for negative events. Net of transaction costs, the abnormal returns are a statistically significant 1.04% for positive events and a statistically significant -1.51% for negative events. / Ph. D.

Several possible approaches to creating a model for predicting satellite signal levels inside buildings are examined. These models make use of resonant cavity modes and vector ray addition. The cavity mode approach yields inconclusive results because of a problem with uniqueness, there are too many potential modes and no obvious way to decide between them. The ray model uses vector representation. It tracks changes, and combines rays at the receiver. Signal levels are normalized to free space values. An algorithm for the construction of such a model is developed and results are obtained. A three ray model, incorporating LOS, floor, and ceiling reflected rays gives reasonable agreement with experimental data. The types of information needed are the room height, the receiver height, if the receiver is in the vicinity of a window, and the elevation angle of the satellite. This model shows that a user has to move only a small distance to find an area where fading is brought to acceptable levels. / Master of Science

We consider the problem of narrow-band signal detection in a passive sonar environment. The collected signals are passed to a fast Fourier Transform (FFT) delay-sum beamformer. In classical signal detection, the output of the FFT spectrum analyser in each frequency bin is the signal power spectrum which is used as the signal feature for detection. The observed signal power is compared to a locally estimated mean noise power and a log likelihood ratio test (LLRT) can then be established. In this thesis, we propose the use of the power spectral density (PSD) matrix of the spectrum analyser output as the feature for detection due to the additional cross-correlation information contained in such matrices. However, PSD matrices are structurally constrained and therefore form a manifold in the signal space. Thus, to find the distance between two matrices, the measurement must be carried out using Riemannian distance (RD) along the tangent of the manifold, instead of using the common Euclidean distance (ED). In this thesis, we develop methods for measuring the Frechet mean of noise PSD matrices using the RD and weighted RD. Further, we develop an optimum weighting matrix for use in signal detection by RD so as to further enhance the detection performance. These concepts and properties are then used to develop a decision rule for the detection of narrow-band sonar signals using PSD matrices. The results yielded by the new detection method are very encouraging. / Thesis / Master of Applied Science (MASc)

<p> The study of the different sleep stages of a patient using his/her recorded EEG signals falls in the area of signal classification. In general, this involves extracting from the EEG signals, a signal feature on which the classification is performed. In this thesis, we apply the techniques of signal classification to the analysis of the sleep of a patient. The feature we use is the power spectral density (PSD) matrices of a multi-channel EEG signal. This not only allows us to examine the power spectrum contents of each signal which complies with what clinical experts use in their visual judgement of EEG signals, but also allows the correlation between the multi-channel signals to be studied. To establish a metric facilitating the classification, we analyze the structure as well as exploit the specific geometric properties of the space of PSD matrices. Specifically, we study this space from the viewpoint of Riemannian manifolds. We apply a Riemannian metric and, with the aid of fibre bundle theory, develop intrinsic (geodesic) distance measures for the PSD matrix manifold. To utilize such new distance measures effectively for EEG signal classification, we need to find a suitable weighting matrix for the PSD matrices so that the distances between similar features are minimized while those between dissimilar features are maximized. A closed form expression for this weighting matrix is obtained by solving an equivalent convex optimization problem. The effectiveness of using these novel weighted distance measures is verified by applying them to the sleep pattern classification of a collection of recorded EEG signals using the k-nearest neighbor decision algorithm with excellent results. </p> / Thesis / Doctor of Philosophy (PhD)

No description available.

No description available.