• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 10
  • 2
  • 1
  • Tagged with
  • 1325
  • 1313
  • 1312
  • 1312
  • 1312
  • 192
  • 164
  • 156
  • 129
  • 99
  • 93
  • 79
  • 52
  • 51
  • 51
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
481

Particle swarm optimization aided MIMO transceiver design

Yao, Wang January 2011 (has links)
In this treatise, we design Particle Swarm Optimization (PSO) aided MIMO transceivers. The employment of multiple antennas leads to the concept of multiple-input multiple-output (MIMO) systems, which constitute an effective way of achieving an increased capacity. When multiple antennas are employed at the Base Station (BS), it is possible to employ Multiuser Detection (MUD) in the uplink. However, in the downlink (DL), due to the size as well as power consumption constraints of mobile devices, so-called Multiuser Transmission (MUT) techniques may be employed at the BS for suppressing the multiuser interference before transmissions, provided that the DL channel to be encountered may be accurately predicted. The MUT scheme using the classic MMSE criterion is popular owing to its simplicity. However, since the BER is the ultimate system performance indicator, in this treatise we are more interested in the Minimum BER MUT (MBER-MUT) design. Unlike the MBER-MUD, the MBER-MUT design encounters a constrained nonlinear optimization problem due to the associated total transmit power constraint. Sequential Quadratic Programming (SQP) algorithms may be used to obtain the precoder’s coefficients. However, the computational complexity of the SQP based MBER-MUT solution may be excessive for high-rate systems. Hence, as an attractive design alternative, continuous-valued PSO was invoked to find the MBER-MUT’s precoder matrix in order to reduce its computational complexity. Two PSO aided MBER-MUTs were designed and explained. The first one may be referred to as a symbol-specific MBER-MUT, while the other one may be termed as the average MBER-MUT. Our simulation results showed that both of our designs achieve an improvement in comparison to conventional linear MUT schemes, while providing a reduced complexity compared to the state-of-art SQP based MBER-MUT. Later, we introduced discrete multi-valued PSO into the context of MMSE Vector Precoding (MMSEVP) to find the optimal perturbation vector. As a nonlinearMUT scheme, the VP provides an attractive BER performance. However, the computational complexity imposed during the search for optimal perturbation vector may be deemed excessive, hence it becomes necessary to find reduced-complexity algorithms while maintaining a reasonable BER performance. Lattice-Reduction-aied (LRA) VP is the most popular approach to reduce the complexity imposed. However, the LRA VP is only capable of achieving a suboptimum BER performance, although its complexity is reduced. Another drawback of LRA VP is that its complexity is fixed, which is beneficial for real-time implenebtations, but it is unable to strike a trade-off between the target BER and its required complexity. Therefore, we developed a discrete multi-valued PSO aided MMSE-VP design, which has a flexible complexity and it is capable of iteratively improving the achievable. In Chapter 5, our contributions in the field of Minimum Bit Error Rate Vector Precoding (MBER-VP) are unveiled. Zero-Forcing Vector Precoding (ZF-VP) and MMSE Vector Precoding (MMSE-VP) had already been proposed in the literature. However, to the best of our knowledge, no VP algorithm was proposed to date based on the direct minimisation of the BER. Our improved MMSE-VP design based on the MBER criterion first invokes a regularised channel inversion technique and then superimposes a discrete-valued perturbation vector for minimising the BER of the system. To further improve the system’s BER performance, an MBER-based generalised continuous-valued VP algorithm was also proposed. Assuming the knowledge of the information symbol vector and the CIR matrix, we consider the generation of the effective symbol vector to be transmitted by directly minimising the BER of the system. Our simulation results show the advantage of these two VP schemes based on the MBER criterion, especially for rank-deficient systems, where the number of BS transmit antennas is lower than the number of MSs supported. The robustness of these two designs to the CIR estimation error are also investigated. Finally, the computational complexity imposed is also quantified in this chapter. With the understanding of the BER criterion of VP schemes, we then considered a new transceiver design by combing uniform channel decomposition and MBER vector precoding, which leads to a joint transmitter and receiver design referred as the UCD-MBER-VP scheme. In our proposed UCD-MBER-VP scheme, the precoding and equalisation matrices are calculated by the UCD method, while the perturbation vector is directly chosen based on the MBER criterion. We demonstrated that the proposed algorithm outperforms the existing benchmark schemes, especially for rank-deficient systems, where the number of users supported is more than the number of transmit antennas employed. Moreover, our proposed joint design approach imposes a similar computational complexity as the existing benchmark schemes.
482

A sensor system to detect events in gait for the correction of abnormalities in neurological patients

Abdul Malik, Noreha January 2010 (has links)
Contraction of the hamstrings or gluteals muscles, using electrical stimulation, could improve the abnormal gait in neurological patients. The stimulation timing which follows the normal muscle activity is impractical to achieve using the traditional sensor (the footswitch). This study focuses on the development of a new sensor system for the detection of events in the gait cycle to trigger stimulation of hamstrings or gluteals muscles for preventing knee hyperextension into early stance or reducing the excessive hip flexion/adduction at heel strike. A sensor unit, consisting of four accelerometers,has been designed to determine the angles and linear accelerations of a segment without the need for integration. Tests have been carried out to verify the error of the sensor unit angle measurement by comparing it with the output from a potentiometer of a simple inverted pendulum. In five healthy subjects during walking, assessments have been carried out to compare the segment angle of the thigh, shank and foot calculated from the sensor unit, with the same angles measured using a motion capture system (ViconTM). The results show that the shank segment angles have a similar pattern. A sensor system to detect gait events has been developed. It consists of the sensor unit, a correlation coefficient calculation and a set of rules with thresholds. The system has detected reliably all heel strikes and a place in the gait cycle representing the tibia vertical position of five healthy subjects. Two sample windows selected from one set of the subject data have been used to detect all of the events. Using the same system, all tibial vertical events have been detected reliably compared to the footswitch in six neurological patients. In five patients, the same sample window, selected from the healthy subject, was used in the detection. For one patient, a sample window selected from the same patient data was used. Further work will be needed to implement the system in real time and evaluate it’s use with electrical stimulation as well as to establish the effect of stimulation in patients using the sensor unit as the trigger
483

Parallel sparse matrix solution for direct circuit simulation on a multiple FPGA system

Nechma, Tarek January 2012 (has links)
SPICE, from the University of California, at Berkeley, is the de facto world standard for circuit simulation. SPICE is used to model the behaviour of electronic circuits prior to manufacturing to decrease defects and hence reduce costs. However, accurate SPICE simulations of today's sub micron circuits can often take days or weeks on conventional processors. In a nutshell, a SPICE simulation is an iterative process that consists of two phases per iteration, namely, model evaluation followed by a matrix solution. The model evaluation phase has been found to be easily parallelisable unlike the subsequent phase, which involves the solution of highly sparse and asymmetric matrices. In this thesis, we present an FPGA implementation of a sparse matrix solver hardware,geared towards matrices that arise in SPICE circuit simulations. As such, we demonstrate how we extract parallelism at di�erent granularities to accelerate the solution process. Our approach combines static pivoting with symbolic analysis to compute an accurate task ow-graph which e�ciently exploits parallelism at multiple granularities and sustains high oating-point data rates. We also present a quantitative comparison between the performance of our hardware protrotype and state-of-the-art software package running on a general purpose PC equipped with a 2.67 GHz six-core 12 thread Intel Core Xeon X5650 microprocessor and 6 GB memory. We report average speedups of 9.65�, 11.83�, 17.21� against UMFPACK, KLU, and Kundert Sparse matrix packages respectively. We also detail our approach to adapt our sparse LU hardware prototype from a single-FPGA architecture to a multi-FPGA system to achieve higher acceleration ratios up to 38� for certain circuit matrices.
484

Noncoherent successive relaying for multi-user wireless systems

Li, Li January 2013 (has links)
A noncoherent detection based successive relaying aided network (SRAN) is proposed and investigated in the context of a multi-user, multi-relay assisted scenario. The potentially excessive complexity of multiple-antenna based power-hungry channel estimation is avoided by replacing the classic coherent detection by multiple-symbol-based noncoherent detection. Then, as the benefit of forming a virtual antenna array (VAA) in a distributed fashion, the proposed cooperative network becomes capable of achieving a substantial spatial diversity gain in the uplink. Furthermore, the 50% throughput loss incurred by the conventional single relay-aided two-phase cooperative network, which is caused by the half-duplex transmit/receive constraint of practical transceivers is recovered by designing a spectral efficient successive relaying protocol. Hence the proposed noncoherent successive relaying aided multi-user wireless system becomes capable of significantly improving the system’s performance. We demonstrate that multiple-symbol differential detection (MSDD) is capable of eliminating the error floor of the conventional differential detection (CDD), when experiencing severely timeselective Rayleigh fading associated with a high Doppler frequency, since the MSDD algorithm benefits from a higher time-diversity than CDD. However, this is achieved at a potentially excessive complexity, which is unaffordable in many practical applications. As a remedy, the sphere decoding principle is incorporated into the MSDD algorithm. The resultant multiple-symbol differential sphere detection (MSDSD) strikes an attractive trade-off between the achievable BER performance and the complexity imposed. In order to improve the energy-efficiency, the hard-decision-based MSDSD algorithm is further developed to its soft-decision-based version, namely to the soft-input soft-output MSDSD (SISO-MSDSD). Furthermore, for the sake of exploiting the benefits of cooperative communications, we devise a new multiple-path propagation-aided MSDSD algorithm as a beneficial variant of the conventional MSDSD algorithm, which is further developed to the relay-aided MSDSD algorithm. However, relay-assisted transmissions increase the amount of interference imposed. Hence, in order to suppress the successive relaying induced interference, namely both the inter-relay interference (IRI) and the co-channel interference (CCI), we invoke the DS-CDMA multiple access technique. Consequently, a noncoherent successive relaying (NC-SR) aided cooperative DS-CDMA uplink is conceived, where the typical 50% half-duplex relaying induced throughput loss is converted to a potential user-load reduction for the DS-CDMA system, since the SRAN requires two - rather than a single - spreading codes for each user. First, the AF protocol is invoked for the successive AF relaying aided DS-CDMA uplink, where initially a simple single-user scenario and then a realistic multi-user scenario are investigated. The evaluation of the associated noncoherent discreteinput continuous-output memoryless channel (DCMC) capacity indicates that our AF based SRAN is capable of significantly outperforming both the conventional AF relaying and the single-link direct-transmission. Then, as a counterpart, the successive DF relaying aided DS-CDMA uplink is also conceived, where a multi-user scenario is considered. The noncoherent DCMC capacity of the DF based SRAN reveals that the DF based SRAN may outperform the AF based SRAN, especially in the low SNR region. Furthermore, a relay-aided SISO-MSDSD assisted three-stage iterative transceiver is designed for supporting the operation of the proposed DF based SRAN, which is capable of operating close to the system’s capacity, while halving the system complexity imposed by the conventional single-path SISO-MSDSD aided distributed turbo decoder. As a further advance, we also consider a multi-user multi-relay DS-CDMA uplink. In order to efficiently organize the cooperation among the multiple nodes of this large-scale wireless network, we further develop the concept of adaptive network coded cooperation (ANCC) to its generalized version, namely to our “GANC” regime, which allows arbitrary channel coding schemes to serve as the cross-layer network coding, while still adapting to both network topology changes and to link failures. Upon incorporating the proposed GANC regime into the SRAN, we construct the GANC aided SRAN. In the spirit of the joint network-channel coding (JNCC) scheme, we devise a generalized iterative detection based three-stage transceiver architecture for the proposed GANC aided SRAN. The proposed transceiver is also adaptive to both network topology changes as well as to link failures. We demonstrate that combining two DF based SRANs and operating them under the proposed GANC regime is capable of attaining a significant power gain with respect to operating them independently, i.e. without any cooperative between them. Employing the DS-CDMA technique for suppressing the successive relaying-induced interference may lead to a potential user-load reduction for the DS-CDMA system. In order to mitigate the interference without requiring any extra orthogonal channel resources, we proposed a new multiple-symbol differential interference suppression (MS-DIS) regime, which is a novel amalgam of the adaptive modified Newton algorithm and of our SISO-MSDSD algorithm. Consequently, a MS-DIS-assisted plus relay-aided SISO-MSDSD based three-stage concatenated turbo transceiver is designed, which is capable of efficiently suppressing the interference at the expense of imposing as little as 2% training overhead, despite experiencing severely time-selective Rayleigh fading associated with a high Doppler frequency.
485

A Bayesian network model for entity-oriented semantic web search

Koumenides, Christos January 2013 (has links)
The rise of standards for semi-structured machine processable information and the increasing awareness of the potentials of a semantic Web are leading the way towards a more meaningful Web of data. Questions regarding location and retrieval of relevant data remain fundamental in achieving a good integration of disparate resources and the effective delivery of data items to the needs of particular applications and users. We consider the basis of such a framework as an Information Retrieval system that can cope with semi-structured data. This thesis examines the development of an Information Retrieval model to support text-based search over formal Semantic Web knowledge bases. Our semantic search model adapts Bayesian Networks as a unifying modelling framework to represent, and make explicit in the retrieval process, the presence of multiple relations that potentially link semantic resources together or with primitive data values, as it is customary with Semantic Web data. We achieve this by developing a generative model that is capable to express Semantic Web data and expose their structure to statistical scrutiny and generation of inference procedures. We employ a variety of techniques to bring together a unified ranking strategy with a sound mathematical foundation and potential for further extensions and modifications. Part of our goal in designing this model has been to enable reasoning with more complex or expressive information requests, with semantics specified explicitly by users or incorporated via more implicit bindings. The ground foundations of the model offer a rich and extensible setting to satisfy an interesting set of queries and incorporate a variety of techniques for fusing probabilistic evidence, both new and familiar. Empirical evaluation of the model is carried out using conventional Recall/Precision effectiveness metrics to demonstrate its performance over a collection of RDF transposed government catalogue records. Statistical significance tests are employed to compare different implementations of the model over different query sets of relative complexity.
486

Energy- and information-managed wireless sensor networks : modelling and simulation

Merrett, Geoff V. January 2008 (has links)
Wireless Sensor Networks (WSNs) allow the remote and distributed monitoring of parameters in their deployed environment. WSNs are receiving increasing research interest, due to their ability to enable a wide range of applications, and their potential to have a major impact on ubiquitous computing. Many research challenges are encountered in retaining a useful network lifetime under constrains imposed by the limited energy reserves that are inherent in the small, locally-powered sensor nodes. This research addresses some of these challenges through the development and evaluation of energy- and information-managed algorithms leading to increased network lifetime. The first contribution of this research is the development of an Information manageD Energy-aware ALgorithm for Sensor networks with Rule Managed Reporting (IDEALS/RMR). IDEALS/RMR is an application-independent, localised system to control and manage the degradation of a network through the positive discrimination of packets. This is achieved by the novel combination of energy management (through IDEALS) and information management (through RMR) which increases the network lifetime at the possible expense of often trivial data. IDEALS/RMR is particularly suited to applications where sensor nodes are small, energy constrained, embedded devices particularly those that feature energy harvesting) that are required to report data in an unassisted fashion. The second contribution of this research is the analysis of various environmental and physical aspects of WSNs, and the effect that they have on the operation of nodes and networks. These aspects include energy components (stores, sources and consumers), sensing devices, wireless communication, and timing; these aspects are independently modelled and, through simulation, their effect on the operation of the network is quantified. The third contribution of this research is the evaluation of IDEALS/RMR using a simulator that has been specifically developed to integrate both the proposed environmental and physical models, and a novel node architecture that facilitates structured software design. A scenario depicting the use of a WSN to monitor pump temperature in a water pumping station is simulated, and highlights the benefits of the developed algorithms.
487

Enhancing TCP delivery over wireless networks

Lien, Kai-Wen January 2009 (has links)
Wireless communication has become a significant life style in the daily use. The wireless communication can be used to extend the service of wired communication. Based on the idea of simplicity, Transmission Control Protocol (TCP) has been used widely over wired network. When network applications take place using wireless link, TCP is still useful because partial wired connection might be necessary. Although wired and wireless communication do share something in common, they have distinct features. Therefore, TCP needs to be adjusted to fit in the wireless environment. This research aims to enhance wireless TCP performance. In order to study network protocols and behaviours, network simulators are often used for researchers to configure and monitor the network factors and system states. Unfortunately, most network simulators cannot demonstrate what the real network does. They are applications. In this research, a network simulator based on the real Linux TCP/IP stacks is proposed. This simulator is able to not only simulate the wired network, but also allow users to extend its structure for live wireless emulation. By means of simulator and emulator, researchers can understand and configure detail factors for further experiments. Eventually, a new wireless TCP enhancement is proposed. In this research, some contributions are delivered. Firstly, a network simulator based on Linux TCP stacks is implemented. Secondly, a wireless emulator and test environment are built, so the wireless factors can be configured and performance can be monitored. Thirdly, a wireless TCP mechanism, TCP NewZag, is proposed. Finally, several experiments to show the value of NewZag are reported in the thesis.
488

An investigation of delay fault testing for multi voltage design

Zain Ali, Noohul Basheer January 2009 (has links)
Multi Voltage Design(MVD) has been successfully applied in contemporary processors as a technique to reduce energy consumption. This work is aimed at finding a generalised delay testing method for MVD. There has been little work to date on testing such systems, but testing the smallest number of operating voltages reduces testing costs. In the initial stage, the impact of varying supply voltage on different types of physical defects is analysed. Simulation results indicate that it is neces- sary to conduct test at more than one operating voltage and the lowest operating voltage does not necessarily give the best fault coverage. The second part of this work is related to the issues in the testing of level shifters in a MVD environment. The testing of level shifters was analysed to determine if high test coverage can be achieved at a single supply voltage. Resistive opens and shorts were considered and it was shown that, for testing purposes, consideration of purely digital fault effects is sufficient. Multiple faults were also considered. In all cases, it can be concluded that a single supply voltage is sufficient to test the level shifters. To further enhance the quality of test, we have proposed fault modelling and simulations using VHDL-AMS. Our simulation results show that the model derived using simplified VHDL-AMS gives acceptable results and significantly reduces the fault simulations time.
489

Multiple objective sensor management and optimisation

Page, Scott F. January 2009 (has links)
One of the key challenges associated with exploiting modern Autonomous Vehicle technology for military surveillance tasks is the development of Sensor Management strategies which maximise the performance of the on-board Data-Fusion systems. The focus of this thesis is the development of Sensor Management algorithms which aim to optimise target tracking processes. Three principal theoretical and analytical contributions are presented which are related to the manner in which such problems are formulated and subsequently solved. Firstly, the trade-offs between optimising target tracking and other system-level objectives relating to expected operating lifetime are explored in an autonomous ground sensor scenario. This is achieved by modelling the observer trajectory control design as a probabilistic, information--theoretic, multiple-objective optimisation problem. This novel approach explores the relationships between the changes in sensor-target geometry that are induced by tracking performance measures and those relating to power consumption. This culminates in a novel observer trajectory control algorithm based on the minimax approach. The second contribution is an analysis of the propagation of error through a limited-lookahead sensor control feedback loop. In the last decade, it has been shown that the use of such non-myopic (multiple-step) planning strategies can lead to superior performance in many Sensor Management scenarios. However, relatively little is known about the performance of strategies which use different horizon lengths. It is shown that, in the general case, planning performance is a function of the length of the horizon over which the optimisation is performed. While increasing the horizon maximises the chances of achieving global optimality, by revealing information about the substructure of the decision space, it also increases the impact of any prediction error, approximations, or unforeseen risk present within the scenario. These competing mechanisms are demonstrated using an example tracking problem. This provides the motivation for a novel sensor control methodology that employs an adaptive length optimisation horizon. A route to selecting the optimal horizon size is proposed, based on a new non-myopic risk equilibrium which identifies the point where the two competing mechanisms are balanced. The third area of contribution concerns the development of a number of novel optimisation algorithms aimed at solving the resulting sequential decision making problems. These problems are typically solved using stochastic search methods such as Genetic Algorithms or Simulated Annealing. The techniques presented in this thesis are extensions of the recently proposed Repeated Weighted Boosting Search algorithm. In its original form, it is only applicable to continuous, single-objective, optimisation problems. The extensions facilitate application to mixed search spaces and Pareto multiple-objective problems. The resulting algorithms have performance comparable with Genetic Algorithm variants, and offer a number of advantages such as ease of implementation and limited tuning requirements.
490

Inference from binary gene expression data

Tuna, Salih January 2009 (has links)
Microarrays provide a practical method for measuring the mRNA abundances of thousands of genes in a single experiment. Analysing such large dimensional data is a challenge which attracts researchers from many different fields and machine learning is one of them. However, the biological properties of mRNA such as its low stability, measurements being taken from a population of cells rather than from a single cell, etc. should make researchers sceptical about the high numerical precision reported and thus the reproducibility of these measurements. In this study we explore data representation at lower numerical precision, down to binary (retaining only the information whether a gene is expressed or not), thereby improving the quality of inferences drawn from microarray studies. With binary representation, we propose a solution to reduce the effect of algorithmic choice in the pre-processing stages. First we compare the information loss if researchers made the inferences from quantized transcriptome data rather than the continuous values. Classification, clustering, periodicity detection and analysis of developmental time series data are considered here. Our results showed that there is not much information loss with binary data. Then, by focusing on the two most widely used inference tools, classification and clustering, we show that inferences drawn from transcriptome data can actually be improved with a metric suitable for binary data. This is explained with the uncertainties of the probe level data. We also show that binary transcriptome data can be used in cross-platform studies and when used with Tanimoto kernel, this increase the performance of inferences when compared to individual datasets. In the last part of this work we show that binary transcriptome data reduces the effect of algorithm choice for pre-processing raw data. While there are many different algorithms for pre-processing stages there are few guidelines for the users as to which one to choose. In many studies it has been shown that the choice of algorithms has significant impact on the overall results of microarray studies. Here we show in classification, that if transcriptome data is binarized after pre-processed with any combination of algorithms it has the effect of reducing the variability of the results and increasing the performance of the classifier simultaneously.

Page generated in 0.0502 seconds