High-performance parallel systems for data-intensive computing

Thesis (Ph.D.)--University of Washington, 2016-08

Strategies for protecting intellectual property when using CUDA applications on graphics processing units

Recent advances in the massively parallel computational abilities of graphical processing units (GPUs) have increased their use for general purpose computation, as companies look to take advantage of big data processing techniques. This has given rise to the potential for malicious software targeting GPUs, which is of interest to forensic investigators examining the operation of software. The ability to carry out reverse-engineering of software is of great importance within the security and forensics elds, particularly when investigating malicious software or carrying out forensic analysis following a successful security breach. Due to the complexity of the Nvidia CUDA (Compute Uni ed Device Architecture) framework, it is not clear how best to approach the reverse engineering of a piece of CUDA software. We carry out a review of the di erent binary output formats which may be encountered from the CUDA compiler, and their implications on reverse engineering. We then demonstrate the process of carrying out disassembly of an example CUDA application, to establish the various techniques available to forensic investigators carrying out black-box disassembly and reverse engineering of CUDA binaries. We show that the Nvidia compiler, using default settings, leaks useful information. Finally, we demonstrate techniques to better protect intellectual property in CUDA algorithm implementations from reverse engineering.

Development of a real time Image analysis system for traffic monitoring applications.

Image processing offers unparalleled potential for traffic monitoring and control. For many years engineers have attempted to perfect the art of automatic data abstraction from sequences of video images. This paper outlines a research project undertaken at Napier University by the authors in the field of image processing for automatic traffic analysis. A software based system implementing TRIP algorithms to count cars and measure vehicle speed has been developed by members of the Transport Engineering Research Unit (TERU) at the University. The TRIP algorithm has been ported and evaluated on an IBM PC platform with a view to hardware implementation of the pre-processing routines required for vehicle detection. Results show that a software based traffic counting system is realisable for single window processing. Due to the high volume of data required to be processed for full frames or multiple lanes, system operations in real time are limited. Therefore specific hardware is required to be designed. The paper outlines a hardware design for implementation of inter-frame and background differencing, background updating and shadow removal techniques. Preliminary results showing the processing time and counting accuracy for the routines implemented in software are presented and a real time hardware pre-processing architecture is described.

Processing Bio-Argo nitrate concentration at the DAC Level

The only method used to date to measure dissolved nitrate concentration (NITRATE) with sensors mounted on profiling floats is based on the absorption of light at ultraviolet wavelengths by nitrate ion (Johnson and Coletti, 2002; Johnson et al., 2010; 2013; D’Ortenzio et al., 2012). Nitrate has a modest UV absorption band with a peak near 210 nm, which overlaps with the stronger absorption band of bromide, which has a peak near 200 nm. In addition, there is a much weaker absorption due to dissolved organic matter and light scattering by particles (Ogura and Hanya, 1966). The UV spectrum thus consists of three components, bromide, nitrate and a background due to organics and particles. The background also includes thermal effects on the instrument and slow drift. All of these latter effects (organics, particles, thermal effects and drift) tend to be smooth spectra that combine to form an absorption spectrum that is linear in wavelength over relatively short wavelength spans. If the light absorption spectrum is measured in the wavelength range around 217 to 240 nm (the exact range is a bit of a decision by the operator), then the nitrate concentration can be determined. Two different instruments based on the same optical principles are in use for this purpose. The In Situ Ultraviolet Spectrophotometer (ISUS) built at MBARI or at Satlantic has been mounted inside the pressure hull of a Teledyne/Webb Research APEX and NKE Provor profiling floats and the optics penetrate through the upper end cap into the water. The Satlantic Submersible Ultraviolet Nitrate Analyzer (SUNA) is placed on the outside of APEX, Provor, and Navis profiling floats in its own pressure housing and is connected to the float through an underwater cable that provides power and communications. Power, communications between the float controller and the sensor, and data processing requirements are essentially the same for both ISUS and SUNA. There are several possible algorithms that can be used for the deconvolution of nitrate concentration from the observed UV absorption spectrum (Johnson and Coletti, 2002; Arai et al., 2008; Sakamoto et al., 2009; Zielinski et al., 2011). In addition, the default algorithm that is available in Satlantic sensors is a proprietary approach, but this is not generally used on profiling floats. There are some tradeoffs in every approach. To date almost all nitrate sensors on profiling floats have used the Temperature Compensated Salinity Subtracted (TCSS) algorithm developed by Sakamoto et al. (2009), and this document focuses on that method. It is likely that there will be further algorithm development and it is necessary that the data systems clearly identify the algorithm that is used. It is also desirable that the data system allow for recalculation of prior data sets using new algorithms. To accomplish this, the float must report not just the computed nitrate, but the observed light intensity. Then, the rule to obtain only one NITRATE parameter is, if the spectrum is present then, the NITRATE should be recalculated from the spectrum while the computation of nitrate concentration can also generate useful diagnostics of data quality.

CATARINA 2012-Leg 1 - CTD-O2 Data report

The CATARINA Leg1 cruise was carried out from June 22 to July 24 2012 on board the B/O Sarmiento de Gamboa, under the scientific supervision of Aida Rios (CSIC-IIM). It included the occurrence of the OVIDE hydrological section that was performed in June 2002, 2004, 2006, 2008 and 2010, as part of the CLIVAR program (name A25) ), and under the supervision of Herlé Mercier (CNRSLPO). This section begins near Lisbon (Portugal), runs through the West European Basin and the Iceland Basin, crosses the Reykjanes Ridge (300 miles north of Charlie-Gibbs Fracture Zone, and ends at Cape Hoppe (southeast tip of Greenland). The objective of this repeated hydrological section is to monitor the variability of water mass properties and main current transports in the basin, complementing the international observation array relevant for climate studies. In addition, the Labrador Sea was partly sampled (stations 101-108) between Greenland and Newfoundland, but heavy weather conditions prevented the achievement of the section south of 53°40’N. The quality of CTD data is essential to reach the first objective of the CATARINA project, i.e. to quantify the Meridional Overturning Circulation and water mass ventilation changes and their effect on the changes in the anthropogenic carbon ocean uptake and storage capacity. The CATARINA project was mainly funded by the Spanish Ministry of Sciences and Innovation and co-funded by the Fondo Europeo de Desarrollo Regional. The hydrological OVIDE section includes 95 surface-bottom stations from coast to coast, collecting profiles of temperature, salinity, oxygen and currents, spaced by 2 to 25 Nm depending on the steepness of the topography. The position of the stations closely follows that of OVIDE 2002. In addition, 8 stations were carried out in the Labrador Sea. From the 24 bottles closed at various depth at each stations, samples of sea water are used for salinity and oxygen calibration, and for measurements of biogeochemical components that are not reported here. The data were acquired with a Seabird CTD (SBE911+) and an SBE43 for the dissolved oxygen, belonging to the Spanish UTM group. The software SBE data processing was used after decoding and cleaning the raw data. Then, the LPO matlab toolbox was used to calibrate and bin the data as it was done for the previous OVIDE cruises, using on the one hand pre and post-cruise calibration results for the pressure and temperature sensors (done at Ifremer) and on the other hand the water samples of the 24 bottles of the rosette at each station for the salinity and dissolved oxygen data. A final accuracy of 0.002°C, 0.002 psu and 0.04 ml/l (2.3 umol/kg) was obtained on final profiles of temperature, salinity and dissolved oxygen, compatible with international requirements issued from the WOCE program.

Herramientas automáticas de diagnóstico y corrección para aerofotogrametría

En Aerofotogrametría, el proceso de restitución (paso de imagen a formato electrónico vectorizado) es realizado por un operador humano, con asistencia de hardware y Software especializado -- Dicho proceso implica la traducción de accidentes geográficos, detalles topográficos, etc., la cual conlleva errores tanto geométricos (precisión) como topológicos (conectividad) de los datos digitales vectorizados -- Adicionalmente, aun si la vectorizacion es perfecta, los editores en etapas subsecuentes deben realizar tareas repetitivas: formateo, marcado, ajuste de convenciones, etc., que por el tamaño de los archivos de datos se hacen prolongadas y propensas al error -- Tanto los procesos de corrección como de formateo y marcado requieren además la ejecución de entradas / salidas con el usuario en el computador, proceso que es particularmente lento -- Esta investigación presenta el desarrollo de herramientas automáticas de (i) detección y corrección de errores comunes en los planos restituidos, (ii) partición y re-agrupación inteligentes de planos grandes, y (iii) formateo y marcado automático -- El desarrollo de software se hace usando el standard AIS (Application Interface Specification), lo que lo hace portable a los modeladores cuya interface AIS haya sido implementada -- El proyecto se desarrolla para la firma AeroEstudios LTDA de Colombia, la cual lo ha incorporado a sus herramientas de procesamiento de información digital

Three Carrier Ambiguity Resolutions : Generalised problems, models and solutions

In this paper, the problems of three carrier phase ambiguity resolution (TCAR) and position estimation (PE) are generalized as real time GNSS data processing problems for a continuously observing network on large scale. In order to describe these problems, a general linear equation system is presented to uniform various geometry-free, geometry-based and geometry-constrained TCAR models, along with state transition questions between observation times. With this general formulation, generalized TCAR solutions are given to cover different real time GNSS data processing scenarios, and various simplified integer solutions, such as geometry-free rounding and geometry-based LAMBDA solutions with single and multiple-epoch measurements. In fact, various ambiguity resolution (AR) solutions differ in the floating ambiguity estimation and integer ambiguity search processes, but their theoretical equivalence remains under the same observational systems models and statistical assumptions. TCAR performance benefits as outlined from the data analyses in some recent literatures are reviewed, showing profound implications for the future GNSS development from both technology and application perspectives.

A hidden Markov model and relative entropy rate approach to vision-based dim target detection for UAV sense-and-avoid

Uninhabited aerial vehicles (UAVs) are a cutting-edge technology that is at the forefront of aviation/aerospace research and development worldwide. Many consider their current military and defence applications as just a token of their enormous potential. Unlocking and fully exploiting this potential will see UAVs in a multitude of civilian applications and routinely operating alongside piloted aircraft. The key to realising the full potential of UAVs lies in addressing a host of regulatory, public relation, and technological challenges never encountered be- fore. Aircraft collision avoidance is considered to be one of the most important issues to be addressed, given its safety critical nature. The collision avoidance problem can be roughly organised into three areas: 1) Sense; 2) Detect; and 3) Avoid. Sensing is concerned with obtaining accurate and reliable information about other aircraft in the air; detection involves identifying potential collision threats based on available information; avoidance deals with the formulation and execution of appropriate manoeuvres to maintain safe separation. This thesis tackles the detection aspect of collision avoidance, via the development of a target detection algorithm that is capable of real-time operation onboard a UAV platform. One of the key challenges of the detection problem is the need to provide early warning. This translates to detecting potential threats whilst they are still far away, when their presence is likely to be obscured and hidden by noise. Another important consideration is the choice of sensors to capture target information, which has implications for the design and practical implementation of the detection algorithm. The main contributions of the thesis are: 1) the proposal of a dim target detection algorithm combining image morphology and hidden Markov model (HMM) filtering approaches; 2) the novel use of relative entropy rate (RER) concepts for HMM filter design; 3) the characterisation of algorithm detection performance based on simulated data as well as real in-flight target image data; and 4) the demonstration of the proposed algorithm's capacity for real-time target detection. We also consider the extension of HMM filtering techniques and the application of RER concepts for target heading angle estimation. In this thesis we propose a computer-vision based detection solution, due to the commercial-off-the-shelf (COTS) availability of camera hardware and the hardware's relatively low cost, power, and size requirements. The proposed target detection algorithm adopts a two-stage processing paradigm that begins with an image enhancement pre-processing stage followed by a track-before-detect (TBD) temporal processing stage that has been shown to be effective in dim target detection. We compare the performance of two candidate morphological filters for the image pre-processing stage, and propose a multiple hidden Markov model (MHMM) filter for the TBD temporal processing stage. The role of the morphological pre-processing stage is to exploit the spatial features of potential collision threats, while the MHMM filter serves to exploit the temporal characteristics or dynamics. The problem of optimising our proposed MHMM filter has been examined in detail. Our investigation has produced a novel design process for the MHMM filter that exploits information theory and entropy related concepts. The filter design process is posed as a mini-max optimisation problem based on a joint RER cost criterion. We provide proof that this joint RER cost criterion provides a bound on the conditional mean estimate (CME) performance of our MHMM filter, and this in turn establishes a strong theoretical basis connecting our filter design process to filter performance. Through this connection we can intelligently compare and optimise candidate filter models at the design stage, rather than having to resort to time consuming Monte Carlo simulations to gauge the relative performance of candidate designs. Moreover, the underlying entropy concepts are not constrained to any particular model type. This suggests that the RER concepts established here may be generalised to provide a useful design criterion for multiple model filtering approaches outside the class of HMM filters. In this thesis we also evaluate the performance of our proposed target detection algorithm under realistic operation conditions, and give consideration to the practical deployment of the detection algorithm onboard a UAV platform. Two fixed-wing UAVs were engaged to recreate various collision-course scenarios to capture highly realistic vision (from an onboard camera perspective) of the moments leading up to a collision. Based on this collected data, our proposed detection approach was able to detect targets out to distances ranging from about 400m to 900m. These distances, (with some assumptions about closing speeds and aircraft trajectories) translate to an advanced warning ahead of impact that approaches the 12.5 second response time recommended for human pilots. Furthermore, readily available graphic processing unit (GPU) based hardware is exploited for its parallel computing capabilities to demonstrate the practical feasibility of the proposed target detection algorithm. A prototype hardware-in- the-loop system has been found to be capable of achieving data processing rates sufficient for real-time operation. There is also scope for further improvement in performance through code optimisations. Overall, our proposed image-based target detection algorithm offers UAVs a cost-effective real-time target detection capability that is a step forward in ad- dressing the collision avoidance issue that is currently one of the most significant obstacles preventing widespread civilian applications of uninhabited aircraft. We also highlight that the algorithm development process has led to the discovery of a powerful multiple HMM filtering approach and a novel RER-based multiple filter design process. The utility of our multiple HMM filtering approach and RER concepts, however, extend beyond the target detection problem. This is demonstrated by our application of HMM filters and RER concepts to a heading angle estimation problem.

Developing regional precise positioning services using the legacy and future GNSS receivers

This paper presents an overview of technical solutions for regional area precise GNSS positioning services such as in Queensland. The research focuses on the technical and business issues that currently constrain GPS-based local area Real Time Kinematic (RTK) precise positioning services so as to operate in future across larger regional areas, and therefore support services in agriculture, mining, utilities, surveying, construction, and others. The paper first outlines an overall technical framework that has been proposed to transition the current RTK services to future larger scale coverage. The framework enables mixed use of different reference GNSS receiver types, dual- or triple-frequency, single or multiple systems, to provide RTK correction services to users equipped with any type of GNSS receivers. Next, data processing algorithms appropriate for triple-frequency GNSS signals are reviewed and some key performance benefits of using triple carrier signals for reliable RTK positioning over long distances are demonstrated. A server-based RTK software platform is being developed to allow for user positioning computations at server nodes instead of on the user's device. An optimal deployment scheme for reference stations across a larger-scale network has been suggested, given restrictions such as inter-station distances, candidates for reference locations, and operational modes. For instance, inter-station distances between triple-frequency receivers can be extended to 150km, which doubles the distance between dual-frequency receivers in the existing RTK network designs.