Garbage collection optimization for non uniform memory access architectures

Cache-coherent non uniform memory access (ccNUMA) architecture is a standard design pattern for contemporary multicore processors, and future generations of architectures are likely to be NUMA. NUMA architectures create new challenges for managed runtime systems. Memory-intensive applications use the system’s distributed memory banks to allocate data, and the automatic memory manager collects garbage left in these memory banks. The garbage collector may need to access remote memory banks, which entails access latency overhead and potential bandwidth saturation for the interconnection between memory banks. This dissertation makes five significant contributions to garbage collection on NUMA systems, with a case study implementation using the Hotspot Java Virtual Machine. It empirically studies data locality for a Stop-The-World garbage collector when tracing connected objects in NUMA heaps. First, it identifies a locality richness which exists naturally in connected objects that contain a root object and its reachable set— ‘rooted sub-graphs’. Second, this dissertation leverages the locality characteristic of rooted sub-graphs to develop a new NUMA-aware garbage collection mechanism. A garbage collector thread processes a local root and its reachable set, which is likely to have a large number of objects in the same NUMA node. Third, a garbage collector thread steals references from sibling threads that run on the same NUMA node to improve data locality. This research evaluates the new NUMA-aware garbage collector using seven benchmarks of an established real-world DaCapo benchmark suite. In addition, evaluation involves a widely used SPECjbb benchmark and Neo4J graph database Java benchmark, as well as an artificial benchmark. The results of the NUMA-aware garbage collector on a multi-hop NUMA architecture show an average of 15% performance improvement. Furthermore, this performance gain is shown to be as a result of an improved NUMA memory access in a ccNUMA system. Fourth, the existing Hotspot JVM adaptive policy for configuring the number of garbage collection threads is shown to be suboptimal for current NUMA machines. The policy uses outdated assumptions and it generates a constant thread count. In fact, the Hotspot JVM still uses this policy in the production version. This research shows that the optimal number of garbage collection threads is application-specific and configuring the optimal number of garbage collection threads yields better collection throughput than the default policy. Fifth, this dissertation designs and implements a runtime technique, which involves heuristics from dynamic collection behavior to calculate an optimal number of garbage collector threads for each collection cycle. The results show an average of 21% improvements to the garbage collection performance for DaCapo benchmarks.

Catalogue of data and platforms at Network GDAC level, including the example of Copernicus In Situ TAC

This document presents catalogue techniques used at network GDAC level to facilitate the discovery of platforms and data files. Some AtlantOS networks are organized as DAC-GDACs that continuously update a catalogue of metadata on observation datasets and platforms: • A DAC is a Data Assembly Centre operating at national or regional scale. It manages data and metadata for its area with a direct link to Scientifics and Operators. The DAC pushes observations to the network GDAC. • A GDAC is a Global Data Assembly Centre. It is designed for a global observation network such as Argo, OceanSITES, DBCP, EGO, Gosud, etc… The GDAC aggregates data and metadata of an observation network, in real-time and delayed mode, provided by DACs.

The EMSO-ERIC Pan-European Consortium: Data Benefits and Lessons Learned as the Legal Entity Forms

The European Multidisciplinary Seafloor and water-column Observatory (EMSO) European Research Infrastructure Consortium (ERIC) provides power, communications, sensors, and data infrastructure for continuous, high-resolution, (near-)real-time, interactive ocean observations across a multidisciplinary and interdisciplinary range of research areas including biology, geology, chemistry, physics, engineering, and computer science, from polar to subtropical environments, through the water column down to the abyss. Eleven deep-sea and four shallow nodes span from the Arctic through the Atlantic and Mediterranean, to the Black Sea. Coordination among the consortium nodes is being strengthened through the EMSOdev project (H2020), which will produce the EMSO Generic Instrument Module (EGIM). Early installations are now being upgraded, for example, at the Ligurian, Ionian, Azores, and Porcupine Abyssal Plain (PAP) nodes. Significant findings have been flowing in over the years; for example, high-frequency surface and subsurface water-column measurements of the PAP node show an increase in seawater pCO2 (from 339 μatm in 2003 to 353 μatm in 2011) with little variability in the mean air-sea CO2 flux. In the Central Eastern Atlantic, the Oceanic Platform of the Canary Islands open-ocean canary node (aka ESTOC station) has a long-standing time series on water column physical, biogeochemical, and acidification processes that have contributed to the assessment efforts of the Intergovernmental Panel on Climate Change (IPCC). EMSO not only brings together countries and disciplines but also allows the pooling of resources and coordination to assemble harmonized data into a comprehensive regional ocean picture, which will then be made available to researchers and stakeholders worldwide on an open and interoperable access basis.

Trace Oblivious Program Execution

The big data era has dramatically transformed our lives; however, security incidents such as data breaches can put sensitive data (e.g. photos, identities, genomes) at risk. To protect users' data privacy, there is a growing interest in building secure cloud computing systems, which keep sensitive data inputs hidden, even from computation providers. Conceptually, secure cloud computing systems leverage cryptographic techniques (e.g., secure multiparty computation) and trusted hardware (e.g. secure processors) to instantiate a “secure” abstract machine consisting of a CPU and encrypted memory, so that an adversary cannot learn information through either the computation within the CPU or the data in the memory. Unfortunately, evidence has shown that side channels (e.g. memory accesses, timing, and termination) in such a “secure” abstract machine may potentially leak highly sensitive information, including cryptographic keys that form the root of trust for the secure systems. This thesis broadly expands the investigation of a research direction called trace oblivious computation, where programming language techniques are employed to prevent side channel information leakage. We demonstrate the feasibility of trace oblivious computation, by formalizing and building several systems, including GhostRider, which is a hardware-software co-design to provide a hardware-based trace oblivious computing solution, SCVM, which is an automatic RAM-model secure computation system, and ObliVM, which is a programming framework to facilitate programmers to develop applications. All of these systems enjoy formal security guarantees while demonstrating a better performance than prior systems, by one to several orders of magnitude.

A Mapping-Scheduling Algorithm for Hardware Acceleration on Reconfigurable Platforms

Reconfigurable platforms are a promising technology that offers an interesting trade-off between flexibility and performance, which many recent embedded system applications demand, especially in fields such as multimedia processing. These applications typically involve multiple ad-hoc tasks for hardware acceleration, which are usually represented using formalisms such as Data Flow Diagrams (DFDs), Data Flow Graphs (DFGs), Control and Data Flow Graphs (CDFGs) or Petri Nets. However, none of these models is able to capture at the same time the pipeline behavior between tasks (that therefore can coexist in order to minimize the application execution time), their communication patterns, and their data dependencies. This paper proves that the knowledge of all this information can be effectively exploited to reduce the resource requirements and the timing performance of modern reconfigurable systems, where a set of hardware accelerators is used to support the computation. For this purpose, this paper proposes a novel task representation model, named Temporal Constrained Data Flow Diagram (TCDFD), which includes all this information. This paper also presents a mapping-scheduling algorithm that is able to take advantage of the new TCDFD model. It aims at minimizing the dynamic reconfiguration overhead while meeting the communication requirements among the tasks. Experimental results show that the presented approach achieves up to 75% of resources saving and up to 89% of reconfiguration overhead reduction with respect to other state-of-the-art techniques for reconfigurable platforms.

Description of the Argo GDAC File Checks: Data Format and Consistency Checks

Every Argo data file submitted by a DAC for distribution on the GDAC has its format and data consistency checked by the Argo FileChecker. Two types of checks are applied: 1. Format checks. Ensures the file formats match the Argo standards precisely. 2. Data consistency checks. Additional data consistency checks are performed on a file after it passes the format checks. These checks do not duplicate any of the quality control checks performed elsewhere. These checks can be thought of as “sanity checks” to ensure that the data are consistent with each other. The data consistency checks enforce data standards and ensure that certain data values are reasonable and/or consistent with other information in the files. Examples of the “data standard” checks are the “mandatory parameters” defined for meta-data files and the technical parameter names in technical data files. Files with format or consistency errors are rejected by the GDAC and are not distributed. Less serious problems will generate warnings and the file will still be distributed on the GDAC. Reference Tables and Data Standards: Many of the consistency checks involve comparing the data to the published reference tables and data standards. These tables are documented in the User’s Manual. (The FileChecker implements “text versions” of these tables.)

PREVALENCE OF IRON DEFICIENCY ANAEMIA AND DIETARY IRON INTAKE AMONG INFANTS AGED SIX TO NINE MONTHS IN KEIYO SOUTH SUB COUNTY, KENYA

The increased prevalence of iron deficiency among infants can be attributed to the consumption of an iron deficient diet or a diet that interferes with iron absorption at the critical time of infancy, among other factors. The gradual shift from breast milk to other foods and liquids is a transition period which greatly contributes to iron deficiency anaemia (IDA). The purpose of this research was to assess iron deficiency anaemia among infants aged six to nine months in Keiyo South Sub County. The specific objectives of this study were: to establish the prevalence of infants with iron deficiency anaemia and dietary iron intake among infants aged 6 to 9 months. The cross sectional study design was adopted in this survey. This study was conducted in three health facilities in Keiyo South Sub County. The infants were selected by use of a two stage cluster sampling procedure. Systematic random sampling was then used to select a total of 244 mothers and their infants. Eighty two (82) infants were selected from Kamwosor sub-district hospital and eighty one (81) from both Nyaru and Chepkorio health facilities. Interview schedules, 24-hour dietary recall and food frequency questionnaires were used for collection of dietary iron intake. Biochemical tests were carried out by use of the Hemo-control photochrometer at the health facilities. Infants whose hemoglobin levels were less than 11g/dl were considered anaemic. Further, peripheral blood smears were conducted to ascertain the type of nutritional anaemia. Data was analyzed using the Statistical Package for Social Sciences (SPSS) computer software version 17, 2009. Dietary iron intake was analyzed using the NutriSurvey 2007 computer software. Results indicated that the mean hemoglobin values were 11.3± 0.84 g/dl. Twenty one percent (21.7%) of the infants had anaemia and further 100% of peripheral blood smears indicated iron deficiency anaemia. Dietary iron intake was a predictor of iron deficiency anaemia in this study (t=-3.138; p=0.01). Iron deficiency anaemia was evident among infants in Keiyo South Sub County. The Ministry of Health should formulate and implement policies on screening for anaemia and ensure intensive nutrition education on iron rich diets during child welfare clinics.

Multiscale Modeling and Simulation of Stepped Crystal Surfaces

A primary goal of this dissertation is to understand the links between mathematical models that describe crystal surfaces at three fundamental length scales: The scale of individual atoms, the scale of collections of atoms forming crystal defects, and macroscopic scale. Characterizing connections between different classes of models is a critical task for gaining insight into the physics they describe, a long-standing objective in applied analysis, and also highly relevant in engineering applications. The key concept I use in each problem addressed in this thesis is coarse graining, which is a strategy for connecting fine representations or models with coarser representations. Often this idea is invoked to reduce a large discrete system to an appropriate continuum description, e.g. individual particles are represented by a continuous density. While there is no general theory of coarse graining, one closely related mathematical approach is asymptotic analysis, i.e. the description of limiting behavior as some parameter becomes very large or very small. In the case of crystalline solids, it is natural to consider cases where the number of particles is large or where the lattice spacing is small. Limits such as these often make explicit the nature of links between models capturing different scales, and, once established, provide a means of improving our understanding, or the models themselves. Finding appropriate variables whose limits illustrate the important connections between models is no easy task, however. This is one area where computer simulation is extremely helpful, as it allows us to see the results of complex dynamics and gather clues regarding the roles of different physical quantities. On the other hand, connections between models enable the development of novel multiscale computational schemes, so understanding can assist computation and vice versa. Some of these ideas are demonstrated in this thesis. The important outcomes of this thesis include: (1) a systematic derivation of the step-flow model of Burton, Cabrera, and Frank, with corrections, from an atomistic solid-on-solid-type models in 1+1 dimensions; (2) the inclusion of an atomistically motivated transport mechanism in an island dynamics model allowing for a more detailed account of mound evolution; and (3) the development of a hybrid discrete-continuum scheme for simulating the relaxation of a faceted crystal mound. Central to all of these modeling and simulation efforts is the presence of steps composed of individual layers of atoms on vicinal crystal surfaces. Consequently, a recurring theme in this research is the observation that mesoscale defects play a crucial role in crystal morphological evolution.

The scientific rationale, design and implementation plan for a Biogeochemical-Argo float array

Biogeochemical-Argo is the extension of the Argo array of profiling floats to include floats that are equipped with biogeochemical sensors for pH, oxygen, nitrate, chlorophyll, suspended particles, and downwelling irradiance. Argo is a highly regarded, international program that measures the changing ocean temperature (heat content) and salinity with profiling floats distributed throughout the ocean. Newly developed sensors now allow profiling floats to also observe biogeochemical properties with sufficient accuracy for climate studies. This extension of Argo will enable an observing system that can determine the seasonal to decadal-scale variability in biological productivity, the supply of essential plant nutrients from deep-waters to the sunlit surface layer, ocean acidification, hypoxia, and ocean uptake of CO2. Biogeochemical-Argo will drive a transformative shift in our ability to observe and predict the effects of climate change on ocean metabolism, carbon uptake, and living marine resource management. Presently, vast areas of the open ocean are sampled only once per decade or less, with sampling occurring mainly in summer. Our ability to detect changes in biogeochemical processes that may occur due to the warming and acidification driven by increasing atmospheric CO2, as well as by natural climate variability, is greatly hindered by this undersampling. In close synergy with satellite systems (which are effective at detecting global patterns for a few biogeochemical parameters, but only very close to the sea surface and in the absence of clouds), a global array of biogeochemical sensors would revolutionize our understanding of ocean carbon uptake, productivity, and deoxygenation. The array would reveal the biological, chemical, and physical events that control these processes. Such a system would enable a new generation of global ocean prediction systems in support of carbon cycling, acidification, hypoxia and harmful algal blooms studies, as well as the management of living marine resources. In order to prepare for a global Biogeochemical-Argo array, several prototype profiling float arrays have been developed at the regional scale by various countries and are now operating. Examples include regional arrays in the Southern Ocean (SOCCOM ), the North Atlantic Sub-polar Gyre (remOcean ), the Mediterranean Sea (NAOS ), the Kuroshio region of the North Pacific (INBOX ), and the Indian Ocean (IOBioArgo ). For example, the SOCCOM program is deploying 200 profiling floats with biogeochemical sensors throughout the Southern Ocean, including areas covered seasonally with ice. The resulting data, which are publically available in real time, are being linked with computer models to better understand the role of the Southern Ocean in influencing CO2 uptake, biological productivity, and nutrient supply to distant regions of the world ocean. The success of these regional projects has motivated a planning meeting to discuss the requirements for and applications of a global-scale Biogeochemical-Argo program. The meeting was held 11-13 January 2016 in Villefranche-sur-Mer, France with attendees from eight nations now deploying Argo floats with biogeochemical sensors present to discuss this topic. In preparation, computer simulations and a variety of analyses were conducted to assess the resources required for the transition to a global-scale array. Based on these analyses and simulations, it was concluded that an array of about 1000 biogeochemical profiling floats would provide the needed resolution to greatly improve our understanding of biogeochemical processes and to enable significant improvement in ecosystem models. With an endurance of four years for a Biogeochemical-Argo float, this system would require the procurement and deployment of 250 new floats per year to maintain a 1000 float array. The lifetime cost for a Biogeochemical-Argo float, including capital expense, calibration, data management, and data transmission, is about $100,000. A global Biogeochemical-Argo system would thus cost about $25,000,000 annually. In the present Argo paradigm, the US provides half of the profiling floats in the array, while the EU, Austral/Asia, and Canada share most the remaining half. If this approach is adopted, the US cost for the Biogeochemical-Argo system would be ~$12,500,000 annually and ~$6,250,000 each for the EU, and Austral/Asia and Canada. This includes no direct costs for ship time and presumes that float deployments can be carried out from future research cruises of opportunity, including, for example, the international GO-SHIP program (http://www.go-ship.org). The full-scale implementation of a global Biogeochemical-Argo system with 1000 floats is feasible within a decade. The successful, ongoing pilot projects have provided the foundation and start for such a system.

An interaction and network perspective of horizontal relationships in the portuguese pharmaceutical industry

Doutoramento em Gestão

Ultrafast transient absorption spectroscopy: principles and applications

The study of photophysical and photochemical processes crosses the interest of many fields of research in physics, chemistry and biology. In particular, the photophysical and photochemical reactions, after light absorption by a photosynthetic pigment-protein complex, are among the fastest events in biology, taking place on timescales ranging from tens of femtoseconds to a few nanoseconds. Among the experimental approaches developed for this purpose, the advent of ultrafast transient absorption spectroscopy has become a powerful and widely used technique.[1,2] Focusing on the process of photosynthesis, it relies upon the efficient absorption and conversion of the radiant energy from the Sun. Chlorophylls and carotenoids are the main players in the process. Photosynthetic pigments are typically arranged in a highly organized fashion to constitute antennas and reaction centers, supramolecular devices where light harvesting and charge separation take place. The very early steps in the photosynthetic process take place after the absorption of a photon by an antenna system, which harvests light and eventually delivers it to the reaction center. In order to compete with internal conversion, intersystem crossing, and fluorescence, which inevitably lead to energy loss, the energy and electron transfer processes that fix the excited-state energy in photosynthesis must be extremely fast. In order to investigate these events, ultrafast techniques down to a sub-100 fs resolution must be used. In this way, energy migration within the system as well as the formation of new chemical species such as charge-separated states can be tracked in real time. This can be achieved by making use of ultrafast transient absorption spectroscopy. The basic principles of this notable technique, instrumentation, and some recent applications to photosynthetic systems[3] will be described. Acknowledgements M. Moreno Oliva thanks the MINECO for a “Juan de la Cierva-Incorporación” research contract. References [1] U. Megerle, I. Pugliesi, C. Schriever, C.F. Sailer and E. Riedle, Appl. Phys. B, 96, 215 – 231 (2009). [2] R. Berera, R. van Grondelle and J.T.M. Kennis, Photosynth. Res., 101, 105 – 118 (2009). [3] T. Nikkonen, M. Moreno Oliva, A. Kahnt, M. Muuronen, J. Helaja and D.M. Guldi, Chem. Eur. J., 21, 590 – 600 (2015).

Typologies and spatialization of agricultural production systems in Rondônia, Brazil: linking land use, socioeconomics and territorial configuration.

The current Amazon landscape consists of heterogeneous mosaics formed by interactions between the original forest and productive activities. Recognizing and quantifying the characteristics of these landscapes is essential for understanding agricultural production chains, assessing the impact of policies, and in planning future actions. Our main objective was to construct the regionalization of agricultural production for Rondônia State (Brazilian Amazon) at the municipal level. We adopted a decision tree approach, using land use maps derived from remote sensing data (PRODES and TerraClass) combined with socioeconomic data. The decision trees allowed us to allocate municipalities to one of five agricultural production systems: (i) coexistence of livestock production and intensive agriculture; (ii) semi-intensive beef and milk production; (iii) semi-intensive beef production; (iv) intensive beef and milk production, and; (v) intensive beef production. These production systems are, respectively, linked to mechanized agriculture (i), traditional cattle farming with low management, with (ii) or without (iii) a significant presence of dairy farming, and to more intensive livestock farming with (iv) or without (v) a significant presence of dairy farming. The municipalities and associated production systems were then characterized using a wide variety of quantitative metrics grouped into four dimensions: (i) agricultural production; (ii) economics; (iii) territorial configuration, and; (iv) social characteristics. We found that production systems linked to mechanized agriculture predominate in the south of the state, while intensive farming is mainly found in the center of the state. Semi-intensive livestock farming is mainly located close to the southwest frontier and in the north of the state, where human occupation of the territory is not fully consolidated. This distributional pattern reflects the origins of the agricultural production system of Rondônia. Moreover, the characterization of the production systems provides insights into the pattern of occupation of the Amazon and the socioeconomic consequences of continuing agricultural expansion.

Direct Write Fabrication of Waveguides and Interconnects for Optical Printed Wiring Boards

Current copper based circuit technology is becoming a limiting factor in high speed data transfer applications as processors are improving at a faster rate than are developments to increase on board data transfer. One solution is to utilize optical waveguide technology to overcome these bandwidth and loss restrictions. The use of this technology virtually eliminates the heat and cross-talk loss seen in copper circuitry, while also operating at a higher bandwidth. Transitioning current fabrication techniques from small scale laboratory environments to large scale manufacturing presents significant challenges. Optical-to-electrical connections and out-of-plane coupling are significant hurdles in the advancement of optical interconnects. The main goals of this research are the development of direct write material deposition and patterning tools for the fabrication of waveguide systems on large substrates, and the development of out-of-plane coupler components compatible with standard fiber optic cabling. Combining these elements with standard printed circuit boards allows for the fabrication of fully functional optical-electrical-printed-wiring-boards (OEPWBs). A direct dispense tool was designed, assembled, and characterized for the repeatable dispensing of blanket waveguide layers over a range of thicknesses (25-225 µm), eliminating waste material and affording the ability to utilize large substrates. This tool was used to directly dispense multimode waveguide cores which required no UV definition or development. These cores had circular cross sections and were comparable in optical performance to lithographically fabricated square waveguides. Laser direct writing is a non-contact process that allows for the dynamic UV patterning of waveguide material on large substrates, eliminating the need for high resolution masks. A laser direct write tool was designed, assembled, and characterized for direct write patterning waveguides that were comparable in quality to those produced using standard lithographic practices (0.047 dB/cm loss for laser written waveguides compared to 0.043 dB/cm for lithographic waveguides). Straight waveguides, and waveguide turns were patterned at multimode and single mode sizes, and the process was characterized and documented. Support structures such as angled reflectors and vertical posts were produced, showing the versatility of the laser direct write tool. Commercially available components were implanted into the optical layer for out-of-plane routing of the optical signals. These devices featured spherical lenses on the input and output sides of a total internal reflection (TIR) mirror, as well as alignment pins compatible with standard MT design. Fully functional OEPWBs were fabricated featuring input and output out-of-plane optical signal routing with total optical losses not exceeding 10 dB. These prototypes survived thermal cycling (-40°C to 85°C) and humidity exposure (95±4% humidity), showing minimal degradation in optical performance. Operational failure occurred after environmental aging life testing at 110°C for 216 hours.

Sustainable Energy Production in the United States: Life Cycle Assessment of Biofuels and Bioenergy

The United States of America is making great efforts to transform the renewable and abundant biomass resources into cost-competitive, high-performance biofuels, bioproducts, and biopower. This is the key to increase domestic production of transportation fuels and renewable energy, and reduce greenhouse gas and other pollutant emissions. This dissertation focuses specifically on assessing the life cycle environmental impacts of biofuels and bioenergy produced from renewable feedstocks, such as lignocellulosic biomass, renewable oils and fats. The first part of the dissertation presents the life cycle greenhouse gas (GHG) emissions and energy demands of renewable diesel (RD) and hydroprocessed jet fuels (HRJ). The feedstocks include soybean, camelina, field pennycress, jatropha, algae, tallow and etc. Results show that RD and HRJ produced from these feedstocks reduce GHG emissions by over 50% compared to comparably performing petroleum fuels. Fossil energy requirements are also significantly reduced. The second part of this dissertation discusses the life cycle GHG emissions, energy demands and other environmental aspects of pyrolysis oil as well as pyrolysis oil derived biofuels and bioenergy. The feedstocks include waste materials such as sawmill residues, logging residues, sugarcane bagasse and corn stover, and short rotation forestry feedstocks such as hybrid poplar and willow. These LCA results show that as much as 98% GHG emission savings is possible relative to a petroleum heavy fuel oil. Life cycle GHG savings of 77 to 99% were estimated for power generation from pyrolysis oil combustion relative to fossil fuels combustion for electricity, depending on the biomass feedstock and combustion technologies used. Transportation fuels hydroprocessed from pyrolysis oil show over 60% of GHG reductions compared to petroleum gasoline and diesel. The energy required to produce pyrolysis oil and pyrolysis oil derived biofuels and bioelectricity are mainly from renewable biomass, as opposed to fossil energy. Other environmental benefits include human health, ecosystem quality and fossil resources. The third part of the dissertation addresses the direct land use change (dLUC) impact of forest based biofuels and bioenergy. An intensive harvest of aspen in Michigan is investigated to understand the GHG mitigation with biofuels and bioenergy production. The study shows that the intensive harvest of aspen in MI compared to business as usual (BAU) harvesting can produce 18.5 billion gallons of ethanol to blend with gasoline for the transport sector over the next 250 years, or 32.2 billion gallons of bio-oil by the fast pyrolysis process, which can be combusted to generate electricity or upgraded to gasoline and diesel. Intensive harvesting of these forests can result in carbon loss initially in the aspen forest, but eventually accumulates more carbon in the ecosystem, which translates to a CO2 credit from the dLUC impact. Time required for the forest-based biofuels to reach carbon neutrality is approximately 60 years. The last part of the dissertation describes the use of depolymerization model as a tool to understand the kinetic behavior of hemicellulose hydrolysis under dilute acid conditions. Experiments are carried out to measure the concentrations of xylose and xylooligomers during dilute acid hydrolysis of aspen. The experiment data are used to fine tune the parameters of the depolymerization model. The results show that the depolymerization model successfully predicts the xylose monomer profile in the reaction, however, it overestimates the concentrations of xylooligomers.

Off-line and On-line Affective Recognition of a Computer User through A Biosignal Processing Approach

Physiological signals, which are controlled by the autonomic nervous system (ANS), could be used to detect the affective state of computer users and therefore find applications in medicine and engineering. The Pupil Diameter (PD) seems to provide a strong indication of the affective state, as found by previous research, but it has not been investigated fully yet. In this study, new approaches based on monitoring and processing the PD signal for off-line and on-line affective assessment (“relaxation” vs. “stress”) are proposed. Wavelet denoising and Kalman filtering methods are first used to remove abrupt changes in the raw Pupil Diameter (PD) signal. Then three features (PDmean, PDmax and PDWalsh) are extracted from the preprocessed PD signal for the affective state classification. In order to select more relevant and reliable physiological data for further analysis, two types of data selection methods are applied, which are based on the paired t-test and subject self-evaluation, respectively. In addition, five different kinds of the classifiers are implemented on the selected data, which achieve average accuracies up to 86.43% and 87.20%, respectively. Finally, the receiver operating characteristic (ROC) curve is utilized to investigate the discriminating potential of each individual feature by evaluation of the area under the ROC curve, which reaches values above 0.90. For the on-line affective assessment, a hard threshold is implemented first in order to remove the eye blinks from the PD signal and then a moving average window is utilized to obtain the representative value PDr for every one-second time interval of PD. There are three main steps for the on-line affective assessment algorithm, which are preparation, feature-based decision voting and affective determination. The final results show that the accuracies are 72.30% and 73.55% for the data subsets, which were respectively chosen using two types of data selection methods (paired t-test and subject self-evaluation). In order to further analyze the efficiency of affective recognition through the PD signal, the Galvanic Skin Response (GSR) was also monitored and processed. The highest affective assessment classification rate obtained from GSR processing is only 63.57% (based on the off-line processing algorithm). The overall results confirm that the PD signal should be considered as one of the most powerful physiological signals to involve in future automated real-time affective recognition systems, especially for detecting the “relaxation” vs. “stress” states.