Coordination is important in groupware because it helps users collaborate efficiently. However, groupware systems in which activities occur at a faster pace need faster coordination in order to keep up with the speed of the activity. Faster coordination is especially needed when actions are dependent on one another (i.e., they are tightly-coupled) and when each user can see and interact with other users actions as they occur (i.e., real time). There is little information available about this type of fast coordination (also named high-speed coordination or HSC) in groupware. In this thesis, I addressed this problem by providing a body of principles and information about high-speed coordination. This solution was achieved by creating a groupware game called RTChess and then conducting an exploratory evaluation in which high-speed coordination was investigated. The results of this evaluation show that there were small amounts of high-speed coordination in the game and that high-speed coordination was difficult to achieve. In addition, HSC was affected by five main characteristics of the groupware environment: user experience, level of awareness of the partners interactions, communication between partners, number of dependencies that affect the users interactions, and pace of activities in the system.

Heparan sulfate proteoglycan (HSPG) or heparan sulfate (HS) degradation may contribute to endothelial cell (EC) dysfunction in diabetes. HSPGs, syndecan and perlecan, contain a protein core with mainly HS glycosaminoglycans (GAGs) attached. HSPGs modulate growth factors and function in membrane filtering. Heparanase induction is likely responsible for diabetic HS degradation. Heparin protects endothelium and insulin regulates glucose metabolism. Our objectives were to observe HSPG changes by studying EC GAG content and gene expression of syndecan, perlecan and heparanase under hyperglycemic conditions with insulin and/or heparin treatment.<p> GAGs, including HS, were determined by the carbazole assay and visualized by agarose gel electrophoresis in porcine aortic EC cultures treated with high glucose (30 mM) and/or insulin (0.01 U/ml) for 24, 48 and 72 hours and/or heparin (0.5 µg/ml) for 72 hours. High glucose decreased cell GAGs and increased medium GAGs. GAGs increased with time in control cultures and in high glucose plus insulin treated medium. GAGs were decreased with insulin but increased with insulin or heparin plus high glucose.<p> Confluent cultured human aortic ECs were incubated with control medium, high glucose and/or insulin and/or heparin for 24 hours. Real time PCR determination showed that: high glucose increased heparanase, decreased syndecan and had no effect on perlecan mRNA; insulin or heparin with/without high glucose decreased and insulin and heparin with high glucose increased heparanase mRNA; heparin and insulin with high glucose increased but insulin decreased syndecan mRNA. Actinomycin D (10 µg/ml) inhibited heparanase and syndecan mRNA with high glucose plus insulin plus heparin and inhibited heparanase mRNA with high glucose compared to time 0 but not â-actin after addition for 0, 2, 4, 8 and 24 hours. Bioinformatic studies revealed that transcription factor Sp1 activates heparanase promoter by high glucose and may play a role in regulation of perlecan and syndecan promoters.<p> Insulin or heparin inhibited the reduction in EC GAGs and syndecan mRNA and induction in heparanase by high glucose, indicating their protective effect. Decreased GAGs by insulin may relate to the pathology of hyperinsulinemia. Transcriptional regulation by heparin and/or insulin may cause variation in gene expression of heparanase, syndecan and perlecan.

Plants respond to and defend themselves against a wide range of disease-causing microbes. In order to do so, massive reprogramming of cellular protein expression patterns, which underpin various defense pathways, must occur. A family of basic leucine zipper transcription factors, called TGA factors, has been implicated in mediating this response. The TGA factors themselves are subject to complex regulation; of note, TGA1 and TGA4 are regulated via a reduction of conserved cysteines after treatment with the phenolic signaling molecular salicylic acid, which accumulates following pathogen challenge. Previous studies indicate that TGA factors physically interact in the yeast two-hybrid system with the plant-specific CC-type of glutaredoxin (Grx)-like proteins. Grx are a family of oxidoreductases that are important for maintaining the cellular redox status and often are required to modulate protein activity. The goal of this study was to ascertain the role of these Grx-like proteins in regulating TGA1 redox state. To this end, the expression patterns of several Grx genes were analyzed.<p> Quantitative-reverse-transcriptase PCR (q-RT-PCR) experiments indicated that TGA1 and TGA4 may be involved in down-regulating levels Grx-like gene transcripts after exposure to pathogens or salicylic acid (SA). Furthermore, qRT-PCR experiments also indicated that expression of some Grx-like genes is induced by SA, jasmonic acid (JA), and <i>Pseudomonas syringae</i>. Overexpression of the Grx-like protein, CXXC9, in <i>Arabidopsis thaliana</i> revealed that it is a regulatory factor in the cross-talk between vi theSA/JA pathways as it is able to suppress expression of PDF1.2, a marker for the JA defense pathway, as determined by qRT-PCR. The â-hydroxy ethyl disulfide (HED) assay was utilized to determine if the CC-type of Grx-like proteins have oxidoreductase activity <i>in vitro</i>. These studies revealed that that the Grx-like proteins do not exhibit oxidoreductase activity in this assay.

The most popular methods for solving the shortest-path problem in Artificial Intelligence are heuristic search algorithms. The main contributions of this research are new heuristic search algorithms that are either faster or scale up to larger problems than existing algorithms. Our contributions apply to both online and offline tasks. For online tasks, existing real-time heuristic search algorithms learn better informed heuristic values and in some cases eventually converge to a shortest path by repeatedly executing the action leading to a successor state with a minimum cost-to-goal estimate. In contrast, we claim that real-time heuristic search converges faster to a shortest path when it always selects an action leading to a state with a minimum f-value, where the f-value of a state is an estimate of the cost of a shortest path from start to goal via the state, just like in the offline A* search algorithm. We support this claim by implementing this new non-trivial action-selection rule in FALCONS and by showing empirically that FALCONS significantly reduces the number of actions to convergence of a state-of-the-art real-time search algorithm. For offline tasks, we improve on two existing ways of scaling up best-first search to larger problems. First, it is known that the WA* algorithm (a greedy variant of A*) solves larger problems when it is either diversified (i.e., when it performs expansions in parallel) or committed (i.e., when it chooses the state to expand next among a fixed-size subset of the set of generated but unexpanded states). We claim that WA* solves even larger problems when it is enhanced with both diversity and commitment. We support this claim with our MSC-KWA* algorithm. Second, it is known that breadth-first search solves larger problems when it prunes unpromising states, resulting in the beam search algorithm. We claim that beam search quickly solves even larger problems when it is enhanced with backtracking based on limited discrepancy search. We support this claim with our BULB algorithm. We show that both MSC-KWA* and BULB scale up to larger problems than several state-of-the-art offline search algorithms in three standard benchmark domains. Finally, we present an anytime variant of BULB and apply it to the multiple sequence alignment problem in biology.

The aggressive evolution of the semiconductor industry smaller process geometries, higher densities, and greater chip complexity has provided design engineers the means to create complex, high-performance System-on-a-Chip (SoC) designs. Such SoC designs typically have more than one processor and huge (tens of Mega Bytes) amount of memory, all on the same chip. Dealing with the global on-chip memory allocation/deallocation in a dynamic yet deterministic way is an important issue for upcoming billion transistor multiprocessor SoC designs. To achieve this, we propose a memory management hierarchy we call Two-Level Memory Management. To implement this memory management scheme which presents a shift in the way designers look at on-chip dynamic memory allocation we present the System-on-a-Chip Dynamic Memory Management Unit (SoCDMMU) for allocation of the global on-chip memory, which we refer to as Level Two memory management (Level One is the management of memory allocated to a particular on-chip Processing Element, e.g., an operating systems management of memory allocated to a particular processor). In this way, processing elements (heterogeneous or non-heterogeneous hardware or software) in an SoC can request and be granted portions of the global memory in a fast and deterministic time. A new tool is introduced to generate a custom optimized version of the SoCDMMU hardware. Also, a real-time operating system is modified support the new proposed SoCDMMU. We show an example where shared memory multiprocessor SoC that employs the Two-Level Memory Management and utilizes the SoCDMMU has an overall average speedup in application transition time as well as normal execution time.

Combustion instabilities are a significant problem in combustion systems, particularly in Low NOx Gas Turbine combustors. These instabilities result in large-scale pressure oscillations in the combustor, leading to degraded combustor performance, shortened lifetime, and catastrophic combustor failure. The objective of this research was to develop a practical adaptive active control system that, coupled with an appropriate actuator, is capable of controlling the combustor pressure oscillations without a priori knowledge of the combustor design, operating conditions or instability characteristics. The adaptive controller utilizes an observer that determines the frequencies, phases and amplitudes of the dominant modes of the oscillations in real time. The research included development and testing of the adaptive controller on several combustors and on an unstable acoustic feedback system in order to analyze its performance. The research also included investigations of combustor controllability and combustor stability margin, which are critical issues for practical implementation of an active control system on an industrial combustor. The results of this research are directly applicable to a variety of combustors and can be implemented on full-scale industrial combustion systems.

As control systems are becoming more and more complex, system complexity is rapidly becoming a limiting factor in the efficacy of established techniques for control systems design. To cope with the growing complexity, control architectures often have a hierarchical structure. At the base of the system pyramid lie feedback loops with simple closed-loop control laws. These are followed, at a higher level, by discrete control logics. Such hierarchical systems typically have a hybrid nature. A common approach to addressing these types of complexity consists of decomposing, in the time domain, the control task into a number of modes, i.e. control laws dedicated to carrying out a limited task. This type of control generally involves switching laws among the various modes, and its design poses a major challenge in many application domains. The primary goal of this thesis is to develop a unified framework for addressing this challenge. To this end, the contribution of this thesis is threefold: 1. An algorithmic framework for how to optimize the performance of switched autonomous systems is derived. The optimization concerns both the sequence in which different modes appear in and the duration of each mode. The optimization algorithms are presented together with detailed convergence analyses. 2. Control strategies for how to optimize switched autonomous systems operating in real time, and when the initial state of the system is unknown, are presented. 3. A control strategy for how to optimally navigate an autonomous mobile robot in real-time is presented and evaluated on a mobile robotics platform. The control strategy uses optimal switching surfaces for when to switch between different modes of operations (behaviors).

Because of the shortage of domestic energy resources in Taiwan, more than 97% of the energy has to be imported. The energy price has been increased dramatically during recent years due to the limited supply of conventional primary fossil energy resources. With the economic development and upgrade of people living standard, the electricity power consumption is increased significantly. To solve the problem, different strategies of energy conservation and CO2 emission reduction have been promoted by government to reduce that the peak loading growth and achieve better usage of electricity with more effective load management. This thesis proposes a digital smart meter which integrates the energy metering IC, microprocessor and hybrid communication schemes (Power Line Carrier/ZigBee/RS-485). The load control module and communication module are included in the smart meter to support various application functions. The embedded power management system (PMS) is also proposed to integrate with the smart meter to perform the demand response according to the real-time pricing and load management for residential and commercial customers. The master station can supervise the real-time power consumption of various load components to analyze the power consumption model of customers served and execute the demand load control. The actual demonstration system of embedded PMS has been set up to verify the function of energy management so that the customers have better understanding of power consumption by each appliance. In the future, the implementation of intelligent load control with an emergency load shedding of capability can help utility companies to achieve virtual power generation to enhance the power systems reliability. The customers may also reduce the electricity charge by executing demand response function, which disconnects the electricity service for non essential loads for either system emergency or high electricity peak pricing

Embedded in circuit emulator (EICE) is the most common and widely used debugging techniques for microprocessors. Because the ICE is capable to provide diverse debugging and testing mechanisms, such as: single-step debugging, breakpoints setting and detection, monitoring, and modification of internal resources. However, the shortcoming of the conventional embedded in circuit emulator (EICE) is that the operation of the processor has to be suspended during debugging, which is categorized as static debugging (Static Debug) and is infeasible for real-time debugging. Therefore, this paper proposes a design alternative to support the real-time system debugging without suspending the microprocessor via the debug hardware Coprocessor14 (the Debug Coprocessor). In this paper, the embedded in circuit emulator is combined with Coprocessor 14 to provide both the static debugging and Run-time system debugging. After incorporating CP14 with the debugging mechanism, the control of the debug hardware is no longer limited to use the IEEE 1149.1 test port during debugging. On the other hand, the set of debugging constraints and the observation of the internal state of the microprocessor can be achieved by inserting the Coprocessor instruction at the program level.

¡@¡@Due to wireless communication technologies, positioning technologies, and mobile computing develop quickly, mobile services are becoming practical and important on big spatiotemporal databases management. Mobile service users move only inside a spatial space, e:g: a country. They often issue the K Nearest Neighbor (kNN) query to obtain data objects reachable through the spatial database. The challenge problem of mobile services is how to efficiently answer the data objects which users interest to the corresponding mobile users. One type of kNN query problems is the order-sensitive moving kNN (order-sensitive MkNN) query problem. In the order-sensitive MkNN query problem, the query point is dynamic and unpredictable, the kNN answers should be responded in real time and sorted by the distance in the ascending order. Therefore, how to respond the kNN answers effectively, incrementally and correctly is an important issue. Nutanong et al: have proposed the V*-kNN algorithm to process the order-sensitive MkNN query. The V*-kNN algorithm uses their the V*-diagram algorithm to generate the safe region. It also uses the Incremental Rank Updates algorithm (IRU) to handle the events while the query point passing the bisectors or the boundary of the safe region. However, the V*-kNN algorithm uses the BF-kNN algorithm to retrieve NNs, which is non-incremental. This makes the search time increase while the density of the object increases. Moreover, they do not consider the situation that there are multiple objects at the same order, and the situation that there are multiple events happen in a single step. These situations may cause that the kNN answers are incorrect. Therefore, in this thesis, we propose the Hilbert curve-based kNN algorithm (HC-kNN) algorithm to process the ordersensitive MkNN query. The HC-kNN algorithm can handle the situation that there are multiple events happen in a single step. We also propose new data structure of the kNN answers. Next, we propose the Intersection of Perpendicular Bisectors algorithm (IPB) in order to handle order update events of the kNN answers. The IPB algorithm handles the situation which there are multiple objects at the same order. Finally, based on the Hilbert curve index, we propose the ONHC-kNN algorithm to get NNs incrementally and to generate the safe region. The safe region will not be affected while the density of the object increases. The safe region of our algorithm is larger than that of the V*-kNN algorithm. From our simulation result, we show that the HC-kNN algorithm provides better performance than the V*-kNN algorithm.