No description available.

No description available.

No description available.

In today’s technically advanced society the collection and study of digital images has become an important aspect of various off-line applications that range from medical diagnosis to exploring the Martian surface for traces of water. Various industries have recently started moving towards vision based systems to monitor several of their manufacturing processes. Except for some simple on-line applications, these systems are primarily used to analyze the digital images off-line. This thesis is concerned with developing a more powerful on-line digital image analysis technique which links the fields of traditional digital image processing with a recently devised statistically based image analysis method called multivariate image analysis (MIA). The first part of the thesis introduces the area of traditional digital image processing techniques through a brief literature review of three of its five main classes (image enhancement, restoration, analysis, compression, & synthesis) which contain most of the commonly used operations in this area. This introduction is intended as a starting point for readers who have little background in this field, and as a means of providing sufficient details on these techniques so that they can be used in conjunction with other advanced MIA on-line monitoring operations. MIA of multispectral digital images using latent variable statistical methods (Multi-Way PCA / PLS) is the main topic covered by the second part of this thesis. After reviewing the basic theory of feature extraction using MIA for off-line analyses, a new technique is introduced that extends these ideas for image analyses in on-line applications. Instead of directly using the updated images themselves to monitor a time- varying process, this new technique uses the latent variable space of the image to monitor the increase or decline in the number of pixels belonging to various features of interest. The ability to switch between the images and their latent variable space then allows the user to determine the exact spatial locations of any features of interest. This new method is shown to be ideal for monitoring interesting features from time-varying processes equipped with multispectral sensors. It forms a basis for future on-line industrial process monitoring schemes in those industries that are moving towards automatic vision systems using multispectral digital imagery. / Thesis / Master of Engineering (ME)

No / In many production processes real time information may be obtained from process control computers and other monitoring systems, but most existing scheduling models are unable to use this information to effectively influence scheduling decisions in real time. In this paper we develop a general framework for using real time information to improve scheduling decisions, which allows us to trade off the quality of the revised schedule against the production disturbance which results from changing the planned schedule. We illustrate how our framework can be used to select a strategy for using real time information for a single machine scheduling model and discuss how it may be used to incorporate real time information into scheduling the complex production processes of steel continuous caster planning.

The recent trend of migrating legacy computer systems to a virtualized, cloud-based environment has expanded to real-time systems. Unfortunately, modern hypervisors have no mechanism in place to guarantee the real-time performance of applications running on virtual machines. Past solutions to this problem rely on either spatial or temporal resource partitioning, both of which under-utilize the processing capacity of the host system. Paravirtualized solutions in which the guest communicates its real-time needs have been proposed, but they cannot support legacy operating systems. This thesis demonstrates the shortcomings of resource partitioning using temporally-isolated servers, presents an alternative solution to the scheduling problem called the KairosVM Flattening Scheduling Algorithm, and provides an implementation of the algorithm based on Linux and KVM. The algorithm is analyzed theoretically and an exact schedulability test for the algorithm is derived. Simulations show that the algorithm can schedule more than 90% of all randomly generated tasksets with a utilization less than 0.95. In comparison to the state-of-the-art server based approach, the KairosVM Flattening Scheduling Algorithm is able to schedule more than 20 times more tasksets with utilization of 0.95. Experimental results demonstrate that the Linux-based implementation is able to match the deadline satisfaction ratio of a state-of-the-art server-based approach when the taskset is schedulable using the state-of-the-art approach. When tasksets are unschedulable, the implementation is able to increase the deadline satisfaction ratio of Vanilla KVM by up to 400%. Furthermore, unlike paravirtualized solutions, the implementation supports legacy systems through the use of introspection. / Master of Science

Real-Time systems are defined not only by their logical correctness but also timeliness. Modern real-time systems, such as those controlling industrial plants or the flight controller on UAVs, are no longer isolated. The same computing resources are shared with a variety of other systems and software. Further, these systems are increasingly being connected and made available over the internet with the rise of Internet of Things and the need for automation. Many real-time systems contain sensitive code and data, which not only need to be kept confidential but also need protection against unauthorized access and modification. With the cheap availability of hardware supported Trusted Execution Environments (TEE) in modern day microprocessors, securing sensitive information has become easier and more robust. However, when applied to real-time systems, the overheads of using TEEs make scheduling untenable. However, this issue can be mitigated by judiciously utilizing TEEs and capturing TEE operation peculiarities to create better scheduling policies. This thesis provides a new task model and scheduling approach, Split-TEE task model and a scheduling approach ST-EDF. It also presents simulation results for 2 previously proposed approaches to scheduling TEEs, T-EDF and CT-RM. / Master of Science / Real-Time systems are computing systems that not only maintain the traditional purpose of any computer, i.e, to be logically correct, but also timeliness, i.e, guaranteeing an output in a given amount of time. While, traditionally, real-time systems were isolated to reduce interference which could affect the timeliness, modern real-time systems are being increasingly connected to the internet. Many real-time systems, especially those used for critical applications like industrial control or military equipment, contain sensitive code or data that must not be divulged to a third party or open to modification. In such cases, it is necessary to use methods to safeguard this information, regardless of the extra processing time/resource consumption (overheads) that it may add to the system. Modern hardware support Trusted Execution Environments (TEEs), a cheap, easy and robust mechanism to secure arbitrary pieces of code and data. To effectively use TEEs in a real-time system, the scheduling policy which decides which task to run at a given time instant, must be made aware of TEEs and must be modified to take as much advantage of TEE execution while mitigating the effect of its overheads on the timeliness guarantees of the system. This thesis presents an approach to schedule TEE augmented code and simulation results of two previously proposed approaches.

Full-motion driving simulators require efficient real-time high fidelity vehicle models in order to provide a more realistic vehicle response. Typically, multi-body models are used to represent the vehicle dynamics, but these have the unfortunate drawback of requiring the solution of a set of coupled differential algebraic equations (DAE). DAE's are not conducive to real-time implementation such as in a driving simulator, without a very expensive processing capability. The primary objective of this thesis is to show that multi-body models constructed from DAE's can be reasonably approximated with linear models using suspension elements that have nonlinear constitutive relationships. Three models were compared in this research, an experimental quarter-car test rig, a multi-body dynamics differential algebraic equation model, and a linear model with nonlinear suspension elements. Models constructed from differential algebraic equations are computationally expensive to compute and are difficult to realize for real-time simulations. Instead, a linear model with nonlinear elements was proposed for a more computationally efficient solution that would retain the nonlinearities of the suspension. Simplifications were made to the linear model with nonlinear elements to further reduce computation time for real-time simulation. The development process of each model is fully described in this thesis. Each model was excited with the same input and their outputs were compared. It was found that the linear model with nonlinear elements provides a reasonably good approximation of actual model with the differential algebraic equations. / Master of Science

As embedded software and platforms become more complicated, many safety properties are left to simulation and testing. MRICDF is a formal polychronous language used to guarantee certain safety properties and alleviate the burden of software development and testing. We propose real-time extensions to MRICDF so that temporal properties of embedded systems can also be proven. We adapt the extended precedence encoding technique of Prelude and expand upon current schedulability analysis techniques for multi-periodic real-time systems. / Master of Science

The demand and interest in serving Computer Vision applications at the Edge, where Edge Devices generate vast quantities of data, clashes with the reality that many Devices are largely unable to process their data in real time. While computational offloading, not to the Cloud but to nearby Edge Nodes, offers convenient acceleration for these applications, such systems are not without their constraints. As Edge networks may be unreliable or wireless, offloading quality is sensitive to communication bottlenecks. Unlike seemingly unlimited Cloud resources, an Edge Node, serving multiple clients, may incur delays due to resource contention. This project describes relevant Computer Vision workloads and how an effective offloading framework must adapt to the constraints that impact the Quality of Service yet have not been effectively nor properly addressed by previous literature. We design an offloading controller, based on closed-loop control theory, that enables Devices to maximize their throughput by appropriately offloading under variable conditions. This approach ensures a Device can utilize the maximum available offloading bandwidth. Finally, we constructed a realistic testbed and conducted measurements to demonstrate the superiority of our offloading controller over previous techniques. / Master of Science / Devices like security cameras and some Internet of Things gadgets produce valuable real-time video for AI applications. A field within AI research called Computer Vision aims to use this visual data to compute a variety of useful workloads in a way that mimics the human visual system. However, many workloads, such as classifying objects displayed in a video, have large computational demands, especially when we want to keep up with the frame rate of a real-time video. Unfortunately, these devices, called Edge Devices because they are located far from Cloud datacenters at the edge of the network, are notoriously weak for Computer Vision algorithms, and, if running on a battery, will drain it quickly. In order to keep up, we can offload the computation of these algorithms to nearby servers, but we need to keep in mind that the bandwidth of the network might be variable and that too many clients connected to a single server will overload it. A slow network or an overloaded server will incur delays which slow processing throughput. This project describes relevant Computer Vision workloads and how an effective offloading framework that effectively adapts to these constraints has not yet been addressed by previous literature. We designed an offloading controller that measures feedback from the system and adapts how a Device offloads computation, in order to achieve the best possible throughput despite variable conditions. Finally, we constructed a realistic testbed and conducted measurements to demonstrate the superiority of our offloading controller over previous techniques.