• 3-D printed heterogenous substrate bandpass filters

      Nesheiwat, Issa (2021-09)
      With the demand for increasing frequencies in today’s communications systems, compact integrated circuits are challenging to achieve. Compact filters have typically been realized by modifying the circuit design including using LC resonators, defective ground structures, and adjusting the length ratios of resonators. Heterogenous substrates with controlled regions of dielectric loading offer a new design approach when it comes to manufacturing an RF component. In this thesis, additive manufacturing is used to selectively place low-K and high-K dielectric materials to achieve a compact form factor, improved bandwidth, and higher suppression in re-entry modes. First, microstrip coupled strip lines are simulated to model the basic coupling effects of loading a substrate. Next, three 2.45GHz parallel coupled bandpass microstrip filters are designed with differing substrates: low-K, high-K and high-K loaded to analyze the impact of loading within the substrate. The filter substrates are manufactured using a dual-extrusion FDM 3-D printer to combine both dielectrics, low-K ABS, and high-K PrePerm ABS1000, into a single heterogeneous substrate. Compared to the low-K dielectric alternative, the high-K loaded filter demonstrated a 30.8% decrease in length, while maintaining similar bandwidth and suppression of re-entry modes. Compared to the high-K filter, the high-K loaded filter showed a 9.4dB reduction in re-entry mode suppression, while maintaining similar footprint size.
    • Analysis of ground plane size, topography and location on a monopole antenna's performance utilizing 3-D printing

      Ciraco, Vito (2021-09)
      The monopole antenna is widely used in communication applications and is typically mounted on various surfaces that act as ground planes; a prime example being the roof of a car. The shape of the ground plane can drastically change the patterns of the electromagnetic radiation of a monopole antenna as well as its RF performance. Extensive work [1,12-13] has been done on the numerical modeling of arbitrarily shaped ground planes. However, due to their geometric complexity, there is very little work reported on the practical testing component of physical antennas with these obscure ground plane structures. This thesis illustrates how the additive manufacturing process presented can be used to physically realize arbitrarily shaped ground planes and provides a low-cost process to verify the numerical model. Ground Planes were modified while maintaining the same antenna length to evaluate the impact on antenna performance. The antenna was not optimized or changed to a standard antenna design. Varying radius spherical ground planes are modelled, as well as modified ground plane structures to evaluate the impact of the ground plane on a 1.3GHz monopole antenna's performance and in some cases to modify the antenna's performance in terms of gain, bandwidth, and radiation pattern. Designs such as the planar ground with horn was found to enhance monopole bandwidth by more than 5 times that of a standard planar ground but significantly deteriorate the antenna's radiation pattern. Moreover, complex geometry such as the fin sphere ground plane offered a 25% increase in gain relative to the standard sphere ground. Designs like the edge-mounted sphere can offer directive gain and radiation characteristics simply by altering the antennas' location mount location with respect to its ground plane. The techniques presented in this thesis offer new ways of producing 3-D printed ground planes for RF applications that are easier to manufacture, lighter in weight, and can enhance antenna performance over their conventional counterparts.
    • Bypassing fingerprint scanners using artificial fingerprints

      Ford, Kerry C. (2021-05)
      Although fingerprint scanning technology is a convenient and user-friendly method of securing many modern devices, it is not without its flaws. In this paper, a methodology for creating artificial fingerprints is presented, as well as the experimental results, in order to display several low-cost techniques that can be used to bypass modern fingerprint sensors. Three methods are employed: direct collection, indirect collection (mold), and indirect collection (copy). First, using direct collection, a mold and cast of a physical fingerprint is created using very low-cost materials. Second, a fingerprint is indirectly collected from a surface and is used to create a 3D printed mold. Finally, a fingerprint is gathered using the indirect collection method, but is then inverted to achieve a higher resolution 3D printed copy of the original finger. Experimental results are presented, showing the effectiveness of the three fingerprint fabrication techniques on optical and capacitive sensors. Experimental results reveal that it is possible to bypass most sensors 80-100% of the time. The artificial fingerprints produced this way are reusable for many months. This was accomplished using widely available tools, and at a lower cost than that which has been previously reported in other research.
    • Creating a mesh sensor network using Raspberry Pi and XBee radio modules

      Forcella, Michael (2017-05)
      A mesh network is a type of network topology in which one or more nodes are capable of relaying data within the network. The data is relayed by the router nodes, which send the messages via one or more 'hops' until it reaches its intended destination. Mesh networks can be applied in situations where the structure or shape of the network does not permit every node to be within range of its final destination. One such application is that of environmental sensing. When creating a large network of sensors, however, we are often limited by the cost of such sensors. This thesis presents a low-cost mesh network framework, to which any number of different sensors can be attached. The hardware configuration is detailed in such a way that anyone with a modest understanding of technology will be able to reproduce it. The software setup required by the user has also been minimized and clearly documented. Details specific to the user's setup can be entered into a configuration file and the majority of software scripts are scheduled to run automatically via Linux Cron jobs. I conclude by outlining several potential modifications to the framework, including further automation of the software setup, inclusion of additional hardware, and alternate methods for downloading data from the network.
    • The design of high quality factor bifilar archimedean coil geometries for wireless power transfer applications

      Feenaghty, Michael (2019)
      This thesis explores the optimization of a planar coil's geometry for wireless power transfer applications. Wireless power transfer is a popular field of study today due to its wide range of uses in professional and consumer applications. The transfer of data or power without the need for a wired connection allows for the design of increasingly robust and convenient electronic devices. However, wireless power transfer is still limited by poor power transfer efficiency and skew sensitivity under suboptimal conditions. For planar coils, optimal power transfer occurs when the transmitter and receiver coil are very closely spaced, with minimal misalignment between the two coils. This thesis proposes novel planar coil geometries which reduce the sensitivity of the coils to these attributes. The proposed geometries all have the same spatial footprint as the original planar coil to make the proposed changes practical in cases where the available area for the planar coil is limited, such as consumer smartphones. The best coil design exhibits an improvement in power transfer of up to 20% over separation distance, and up to 13% overall with lateral misalignment. The proposed designs enhance the performance of planar coils for wireless power transfer without requiring more board real estate to be dedicated to the coil geometry, maintaining a compact system design.
    • Feasibility of solar panel production using renewables

      Mazzurco, Anthony (2021-12)
      The purpose of this paper is to cover a range of topics related to the current energy issue that we have at hand. It will cover the foundation of our main energy sources, if we have reached peak oil, energy economics in relation to renewable energy, the rate of consumption of energy, other bi-products of oil that we use in everyday life, and the feasibility of producing solar panels from a completely renewable energy power plant. When most people think of oil, they do not consider that it is our main source of energy that drives society. There are other energy sources that we use that include coal, other forms of oil like substances such as biodiesel, ethanol corn, and renewable energy. In the past twenty years, the growth in solar and wind technologies has grown rapidly. In order to use less fossil fuels, there has also been an increase in electric vehicles. The movement towards solar, wind, and electric vehicles may sound like a viable solution, but the embodied energy in these technologies is not emphasized enough on the engineering side. In energy economics there is a term called Energy Returned On Energy Invested (EROEI, or EROI). This field of economics focuses on the amount of energy it takes to produce an energy source, and what that energy output is in relation to production. While looking into the EROEI for the more popular energy alternatives, it can be seen that solar and wind have various values of return. EROI should also be considered with electric vehicles, but there are many other variables to be considered. We are now realizing that peak oil production will be an issue, so alternative energy and transportation technologies are being focused on. One of the issues is if we use certain types of elements for these fossil fuel alternatives, we will eventually exhaust those resources as well. That being said, we should reconsider better alternatives, and reduce wasteful resource industries.
    • Meltdown detection in autistic children combining stress sensors and machine learning

      Singh, Sarah (2022-05)
      Children with autism spectrum disorder face many challenges on a daily basis, including their struggle to communicate their needs, especially in times of distress. This can lead to meltdowns, making it difficult for them to learn, make friends, or have a positive social or educational experience. Existing research detecting meltdowns, specifically using deep learning combined with either facial recognition [1] or a variety of sensors such as heart rate, electrodermal, and temperature sensors [2], have proved successful. However, optimization for practical application utilizing more affordable technology could improve upon the accessibility of these tools for the autistic community, especially working class families. This thesis provides a method to detect and prevent autistic meltdowns inspired by my son, aiming to make a wearable device that can be used whenever and wherever by combining heart rate monitors and electrodermal sensors as a more practical means of detection, as well as a more cost friendly option using low power equipment. The device was built on an STM32-F446RE nucleo board using the kernel based operating system FreeRTOS. A bluetooth android application was created using MIT APP Inventor 2, allowing easy access to sensor data. The device was tested on a child diagnosed with autism by wearing a finger glove with sensors attached during their every day homework routine. A simple logistic regression model was applied to calculate the slope of the sensor's data. The logistic regression model showed promising results with an accuracy score of 0.82 and a recall score of 0.83. This device can be easily modified into a wrist watch interface, making it more comfortable and practical for autistic children to wear. The low cost sensors and processor, combined with a lower cost method of machine learning gives families a better chance at owning a device that could help their child. Meltdown predictions will allow teachers and guardians an opportunity for early intervention and meltdown mitigation.
    • Temperature and energy aware scheduling of heterogeneous processors using machine learning

      Parikh, Harsh (2017-12)
      In the past 20-some years, the entire lifetime of Data Center, the hymn computer engineers and end users have chanted in harmony has been "faster. . .smaller. . . cheaper. . . lower power. . . ," with the most recently added "and lower temperature. . ." significantly complicating the whole scenario. The trade offs among performance, complexity, cost, power and temperature have created exciting challenges and opportunities. All modern data centers face the widespread problem "High performance without trading energy, power and most important temperature". Previous research on scheduling algorithms of processors have focused on static implementation to minimize energy consumption and heat dissipation, but never used Machine Learning to dynamically apply the algorithm. We use Naive Bayesian Classifiers (NBCs) to select the processor combination for the Temperature and Energy Aware Dynamic Level Scheduling algorithm that satisfies a particular user defined condition such as a deadline, energy or temperature budget. Our simulation results exhibit significant energy and temperature savings at a reasonable increase in overall execution time, the learning algorithm selects the desired processors significantly faster than random selection.