Systems and Communications Commons^™

Meltpondnet: A Swin Transformer U-Net For Detection Of Melt Ponds On Arctic Sea Ice, Ivan Sudakow, Vijayan K. Asari, Ruixu Liu, Denis Demchev

Electrical and Computer Engineering Faculty Publications

High-resolution aerial photographs of Arctic region are a great source for different sea ice feature recognition, which are crucial to validate, tune, and improve climate models. Melt ponds on the surface of melting Arctic sea ice are of particular interest as they are sensitive and valuable indicators and are proxy to the processes in the Arctic climate system. Manual analysis of this remote sensing data is extremely difficult and time-consuming due to the complex shapes and unpredictable boundaries of the melt ponds, and that leads to the necessity for automatizing the processes. In this study, we propose a robust and …

Circuit Optimization Techniques For Efficient Ex-Situ Training Of Robust Memristor Based Liquid State Machine, Alex Henderson, Christopher Yakopcic, Cory Merkel, Steven Harbour, Tarek M. Taha, Hananel Hazan

Electrical and Computer Engineering Faculty Publications

Spiking neural network hardware offers a high performance, power-efficient and robust platform for the processing of complex data. Many of these systems require supervised learning, which poses a challenge when using gradient-based algorithms due to the discontinuous properties of SNNs. Memristor based hardware can offer gains in portability, power reduction, and throughput efficiency when compared to pure CMOS. This paper proposes a memristor-based spiking liquid state machine (LSM). The inherent dynamics of the LSM permit the use of supervised learning without backpropagation for weight updates. To carry out the design space evaluation of the LSM for optimal hardware performance, several …

Detecting The Presence Of Electronic Devices In Smart Homes Using Harmonic Radar, Beatrice Perez, Gregory Mazzaro, Timothy J. Pierson, David Kotz

Dartmouth Scholarship

Data about users is collected constantly by phones, cameras, Internet websites, and others. The advent of so-called ‘Smart Things' now enable ever-more sensitive data to be collected inside that most private of spaces: the home. The first step in helping users regain control of their information (inside their home) is to alert them to the presence of potentially unwanted electronics. In this paper, we present a system that could help homeowners (or home dwellers) find electronic devices in their living space. Specifically, we demonstrate the use of harmonic radars (sometimes called nonlinear junction detectors), which have also been used in …

Multimodal Adversarial Learning, Uche Osahor

Graduate Theses, Dissertations, and Problem Reports (ETD)

Deep Convolutional Neural Networks (DCNN) have proven to be an exceptional tool for object recognition, generative modelling, and multi-modal learning in various computer vision applications. However, recent findings have shown that such state-of-the-art models can be easily deceived by inserting slight imperceptible perturbations to key pixels in the input. A good target detection systems can accurately identify targets by localizing their coordinates on the input image of interest. This is ideally achieved by labeling each pixel in an image as a background or a potential target pixel. However, prior research still confirms that such state of the art targets models …

Autonomous Wildfire Detection System, Matthew Pyle, Nick Garrett

Williams Honors College, Honors Research Projects

According to the National Fire Protection Association Journal, federal wildfire suppression costs in the United States have risen from an annual average of about $425 million from 1985 to 1999 up to $1.6 billion from 2000 to 2019. On average, more than 200,000 acres in the United States are burned per year due to wildfires, with more than 700,000 acres burned in 2020 alone. With the risk of wildfire ever rising, there is a need for better early detection of remote wildfires, as existing methods often include long delays like satellites or rely on human lookout towers. The objective is …

Smart Uv-C Disinfectant Module, Nicole Baldy, Luke Rogers, Haitham Saleh

Williams Honors College, Honors Research Projects

The Smart UV Disinfectant device shall sanitize objects which are 18”x14”x8” or smaller and less than 20 lbs. using UV-C light. This device should contain many safety measures to prevent human and animal exposure to the UV-C light and have no public touchpoints to operate the interface. In order to achieve the first objective, this device shall contain a "sanitizing chamber" which completely encloses the object to be sanitized to prevent outside exposure with detection of any lifeforms inside of the chamber; for the second objective, it will contain a wireless interface to an Android application which can be used …

Security Concerns On Machine Learning Solutions For 6g Networks In Mmwave Beam Prediction, Ferhat Ozgur Catak, Murat Kuzlu, Evren Catak, Umit Cali, Devrim Unal

Engineering Technology Faculty Publications

6G – sixth generation – is the latest cellular technology currently under development for wireless communication systems. In recent years, machine learning (ML) algorithms have been applied widely in various fields, such as healthcare, transportation, energy, autonomous cars, and many more. Those algorithms have also been used in communication technologies to improve the system performance in terms of frequency spectrum usage, latency, and security. With the rapid developments of ML techniques, especially deep learning (DL), it is critical to consider the security concern when applying the algorithms. While ML algorithms offer significant advantages for 6G networks, security concerns on artificial …

Energy-Efficient Dynamic Directional Modulation With Electrically Small Antennas, Adam Narbudowicz, Abel Zandamela, Nicola Marchetti, Max Ammann

Articles

This letter proposes a compact and energy-efficient directional modulation scheme for small and power constrained devices (e.g., Internet of Things wireless sensors). The scheme uses a uniform circular array of monopole antennas, with a single radiofrequency chain and antenna active at a time. The antennas can be located in close proximity, offering significant size reduction. Furthermore, the schemewas demonstrated to operate successfully with an array of 0.6 λ diameter. The system does not generate additional artificial noise, limiting interference to other systems. Finally, the antenna switching sequence can be randomly generated and without fine-tuned synchronization with the transmitter.

Leveraging Machine Learning For Detecting Iot-Based Interference In Operational Wifi Networks, Josh Pulse

Honors Program Theses

IoT (Internet of Things) devices have become increasingly popular in recent years while WiFi continues to serve as primary network provider indoors. With the advancements in technology, the networks of IoT devices continue to weave closely with indoor WiFi network deployments. Both kinds of these networks primarily operate in 2.4 GHz ISM Band (though latest WiFi standards can operate in 5 GHz and 60 GHz bands, too). With the multitude of tiny IoT sensors being deployed indoors alongside operational WiFi networks, severe interference scenarios cannot be ruled out. As a result of this interference, performance of WiFi networks is bound …

Lapnitor: A Web Service That Protects Your Laptop From Theft., Michael Ameteku

Williams Honors College, Honors Research Projects

Laptop theft is an issue worldwide. According to an article from 2018, Security Boulevard stated that a laptop is stolen every 53 seconds. Using a laptop's camera, we can monitor the surroundings of the laptop and reduce a laptop's probability of being stolen. According to the University of Pittsburgh, a laptop has a 1-in- 10 chance of being stolen and nearly half of these thefts occur in offices or classrooms. These thefts mostly occur when a laptop owner leaves their device unattended for a certain period of time to maybe go visit the restroom or attend to a call when …

Smart Kettlebell, Cole Russell, Patrick Davis, Ryan Deboer, Isaac Stecker

Williams Honors College, Honors Research Projects

The purpose of design team 08’s senior design project is to create and implement a workout device that will be able to calculate the caloric expenditure, repetitions, and length of a workout. Once the duration of a workout is completed, the user metrics will be communicated from the kettlebell system to a user application via Bluetooth. The caloric expenditure will be used using the data that is provided from the accelerometer. The duration of the workout will be controlled using a push button that is connected to the kettlebell system that will allow for the start and stop of a …

Effect Of Freestream Noise On Hypersonic Crossflow-Induced Boundary-Layer Transition, Andrew Bustard, Thomas J. Juliano, Harrison B. Yates, Mark Noftz, Joseph Jewell

Publications

Between the windward and leeward rays, boundary-layer transition on cones at angle of attack in hypersonic flow is dominated by the crossflow instability [1, 2], wherein stationary and/or traveling crossflow vortices develop and can breakdown into turbulence [3]. Stationary crossflow modes are caused by surface roughness or any other steady forcing and produce disturbances that are fixed in place on the surface [3–6]. Stationary vortices create a generalized inflection point in the velocity profile that is inviscidly unstable [7]. This distorted mean flow develops secondary instabilities, whose growth and breakdown is one path to turbulence [8–10]. Traveling crossflow modes are …

Removing Physical Presence Requirements For A Remote And Automated World - Api Controlled Patch Panel For Conformance Testing, Hunter George Wells

Honors Theses and Capstones

Quality assurance test engineers at the UNH-InterOperability Lab must run tests that require driving and monitoring a selection of DC signals. While the number of signals is numerous, there are limited ports on the test equipment, and only a few signals need patching for any given test. The selection of signals may vary between the 209 different tests and must be re-routed frequently. Currently, testers must leave their desk to manually modify the test setup in another room. This posed a considerable issue at the onset of the COVID-19 Pandemic when physical access was not possible. In order to enable …

Design And Performance Analysis Of A Chemically-Etched Flexible Nfc Tag Antenna, Muhammad Enayetur Rahman, Marwan Abdelatti, Manbir Singh Sodhi, Kunal Mankodiya

Electrical & Computer Engineering Faculty Publications

Near Field Communication (NFC) is a perfect example of ubiquitous computing that is secured, short-ranged, low-powered contactless communication. High demand is predicted for NFC, especially with wearables, and the variety of applications may require that this technology be fabricated onto different materials. In this research, we first designed flexible NFC antennas based on mathematical models. A simulation of the NFC coil inductance was performed and verified using the predictive models. Later, two flexible NFC antennas, including 160x80 mm2 (rectangular) and 80x80 mm2 (square) were fabricated through a chemical etching process and verified at 13.56 MHz frequency. Two experiments 1) Read …

A Primer On Software Defined Radios, Dimitrie C. Popescu, Rolland Vida

Electrical & Computer Engineering Faculty Publications

The commercial success of cellular phone systems during the late 1980s and early 1990 years heralded the wireless revolution that became apparent at the turn of the 21st century and has led the modern society to a highly interconnected world where ubiquitous connectivity and mobility are enabled by powerful wireless terminals. Software defined radio (SDR) technology has played a major role in accelerating the pace at which wireless capabilities have advanced, in particular over the past 15 years, and SDRs are now at the core of modern wireless communication systems. In this paper we give an overview of SDRs that …

Stress Classification Using Deep Learning With 1d Convolutional Neural Networks, Abdulrazak Yahya Saleh, Lau Khai Xian

Knowledge Engineering and Data Science

Stress has been a major problem impacting people in various ways, and it gets serious every day. Identifying whether someone is suffering from stress is crucial before it becomes a severe illness. Artificial Intelligence (AI) interprets external data, learns from such data, and uses the learning to achieve specific goals and tasks. Deep Learning (DL) has created an impact in the field of Artificial Intelligence as it can perform tasks with high accuracy. Therefore, the primary purpose of this paper is to evaluate the performance of 1D Convolutional Neural Networks (1D CNNs) for stress classification. A Psychophysiological stress (PS) dataset …

Machine Learning Based Critical Resource Allocation In Mixed-Traffic Cellular Networks, Mohamed Nomeir

Theses and Dissertations

The proliferation of cellular networks over the past two decades has encouraged the expansion of their use in many modern applications. These applications involve the use of data traffic of different quality of service (QoS) requirements. Some of these requirements are quite stringent such as in the case of critical Internet of Things (IoT) health care, military and homeland security applications. This situation resulted in imposing a variety of resource allocation requirements on the cellular network operation in a simultaneous manner.

In this thesis, we consider the challenging problem of mixed-traffic resource allocation, or scheduling, in cellular networks. We focus …

Evaluating Deep-Learning Models For Debris-Covered Glacier Mapping, Zhiyuan Xie, Vijayan K. Asari, Umesh K. Haritashya

Electrical and Computer Engineering Faculty Publications

In recent decades, mountain glaciers have experienced the impact of climate change in the form of accelerated glacier retreat and other glacier-related hazards such as mass wasting and glacier lake outburst floods. Since there are wide-ranging societal consequences of glacier retreat and hazards, monitoring these glaciers as accurately and repeatedly as possible is important. However, the accurate glacier boundary, especially the debriscovered glacier (DCG) boundary, which is one of the primary inputs in many glacier analyses, remains a challenge even after many years of research using conventional remote sensing methods or machine-learning methods. The GlacierNet, a deep-learning-based approach, utilized the …

Parallel Approach Of Adaptive Image Thresholding Algorithm On Gpu, Adhi Prahara, Andri Pranolo, Nuril Anwar, Yingchi Mao

Knowledge Engineering and Data Science

Image thresholding is used to segment an image into background and foreground using a given threshold. The threshold can be generated using a specific algorithm instead of a pre-defined value obtained from observation or experiment. However, the algorithm involves per pixel operation, histogram calculation, and iterative procedure to search the optimum threshold that is costly for high-resolution images. In this research, parallel implementations on GPU for three adaptive image thresholding methods, namely Otsu, ISODATA, and minimum cross-entropy, were proposed to optimize their computational times to deal with high-resolution images. The approach involves parallel reduction and parallel prefix sum (scan) techniques …

A Comprehensive Analysis Of Reward Function For Adaptive Traffic Signal Control, Abu Rafe Md Jamil, Naushin Nower

Knowledge Engineering and Data Science

Adaptive traffic control systems (ATCS) can play an essential role in reducing traffic congestion in urban areas. The main challenge for ATSC is to determine the proper signal timing. Recently, Deep Reinforcement Learning (DRL) has been used to determine proper signal timing. However, the success of the DRL algorithm depends on the appropriate reward function design. There exist various reward functions for ATSC in the existing research. This research presents a comprehensive analysis of the widely used reward function. The pros and cons of various reward algorithms were discussed, and experimental analysis shows that the multi-objective reward function enhances the …