Search

Implantable medical devices (IMDs) are medically necessary devices embedded in a human body that monitor chronic disorders or automatically deliver therapies, such as insulin pumps or pacemakers. Typically, they are small form-factor devices with limited battery and processing power. Most IMDs have wireless capabilities that allow them to share data with an offboard programming device, such as a smartphone application, that has more storage and processing power than the IMD itself. Additionally, the programming device can send commands back to the IMD to change its settings according to the treatment plan. As such, wirelessly sharing information between an IMD and offboard device can help medical providers monitor the patient's health remotely while giving the patient more insight into their condition, more autonomy, and fewer in-person appointments. However, serious security concerns have arisen as researchers have demonstrated it is possible to hack these devices to obtain sensitive information or potentially harm the patient. This is particularly easy to do as most IMDs transmit their data in the clear to avoid allocating their limited resources to encrypting their packets. As these concerns and the percentage of the American population with IMDs grows, there is another fear that bad actors could exploit the link between the programming device and IMD. Theoretically, a hacker could launch a man in the middle attack to send the IMD unauthorized commands, reprogramming it to act as a radio, sniffing signals of interest in the environment. As such, the hacker could use the IMD as a software defined radio (SDR) that captures sensitive or even classified information without the patient's knowledge. If this were to happen, it is possible an unwitting person with an IMD who has access to classified or sensitive information could be used to exfiltrate data that, in the wrong hands, could be used for corporate espionage or to the detriment of national security. While governing bodies agree that cybersecurity risks are present in IMD systems, there are no requirements for IMD manufacturers to create their devices with security measures that mitigate these risks. Researchers have proposed physical, technical, and administrative security measures for IMDs, but other existing wireless security techniques may apply to the healthcare space and need to be explored. Beamforming is an array signal processing technique that relies on individual elements of antenna arrays adjusting their phase and amplitude to create an overall effect of directing RF energy in a particular direction. Similarly, cooperative beamforming uses several physically separate "friendly" beamforming-capable devices to collectively send artificial noise to eavesdroppers while ensuring the signal is successfully received by the intended receiver. Although there are several cooperative jamming algorithms, they share the underlying principles of minimizing SINR at potential eavesdroppers while maximizing the SINR at the intended receiver. Researchers exploring cooperative jamming have largely used models to estimate its impact on channel secrecy. While RF propagation and communication system modeling provides valuable insight into system performance, many theoretical and empirical models are limited by the extent to which the operational environment matches that of the model itself. Ray tracing, alternatively, is more widely applicable as it accounts for a 3D environment and the objects a signal interacts with in that space. A ray is defined as an individual RF signal that travels in a straight line through a uniform medium; obeys the laws of reflection, refraction, and diffraction; and carries energy. As the ray interacts with objects in the environment, its energy will decrease by some amount that depends on the materials and geometry of the object. While research has predominantly focused on applications like cellular communications, the same principles of minimizing SINR at potential eavesdroppers while maximizing the SINR at the intended receiver can be applied to IMDs. As IMD use cases assume the programmer is nearby, the friendly nodes will not need to act as relays and can instead focus all their power on jamming. The number of cooperative jammers will be low to simulate the number of devices an individual might have in a workspace or office setting, like a personal phone, smart watch, or laptop, and realistic power constraints will be observed. Further, ray tracing software will provide additional visual insights into how various building materials like drywall, concrete, brick, and glass impact cooperative jamming. Through these simulations, the trade-off between secrecy rate and physical separation and layout of friendly nodes can be determined, which in turn may inform how companies or individuals can protect their proprietary and personal information. Master of Science Implantable medical devices (IMDs) are medically necessary devices embedded in a human body that monitor chronic disorders or automatically deliver therapies, such as insulin pumps or pacemakers. The data on IMDs need to be processed and their settings might need to be adjusted, but IMDs themselves usually cannot support direct user input, such as through screens or buttons, as they are inaccessible without surgery or generally too small to have space for displays. Further, they lack processing power and battery life due to their small form-factors, so relatively little data remains onboard. Instead, it is more convenient for the IMDs to wirelessly send their data to a more powerful external device like a smartphone. Since smartphones have more battery and processing resources available, and are easily recharged, they can store more data, monitor trends in the patient's health records, and upload the data to a server which the doctors can access. Additionally, these devices can send commands back to the IMD to change its settings according to the treatment plan. As such, wirelessly sharing information between an IMD and offboard programming device can help medical providers monitor the patient's health remotely while giving the patient more insight into their condition, more autonomy, and fewer in-person appointments. However, serious security concerns have arisen as researchers have demonstrated it is possible to hack these devices to obtain sensitive information or potentially harm the patient. As these concerns and the percentage of the American population with IMDs grows, there is another fear that bad actors could exploit the link between the programming device and IMD. Theoretically, a hacker could send the IMD unauthorized commands that change the IMD's behavior so that they are reprogrammed to act as radios listening for signals in the environment in order to steal sensitive or even classified information. While governing bodies agree that cybersecurity risks are present in IMD systems, there are no requirements for IMD manufacturers to create their devices with security measures that mitigate these risks. Researchers have proposed physical, technical, and administrative security measures for IMDs, but other existing wireless security techniques may apply to the healthcare space and need to be explored. Cooperative jamming is an existing defensive wireless technique that reduces the likelihood of an eavesdropper gaining access to unauthorized information. A known set of "friendly" transmitters each transmit noise to eavesdroppers while ensuring the signal is successfully received by the intended receiver. Researchers exploring cooperative jamming have largely used models to estimate its impact on channel secrecy. While RF propagation and communication system modeling provides valuable insight into system performance, many theoretical and empirical models are limited by the extent to which the operational environment matches that of the model itself. Ray tracing, alternatively, is more widely applicable as it accounts for a 3D environment and the objects a signal interacts with in that space. A ray is defined as an individual RF signal that travels in a straight line through a uniform medium; obeys the laws of reflection, refraction, and diffraction; and carries energy. As the ray interacts with objects in the environment, its energy will decrease by some amount that depends on the materials and geometry of the object. Thus, using ray tracing to model cooperative jamming will provide new insights into the degree to which cooperative jamming could be used to protect an IMD from eavesdroppers, and how companies or individuals can protect their proprietary and personal information.

Driver Assistance and Autonomous Driving features are becoming nearly ubiquitous in new vehicles. The intent of the Driver Assistant features is to assist the driver in making safer decisions. The intent of Autonomous Driving features is to execute vehicle maneuvers, without human intervention, in a safe manner. The overall goal of Driver Assistance and Autonomous Driving features is to reduce accidents, injuries, and deaths with a comforting driving experience. However, different drivers can react differently to advanced automated driving technology. It is therefore important to consider and improve the adaptability of these advances based on driver behavior. In this thesis, a human-centric approach is adopted in order to provide an enriching driving experience. The thesis investigates the natural behavior of drivers when changing lanes in terms of preferences of vehicle kinematics parameters using a real-world driving dataset collected as part of the Second Strategic Highway Research Program (SHRP2). The SHRP2 Naturalistic Driving Study (NDS) set is mined for lane change events. This work develops a way to detect reliable lane changing instances from a huge NDS dataset with more than 5,400,000 data files. The lane changing instances are distinguished from noisy and erroneous data by using machine vision lane tracking system variables such as left lane marker probability and right lane marker probability. We have shown that detected lane changing instances can be validated using only vehicle kinematics data. Kinematic vehicle parameters such as vehicle speed, lateral displacement, lateral acceleration, steering wheel angle, and lane change duration are then extracted and examined from time series data to characterize these lane-changing instances for a given driver. We have shown how these vehicle kinematic parameters change and exhibit patterns during lane change maneuvers for a specific driver. The thesis shows the limitations of analyzing vehicle kinematic parameters separately and develops a novel metric, Lane Change Dynamic Score(LCDS) that shows the collective effect of these vehicle kinematic parameters. LCDS is used to classify each lane change and thereby different driving styles. Master of Science The current tendency of car manufacturers is to create vehicles that will offer the user the most comfortable ride possible. The user experience is given a lot of attention to ensure it is up to par. With technological advancements, we are moving closer to an era in which automobiles perform many functions autonomously. However, different drivers may react differently to highly automated driving technologies. Therefore, adapting to different driving styles is critical to increasing the acceptance of autonomous vehicle features. In this work, we examine one of the stressful maneuvers of lane changes. The analysis of various drivers' lane-changing behaviors and the value of personalization are the main subjects of this study based on actual driving scenarios. To achieve this, we have provided an algorithm to identify occurrences of lane-changing from real driving trip data files. Following that, we investigated parameters such as lane change duration, vehicle speed, displacement, acceleration, and steering wheel angle when changing lanes. We have demonstrated the patterns and changes in these vehicle kinematic characteristics that occur when a particular driver performs lane change operations. The thesis shows the limitations of analyzing vehicle kinematic parameters separately and develops a novel metric, Lane Change Dynamic Score(LCDS) that shows the collective effect of these vehicle kinematic parameters. LCDS is used to classify each lane change and thereby different driving styles.

As Moore's law nears its limit, a new form of signal processing is needed. Neuromorphic computing has used inspiration from biology to produce a new form of signal processing by mimicking biological neural networks using electrical components. Neuromorphic computing requires less signal preprocessing than digital systems since it can encode signals directly using analog temporal encoders from Spiking Neural Networks (SNNs). These encoders receive an analog signal as an input and generate a spike or spike trains as their output. The proposed temporal encoders use latency and Inter-Spike Interval (ISI) encoding and are expected to produce a highly sensitive hardware implementation of time encoding to preprocess signals for dynamic neural processors. Two ISI and two latency encoders were designed using Integrate-and-Fire (IF) and Leaky Integrate-and-Fire (LIF) neurons and optimized to produce low area designs. The IF and LIF neurons were designed using the Global Foundries 180nm CMOS process and achieved an area of 186µm2 and 182µm2, respectively. All four encoders have a sampling frequency of 50kHz. The latency encoders achieved an average energy consumption per spike of 277nJ and 316pJ for the IF-based and LIF-based latency encoders, respectively. The ISI encoders achieved an average energy consumption per spike of 1.07uJ and 901nJ for the IF-based and LIF-based ISI encoders, respectively. Power consumption is proportional to the number of neurons employed in the encoder and the potential to reduce power consumption through layout-level simulations is presented. The LIF neuron is able to use a smaller membrane capacitance to achieve similar operability as the IF neuron and consumes less area despite having more components. This demonstrates that capacitor sizes are the main limitations of a small size in spiking neurons for SNNs. An overview of the design and layout process of the two presented neurons is discussed with tips for overcoming problems encountered. The proposed designs can result in a fast neuromorphic process by employing a frequency higher than 10kHz and by providing a hardware implementation that is efficient in multiple sectors like machine learning, medical implementations, or security systems since hardware is safer from hacks. Master of Science As Moore's law nears its limit, a new form of signal processing is needed. Moore's law anticipated that transistor sizes will decrease exponentially as the years pass but CMOS technology is reaching physical limitations which could mean an end to Moore's prediction. Neuromorphic computing has used inspiration from biology to produce a new form of signal processing by mimicking biological neural networks using electrical components. Biological neural networks communicate through interconnected neurons that transmit signals through synapses. Neuromorphic computing uses a subdivision of Artificial Neural Networks (ANNs) called Spiking Neural Networks (SNNs) to encode input signals into voltage spikes to mimic biological neurons. Neuromorphic computing reduces the preprocessing step needed to process data in the digital domain since it can encode signals directly using analog temporal encoders from SNNs. These encoders receive an analog signal as an input and generate a spike or spike trains as their output. The proposed temporal encoders use latency and Inter-Spike Interval (ISI) encoding and are expected to produce a highly sensitive hardware implementation of time encoding to preprocess signals for dynamic neural processors. Two ISI and two latency encoders were designed using Integrate-and-Fire (IF) and Leaky Integrate-and-Fire (LIF) neurons and optimized to produce low area designs. All four encoders have a sampling frequency of 50kHz. The latency encoders achieved an average energy consumption per spike of 277nJ and 316pJ for the IF-based and LIF-based latency encoders, respectively. The ISI encoders achieved an average energy consumption per spike of 1.07uJ and 901nJ for the IF-based and LIF-based ISI encoders, respectively. Power consumption is proportional to the number of neurons employed in the encoder and the potential to reduce power consumption through layout-level simulations is presented. The LIF neuron is able to use a smaller membrane capacitance to achieve similar operability which consumes less area despite having more components than the IF neuron. This demonstrates that capacitor sizes are the main limitations of small size in neurons for spiking neural networks. An overview of the design and layout process of the two presented neurons is discussed with tips for overcoming problems encountered. The proposed designs can result in a fast neuromorphic process by employing a frequency higher than 10kHz and by providing a hardware implementation that is efficient in multiple sectors like machine learning, medical implementations, or security systems since hardware is safer from hacks.

The emergence of unmanned aerial vehicles (UAVs) along with the implementation of 5G networks offers exciting opportunities in expanding wireless capabilities. Not only is improved wireless performance expected with traditional devices such as mobile phones, but new use cases such as the internet-of-things and autonomous vehicle operation will rely on 5G and future network generations. In such widespread applications, from transportation to vital business operation, reliable and often guaranteed connectivity is required for safety and commercial approval. Introducing UAVs into network processes has been explored and implemented in certain instances to take advantage of the flexibility drone devices offer in their mobility and control to address these evolving network possibilities. While practical UAV deployment in certain network cases has been demonstrated, including coverage restoration in disaster relief scenarios, more ambitious goals of 5G will have additional considerations. This includes autonomous vehicles (AVs) whose operation is defined by levels representing varying degrees of autonomy. With computational requirements exponentially increasing as a vehicle's autonomy level is increased, 5G is expected to play an integral role in offloading certain vehicle tasks to the cloud. This thesis then proposes UAV based measurement opportunities as a method to characterize 5G coverage as part of autonomous vehicle processes to identify the proper level of autonomy that can operate safely given the current RF environment. This thesis proposes an UAV based measurement system that would provide coverage verification employing a platform capable of precise RF measurements and enhanced spatial sampling of the environment. Methods employed to traditionally characterize available coverage, including cellular drive tests, do not result in accurate enough measurements for AV use cases. Where lack of coverage in common network processes and use cases can result in dropped calls and poor connectivity in mobile devices, autonomous systems proposed in evolving network generations that deal with safety and mission critical functions must have guaranteed and verified coverage. Data produced in this thesis demonstrates that the proposed UAV based measurement system will improve measurement accuracy and enhanced geographic performance over conventional automotive vehicle based measurement systems Master of Science Wireless networks have grown to support vital and everyday processes in modern society. The COVID-19 pandemic proved wireless communication means a necessity to limit daily disruptions, but networks had already been supporting a continuously increasing amount of mobile devices prior to this. Other demonstrations of wireless network capacity include the release of 5G technology, allowing improved performance with traditional devices like smartphones, along with additional use cases this technology enables including the internet-of-things (IoT) and artificial intelligence (AI) leveraged functions for commercial applications. While wireless network capabilities have demonstrated their success in supporting and maintaining some critical functions, it is important to continually look ahead and plan for future network implementations in order to develop and support all desired advancements. Current measurement methods that assist in verifying coverage for current use cases like mobile devices will fall short in verification for more stringent requirements characteristic of AV and other ambitious network goals. The results found in this work then support the need for continuing research of a UAV-leveraged platform in the scope of eventual practical and safe AV integration into society. The focus of this thesis is to then propose and provide initial evaluation of a UAV-leveraged measurement platform to verify the operability of autonomous vehicles (AVs), which are expected to be a major aspect of future network processes. The computational requirements to operate an autonomous vehicle exponentially increase as a vehicle's autonomy level is increased. 5G is then expected to play an integral role in offloading certain vehicle tasks to the cloud. This thesis paper then proposes UAV based measurement opportunities as a method to characterize 5G coverage as part of autonomous vehicle processes to identify the proper level of autonomy that can operate safely given the current RF environment.

With the increasing number of wireless devices and the phenomenal amount of data that is being generated by them, there is a growing interest in the wireless communications community to complement the traditional model-driven design approaches with data-driven machine learning (ML)-based solutions. However, managing the large-scale multi-dimensional data to maintain the efficiency and scalability of the ML algorithms has obviously been a challenge. Tensors provide a useful framework to represent multi-dimensional data in an integrated manner by preserving relationships in data across different dimensions. This thesis studies two new applications of tensors to ML for wireless communications where the tensor structure of the concerned data is exploited in novel ways. The first contribution of this thesis is a tensor learning-based low-complexity precoder codebook design technique for a full-dimension multiple-input multiple-output (FD-MIMO) system with a uniform planar antenna (UPA) array at the transmitter (Tx) whose channel distribution is available through a dataset. Represented as a tensor, the FD-MIMO channel is further decomposed using a tensor decomposition technique to obtain an optimal precoder which is a function of Kronecker-Product (KP) of two low-dimensional precoders, each corresponding to the horizontal and vertical dimensions of the FD-MIMO channel. From the design perspective, we have made contributions in deriving a criterion for optimal product precoder codebooks using the obtained low-dimensional precoders. We show that this product codebook design problem is an unsupervised clustering problem on a Cartesian Product Grassmann Manifold (CPM), where the optimal cluster centroids form the desired codebook. We further simplify this clustering problem to a $K$-means algorithm on the low-dimensional factor Grassmann manifolds (GMs) of the CPM which correspond to the horizontal and vertical dimensions of the UPA, thus significantly reducing the complexity of precoder codebook construction when compared to the existing codebook learning techniques. The second contribution of this thesis is a tensor-based bandwidth-efficient gradient communication technique for federated learning (FL) with convolutional neural networks (CNNs). Concisely, FL is a decentralized ML approach that allows to jointly train an ML model at the server using the data generated by the distributed users coordinated by a server, by sharing only the local gradients with the server and not the raw data. Here, we focus on efficient compression and reconstruction of convolutional gradients at the users and the server, respectively. To reduce the gradient communication overhead, we compress the sparse gradients at the users to obtain their low-dimensional estimates using compressive sensing (CS)-based technique and transmit to the server for joint training of the CNN. We exploit a natural tensor structure offered by the convolutional gradients to demonstrate the correlation of a gradient element with its neighbors. We propose a novel prior for the convolutional gradients that captures the described spatial consistency along with its sparse nature in an appropriate way. We further propose a novel Bayesian reconstruction algorithm based on the Generalized Approximate Message Passing (GAMP) framework that exploits this prior information about the gradients. Through the numerical simulations, we demonstrate that the developed gradient reconstruction method improves the convergence of the CNN model. Master of Science The increase in the number of wireless and mobile devices have led to the generation of massive amounts of multi-modal data at the users in various real-world applications including wireless communications. This has led to an increasing interest in machine learning (ML)-based data-driven techniques for communication system design. The native setting of ML is {em centralized} where all the data is available on a single device. However, the distributed nature of the users and their data has also motivated the development of distributed ML techniques. Since the success of ML techniques is grounded in their data-based nature, there is a need to maintain the efficiency and scalability of the algorithms to manage the large-scale data. Tensors are multi-dimensional arrays that provide an integrated way of representing multi-modal data. Tensor algebra and tensor decompositions have enabled the extension of several classical ML techniques to tensors-based ML techniques in various application domains such as computer vision, data-mining, image processing, and wireless communications. Tensors-based ML techniques have shown to improve the performance of the ML models because of their ability to leverage the underlying structural information in the data. In this thesis, we present two new applications of tensors to ML for wireless applications and show how the tensor structure of the concerned data can be exploited and incorporated in different ways. The first contribution is a tensor learning-based precoder codebook design technique for full-dimension multiple-input multiple-output (FD-MIMO) systems where we develop a scheme for designing low-complexity product precoder codebooks by identifying and leveraging a tensor representation of the FD-MIMO channel. The second contribution is a tensor-based gradient communication scheme for a decentralized ML technique known as federated learning (FL) with convolutional neural networks (CNNs), where we design a novel bandwidth-efficient gradient compression-reconstruction algorithm that leverages a tensor structure of the convolutional gradients. The numerical simulations in both applications demonstrate that exploiting the underlying tensor structure in the data provides significant gains in their respective performance criteria.