Recent innovations in machine learning have paved the way for new capabilities in the field of radio frequency (RF) communications. Machine learning techniques such as reinforcement learning and deep neural networks (DNN) can be leveraged to improve upon traditional wireless communications methods so that they no longer require expertly-defined features. Simultaneously, cybersecurity and electronic warfare are growing areas of focus and concern in an increasingly technology-driven world. Privacy and confidentiality of communication links are both more important and more difficult than ever in the current high threat environment. RF machine learning (RFML) systems contribute to this threat as they have been shown to be successful in gleaning information from intercepted signals, through the use of learning-enabled eavesdroppers. This thesis focuses on a method of defense against such communications threats termed an adversarial evasion attack in which intelligently crafted perturbations of the RF signal are used to fool a DNN-enabled classifier, therefore securing the communications channel. One often overlooked aspect of evasion attacks is the concept of maintaining intended use. In other words, while an adversarial signal, or more generally an adversarial example, should fool the DNN it is attacking, this should not come at the detriment to it's primary application. In RF communications, this manifests in the idea that the communications link must be successfully maintained with friendly receivers, even when executing an evasion attack against malicious receivers. This is a difficult scenario, made even more so by the nature of channel effects present in over-the-air (OTA) communications, as is assumed in this work. Previous work in this field has introduced a form of evasion attack for RFML systems called a communications aware attack that explicitly addresses the reliable communications aspect of the attack by training a separate DNN to craft adversarial signals; however, this work did not utilize the full RF processing chain and left residual indicators of the attack that could be leveraged for defensive capabilities. First, this thesis focuses on implementing forward error correction (FEC), an aspect present in most communications systems, in the training process of the attack. It is shown that introducing this into the training stage allows the communications aware attack to implicitly use the structure of the coding to create smarter and more efficient adversarial signals. Secondly, this thesis then addresses the fact that in previous work, the resulting adversarial signal exhibiting significant out-of-band frequency content, a limitation that can be used to render the attack ineffective if preprocessing at the attacked DNN is assumed. This thesis presents two novel approaches to solve this problem and eliminate the majority of side content in the attack. By doing so, the communications aware attack is more readily applicable to real-world scenarios. / Master of Science / Deep learning has started infiltrating many aspects of society from the military, to academia, to commercial vendors. Additionally, with the recent deployment of 5G technology, connectivity is more readily accessible than ever and an increasingly large number of systems will communicate with one another across the globe. However, cybersecurity and electronic warfare call into question the very notion of privacy and confidentiality of data and communication streams. Deep learning has further improved these intercepting capabilities. However, these deep learning systems have also been shown to be vulnerable to attack. This thesis exists at the nexus of these two problems, both machine learning and communication security. This work expands upon adversarial evasion attacks meant to help elude signal classification at a deep learning-enabled eavesdropper while still providing reliable communications to a friendly receiver. By doing so, this work both provides a new methodology that can be used to conceal communication information from unwanted parties while also highlighting the glaring vulnerabilities present in machine learning systems.

Exploring RF models for an integrated power electronics module (IPEM) is crucial to analyzing and predicting its EMI performance. This thesis deals with the parasitics extraction approach for an active IPEM in a frequency range of 1MHz through 30MHz. Based on the classic electromagnetic field theory, the calculating equations of DC and AC parameters for a 3D conducting structure are derived. The influence of skin effect and proximity effect on AC resistances and inductances is also investigated at high frequencies. To investigate RF models and EMI performance of the IPEM, a 1kW 1MHz series resonant DC-DC converter (SRC) is designed and fabricated in this work. For extracting the stray parameters of the built IPEM, two main software simulation tools ¡ª Maxwell Quick 3D Parameter Extractor (Maxwell Q3D) and Maxwell 3D Field Simulator (Maxwell 3D), prevailing electromagnetic simulation products from Ansoft Corporation, are introduced in this study. By trading off between the numerical accuracy and computational economy (CPU time and consumption of memory size), Maxwell Q3D is chosen in this work to extract the parameters for the full bridge IPEM structure. The step-by-step procedure of using Maxwell Q3D is presented from pre-processing the 3D IPEM structure to post-processing the solutions, and exporting equivalent circuit for PSpice simulations as well. RF modeling of power MOSFETs is briefly introduced. In order to verify extracted parameters, in-circuit impedance measurements for the IPEM using Agilent 4294A Impedance Analyzer together with Agilent 42941A probe are then followed. Measured results basically verify the extracted data, while the discrepancy between measured results and simulated results is also analyzed. / Master of Science

This thesis investigates the interference potential of microwave ovens to Bluetooth data communication. Interference experiments are conducted in the CWT's Bluetooth lab, using CSR™ (Cambridge Silicone Radio) Bluetooth radios and a Tektronics™ Protocol Analyzer to record packet transmissions between the master and slave units. A novel, "quasi-real time" spectral measurement concept is developed to take radio frequency measurements. A LabView program enables a spectrum analyzer to download oven spectral data onto a computer via the instrument's serial port. From this data, three-dimensional plots of microwave radiated power levels versus ISM band frequencies over time periods are produced for different microwave ovens. These plots are compared with the results of interference experiments to explain Bluetooth packet errors. In addition to causing packet errors, emitted oven power levels at certain frequencies are sometimes strong enough to cause data packets to be lost (dropped) as they are transmitted over the air. This is a major problem since the Protocol Analyzer does not "see" these packets and cannot record the transmissions during an experiment. These lost packets can be accounted for if the frequency hopping scheme of the communicating Bluetooth devices is know prior to data transmission. Bluetooth's Frequency Hop Scheme is coded in Matlab for the purpose of predicting a data transmission's hopping sequence. The lost packets on each Bluetooth channel are counted by subtracting the Analyzer's recorded number of data transmissions per channel from the total number of transmissions per channel predicted by Matlab. A method is devised to calibrate the Bluetooth receiver and the spectrum analyzer is used to measure the received power level of Bluetooth signals on a particular frequency (channel). The number of packet errors on a channel is determined from the channel's C/I (carrier-to-interference ratio). If a channel's C/I level falls below the calculated C/I threshold at any instant of time due to oven operation, the packet transmitted at that instant is likely in error. A Matlab program estimates the number of packet errors per channel by counting the number of times the C/I of a channel falls below it's threshold value. The predicted number of packet errors is compared with the measured packet errors from experiments to yield extremely good results. Various oven-interference experiments are conducted in a small building, a large office environment and outdoors. For each experiment, the number of occurrences of transmitted data is plotted for each Bluetooth channel. Composite Excel bar graphs, created from this data, are compared with the oven spectral plots to describe an oven's effect on Bluetooth transmission. It is determined that different ovens cause packet errors on specifically different channels, in addition to channels 52-54 around the oven's 2450 MHz center operating frequency. The interference experiments suggest that placing an oven a radius of 10 m away from Class I (devices in a piconet will not affect data transmission). / Master of Science

No description available.

Widely used design techniques for radio frequency power amplifiers yield results which are approximate; the initial design is usually refined by applying trial-and-error procedures in the laboratory. More accurate design techniques are complicated in their application and have not gained acceptance by practicing engineers. A new design technique for VHF linear power amplifiers using bipolar junction transistors is presented in this report. This design technique is simple in its application but yields accurate results. The design technique is based upon a transistor model which is simple enough to be useful for design, but which is sufficiently accurate to predict performance at high frequencies. Additionally, the model yields insight into many of the processes which take place within the typical RF power transistor. The fundamental aspect of the model is the inclusion of charge storage within the transistor base. This charge storage effect gives rise to a nearly sinusoidal collector current waveform, even in a transistor which ostensibly is biased for class B or nonsaturating class C operation. Methods of predicting transistor input and output impedances are presented. A number of other topics related to power amplifier measurement and design are also included. A unique measurement approach which is ideally suited for use with power amplifiers is discussed. This measurement approach is a hybrid of the common S-parameter measurement technique and the "load-pull" procedure. Practical considerations such as amplifier stability, bias network design, and matching network topology are also included in the report. / Ph. D.

No description available.

No description available.

The objective of this work is to enhance and advance sensing technologies with the design and development of novel radio frequency (RF) sensors based on far-field and near-field principles of the electromagnetic (EM) resonances. In the first part of this thesis, original design and development of a passive RF temperature sensor, a passive RF strain sensor, and a passive RF pressure sensor are presented. The RF temperature sensor is presented in Chapter 3. It is based on split ring resonators loaded with bimorph cantilevers. Its operating principles and equivalent circuits are discussed in Chapter 4, where the design concept is illustrated to be robust and highly adaptable to different sensing ranges, environments, and applicable to other type of sensing beyond temperatures. The passive RF strain sensor, based on a patch antenna loaded with a cantilever-integrated open loop, is presented in Chapter 5, where it is demonstrated to have the highest strain sensitivity in the same remote and passive class of sensors in the state-of-the-art. Chapter 6 describes the passive RF pressure sensor, which is based on a dual-band stacked-patch antenna that allows both identification and sensing to be embedded in its unique dual resonant responses. In the second part of this thesis, an original and first-of-its-kind RF transducer is presented that enables non-touch sensing of human fingers within 3 cm of proximity (based on one unit sensor cell). The RF transducer is based on a slotted microstrip patch coupled to a half-wavelength parallel-coupled microstrip filter operating in the frequency range of 6 – 8 GHz. The sensing mechanism is based on the EM near-field coupling between the resonator and the human finger. Fundamentally different from the electric field capacitive sensing, this new method of sensing, the first of its kind, based on near-field interference that produces a myriad of nonlinearities in the sensing response, can introduce new capabilities for the interface of electronic displays (the detection is based on pattern recognition). What set this sensor and its platform apart from previous proximity sensors and microwave sensing platforms is the low profile planar structure of the system, and its compatibility with mobile applications. The thesis provides both breadth and depth in the proposed design and development and thus presenting a complete research in its contributions to RF sensing.

This dissertation deals with the challenge of power utility in South Africa which is on proactive detection of fallen power line conductors and real time sagging measurement together with slipping of such conductors. Various methods which are currently used for sag detection were characterized and evaluated to the aim of the research. A mathematical reconstruction done to estimate the lowest point of the conductor in a span is presented. Practical simulations and application of radio frequency identification (RFID) for sag detection is attempted through matlab software. RFID radar system is then analyzed in different modes and found to give precision measurement for sag in real time as opposed to global positioning system (GPS) if one dimension of the tag assumed fixed on the power line. Lastly errors detected on the measurements are corrected using a trainable artificial neural network. A conclusion is made by making recommendations in the advancement of the research. / Electrical Engineering / M. Tech. (Electrical Engineering)

A necessidade da medicina por equipamentos eletrônicos portáteis, menores, confiáveis e baratos favorece o desenvolvimento de técnicas biotelemétricas passivas de monitoração nas aplicações biomédicas invasivas. Nesse trabalho são apresentados o desenvolvimento e o teste de um dispositivo sensor biotelemétrico passivo, que utiliza a tecnologia de identificação por radio freqüência - RFID (Radio Frequency Identification). O dispositivo desenvolvido, comumente chamado como tag, é comporto basicamente por três partes: a arquitetura RFID utilizada para baixa freqüência, a unidade de controle responsável pelo processamento dos dados e a unidade sensora responsável pela monitoração da pressão arterial. Como o tag e passivo, ou seja, não contém baterias, para que seja ativado é necessário um aparelho que faça sua energização e também decodifique os dados por ele transmitidos. Este aparelho que faz a leitura e envia um sinal para ativar o transponder é comumente chamado de leitora. A leitora utilizada neste projeto terá seu firmware adaptado de forma a processar o valor da pressão enviada pelo tag. São abordados neste trabalho tanto a modelagem teórica do sistema quanto a especificação prática dos componentes para os testes de validação. Na modelagem teórica são apresentados os modelos matemáticos comportamentais do sistema. Os resultados obtidos validam a metodologia utilizada para o desenvolvimento de um sensor RFID passivo que tem como finalidade mensurar a pressão arterial. / The need of medicine for portable electronic equipments smallers [sic], reliables [sic] and inexpensive supports the development of biotelemetry techniques passive monitoring in invasive biomedical applications. In this work are presents [sic] the development and testing of a passive biotelemetry sensor device, which uses the technology of Radio Frequency Identification - RFID. The developed device, commonly referred to as the tag, is basically composed of three parts: the RFID architecture used for low frequency, the control unit responsible for data processing and sensor unit responsible for monitoring blood pressure. As the tag is passive i.e. does not contain batteries to activate it, It [sic] is necessary a device that makes its energizing and also decode the data transmitted by it. This device that reads and sends a signal to activate the transponder is commonly called a reader. The reader used in this project will have its firmware adapted to handle the pressure value sent by the tag. This work also presents theoretical modeling of the system and the specification of components for practicing the validation tests. In theoretical modeling are presented mathematical models of system behavior. The results validate the methodology used for the development of passive RFID sensor that aims to measure blood pressure.