Present trends in communication links between devices have opted for wireless instead of wired solutions. As a consequence, unlicensed bands have seen a rise in the interference level as more and more devices are introduced into the market place that take advantage of these free bands for their communication needs. Under these conditions, the receiver's ability to recognize and identify the presence of interference becomes increasingly important. In order for the receiver to make an optimal decision on the signal-of-interest, it has to be aware of the type (modulation) of interference as well as how the received signals are affected (channel) by these impediments in order to appropriately mitigate them. This dissertation addresses the blind (unaided) identification of the signal modulations and the channel in a Multiple Input Multiple Output (MIMO) system. The method presented herein takes advantage of the modulation induced periodicities of the signals in the system and uses higher-order cyclostationary statistics to extract the signal and channel unknowns. This method can be used to identify more signals in the system than antenna elements at the receiver (overloaded case). This dissertation presents a system theoretic analysis of the problem as well as describes the development of an algorithm that can be used in the identification of the channel and the modulation of the signals in the system. Linear and non-linear receivers are examined at the beginning of the manuscript in order to review the a priori information that is needed for each receiver configuration to function properly. / Ph. D.

This thesis proposes a blind receiver for the Nyquist pulse shaped quadratureamplitude modulation (QAM) signals. The focus is on single carrier signals. The blind receiver includes the estimation of the symbol rate, the roll-off factor of the filter, the optimal sample phase, the frequency offset, the phase offset and as well as the correction of frequency and phase offsets. The blind receiver is proposed for the cognitive radio applications. Cognitive radios are intelligent devices which can adapt themselves according to its user and its environment, i.e. they are aware of the user and the environment. Another importance of cognitive radios is they can detect the incoming signal and demodulate it and also respond to the transmitting node with the same parameters. In order to demodulate the signal and to respond the transmitter node, there are some parameters which are needed to be known. The estimation starts with the bandwidth and carrier frequency, continued by the estimation of the symbol rate, which is a crucial factor. After the estimation and restrictions of these parameters, the roll-off factor of the filter is estimated for match filtering to remove the inter symbol interference (ISI) effect. Then the optimal sample phase is detected and the signal is downsampled. The following procedures include the modulation identification and estimation and correction of both frequency and phase offsets. The estimation algorithms performance is compared to the performances of the other algorithms available in the literature. These simulation results are presented and discussed in this thesis.

Deep learning has become an ubiquitous part of research in all fields, including wireless communications. Researchers have shown the ability to leverage deep neural networks (DNNs) that operate on raw in-phase and quadrature samples, termed Radio Frequency Machine Learning (RFML), to synthesize new waveforms, control radio resources, as well as detect and classify signals. While there are numerous advantages to RFML, this thesis answers the question "is it secure?" DNNs have been shown, in other applications such as Computer Vision (CV), to be vulnerable to what are known as adversarial evasion attacks, which consist of corrupting an underlying example with a small, intelligently crafted, perturbation that causes a DNN to misclassify the example. This thesis develops the first threat model that encompasses the unique adversarial goals and capabilities that are present in RFML. Attacks that occur with direct digital access to the RFML classifier are differentiated from physical attacks that must propagate over-the-air (OTA) and are thus subject to impairments due to the wireless channel or inaccuracies in the signal detection stage. This thesis first finds that RFML systems are vulnerable to current adversarial evasion attacks using the well known Fast Gradient Sign Method originally developed for CV applications. However, these current adversarial evasion attacks do not account for the underlying communications and therefore the adversarial advantage is limited because the signal quickly becomes unintelligible. In order to envision new threats, this thesis goes on to develop a new adversarial evasion attack that takes into account the underlying communications and wireless channel models in order to create adversarial evasion attacks with more intelligible underlying communications that generalize to OTA attacks. / Master of Science / Deep learning is beginning to permeate many commercial products and is being included in prototypes for next generation wireless communications devices. This technology can provide huge breakthroughs in autonomy; however, it is not sufficient to study the effectiveness of deep learning in an idealized laboratory environment, the real world is often harsh and/or adversarial. Therefore, it is important to know how, and when, these deep learning enabled devices will fail in the presence of bad actors before they are deployed in high risk environments, such as battlefields or connected autonomous vehicle communications. This thesis studies a small subset of the security vulnerabilities of deep learning enabled wireless communications devices by attempting to evade deep learning enabled signal classification by an eavesdropper while maintaining effective wireless communications with a cooperative receiver. The primary goal of this thesis is to define the threats to, and identify the current vulnerabilities of, deep learning enabled signal classification systems, because a system can only be secured once its vulnerabilities are known.