No description available.

No / Spectrum sensing is used in cognitive radio systems to detect the availability of spectrum holes for secondary usage. The simplest and most famous spectrum sensing techniques are based either on energy detection or eigenspace analysis from Random Matrix Theory (RMT) such as using the Marchenko-Pastur law. These schemes suffer from uncertainty in estimating the noise variance which reduces their performance. In this paper we propose a new method to evaluate the noise variance that can eliminate the limitations of the aforementioned schemes. This method estimates the noise variance from a measurement set of noisy signals or noise-only signals. Extensive simulations show that the proposed method performs well in estimating the noise variance. Its performance greatly improves with increasing numbers of measurements and also with increasing numbers of samples taken per measurement.

No / Dynamic Spectrum Access (DSA) for secondary usage of underutilized radio spectrum is currently of great interest for radio regulatory authorities and for cellular network operators. However, the co-existence of multiple devices operating in the same bands, such as wireless microphones which also operate in TV bands, poses a challenge to DSA. Efficient and proactive spectrum sensing could prevent harmful interference between collocated devices, but existing blind spectrum sensing schemes such as energy detection and schemes based on Random Matrix Theory (RMT) have performance limitations. We propose a new blind spectrum sensing scheme for cognitive radio. The proposed scheme uses a new formula for the estimation of noise variance. The scheme has been evaluated through extensive simulations on wireless microphone signals and shows higher performance as compared to energy detection and RMT-based sensing schemes such as MME and EME. It also shows higher performance in terms of probability of detection (Pd).

Recurrent neural networks (RNNs) have been widely used for supervised pattern recognition and exploring the underlying spatio-temporal correlation. However, due to the vanishing/exploding gradient problem, training a fully connected RNN in many cases is very difficult or even impossible. The difficulties of training traditional RNNs, led us to reservoir computing (RC) which recently attracted a lot of attention due to its simple training methods and fixed weights at its recurrent layer. There are three different categories of RC systems, namely, echo state networks (ESNs), liquid state machines (LSMs), and delayed feedback reservoirs (DFRs). In this dissertation a novel structure of RNNs which is inspired by dynamic delayed feedback loops is introduced. In the reservoir (recurrent) layer of DFR, only one neuron is required which makes DFRs extremely suitable for hardware implementations. The main motivation of this dissertation is to introduce an energy efficient, and easy to train RNN while this model achieves high performances in different tasks compared to the state-of-the-art. To improve the energy efficiency of our model, we propose to adopt spiking neurons as the information processing unit of DFR. Spiking neural networks (SNNs) are the most biologically plausible and energy efficient class of artificial neural networks (ANNs). The traditional analog ANNs have marginal similarity with the brain-like information processing. It is clear that the biological neurons communicate together through spikes. Therefore, artificial SNNs have been introduced to mimic the biological neurons. On the other hand, the hardware implementation of SNNs have shown to be extremely energy efficient. Towards achieving this overarching goal, this dissertation presents a spiking DFR (SDFR) with novel encoding schemes, and defense mechanisms against adversarial attacks. To verify the effectiveness and performance of the SDFR, it is adopted in three different applications where there exists a significant Spatio-temporal correlations. These three applications are attack detection in smart grids, spectrum sensing of multi-input-multi-output(MIMO)-orthogonal frequency division multiplexing (OFDM) Dynamic Spectrum Sharing (DSS) systems, and video-based face recognition. In this dissertation, the performance of SDFR is first verified in cyber attack detection in Smart grids. Smart grids are a new generation of power grids which guarantee a more reliable and efficient transmission and delivery of power to the costumers. A more reliable and efficient power generation and distribution can be realized through the integration of internet, telecommunication, and energy technologies. The convergence of different technologies, brings up opportunities, but the challenges are also inevitable. One of the major challenges that pose threat to the smart grids is cyber-attacks. A novel method is developed to detect false data injection (FDI) attacks in smart grids. The second novel application of SDFR is the spectrum sensing of MIMO-OFDM DSS systems. DSS is being implemented in the fifth generation of wireless communication systems (5G) to improve the spectrum efficiency. In a MIMO-OFDM system, not all the subcarriers are utilized simultaneously by the primary user (PU). Therefore, it is essential to sense the idle frequency bands and assign them to the secondary user (SU). The effectiveness of SDFR in capturing the spatio-temporal correlation of MIMO-OFDM time-series and predicting the availability of frequency bands in the future time slots is studied as well. In the third application, the SDFR is modified to be adopted in video-based face recognition. In this task, the SDFR is leveraged to recognize the identities of different subjects while they rotate their heads in different angles. Another contribution of this dissertation is to propose a novel encoding scheme of spiking neurons which is inspired by the cognitive studies of rats. For the first time, the multiplexing of multiple neural codes is introduced and it is shown that the robustness and resilience of the spiking neurons is increased against noisy data, and adversarial attacks, respectively. Adversarial attacks are small and imperceptible perturbations of the input data, which have shown to be able to fool deep learning (DL) models. So far, many adversarial attack and defense mechanisms have been introduced for DL models. Compromising the security and reliability of artificial intelligence (AI) systems is a major concern of government, industry and cyber-security researchers, in that insufficient protections can compromise the security and privacy of everyone in society. Finally, a defense mechanism to protect spiking neurons against adversarial attacks is introduced for the first time. In a nutshell, this dissertation presents a novel energy efficient deep spiking recurrent neural network which is inspired by delayed dynamic loops. The effectiveness of the introduced model is verified in several different applications. At the end, novel encoding and defense mechanisms are introduced which improve the robustness of the model against noise and adversarial attacks. / Doctor of Philosophy / The ultimate goal of artificial intelligence (AI) is to mimic the human brain. Artificial neural networks (ANN) are an attempt to realize that goal. However, traditional ANNs are very far from mimicking biological neurons. It is well-known that biological neurons communicate with one another through signals in the format of spikes. Therefore, artificial spiking neural networks (SNNs) have been introduced which behave more similarly to biological neurons. Moreover, SNNs are very energy efficient which makes them a suitable choice for hardware implementation of ANNs (neuromporphic computing). Despite the many benefits that are brought about by SNNs, they are still behind traditional ANNs in terms of performance. Therefore, in this dissertation, a new structure of SNNs is introduced which outperforms the traditional ANNs in three different applications. This new structure is inspired by delayed dynamic loops which exist in biological brains. The main objective of this novel structure is to capture the spatio-temporal correlation which exists in time-series while the training overhead and power consumption is reduced. Another contribution of this dissertation is to introduce novel encoding schemes for spiking neurons. It is clear that biological neurons leverage spikes, but the language that they use to communicate is not clear. Hence, the spikes require to be encoded in a certain language which is called neural spike encoding scheme. Inspired by the cognitive studies of rats, a novel encoding scheme is presented. Lastly, it is shown that the introduced encoding scheme increases the robustness of SNNs against noisy data and adversarial attacks. AI models including SNNs have shown to be vulnerable to adversarial attacks. Adversarial attacks are minor perturbations of the input data that can cause the AI model to misscalassify the data. For the first time, a defense mechanism is introduced which can protect SNNs against such attacks.

With the increase in spectrum congestion, intelligent spectrum sensing systems have become more important than ever before. In the field of Radio Frequency Machine Learning (RFML), techniques like deep neural networks and reinforcement learning have been used to develop more complex spectrum sensing systems that are not reliant on expert features. Architectures like Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN) have shown great promise for applications like automated modulation classification, signal detection, and specific emitter ID. Research in these areas has primarily focused on "one size fits all" networks that assume a fixed signal length in both training and inference. However, since some signals are more complex than others, due to channel conditions, transmitter/receiver effects, etc., being able to dynamically utilize just enough of the received symbols to make a reliable decision allows for more efficient decision making in applications such as electronic warfare and dynamic spectrum sharing. Additionally, the operator may want to get to the quickest possible decision. Recurrent neural networks have been shown to outperform other architectures when processing temporally correlated data, such as from wireless communication signals. However, compared to other architectures, such as CNNs, RNNs can suffer from drastically longer training and evaluation times due to their inherent sample-by-sample data processing. While traditional usage of both of these architectures typically assumes a fixed observation interval during both training and testing, the sample-by-sample processing capabilities of recurrent neural networks opens the door for "decoupling" these intervals. This is invaluable in real-world applications due to the relaxation of the typical requirement of a fixed time duration of the signals of interest. This work illustrates the benefits and considerations needed when "decoupling" these observation intervals for spectrum sensing applications. In particular, this work shows that, intuitively, recurrent neural networks can be leveraged to process less data (i.e. shorter observation intervals) for simpler inputs (less complicated signal types or channel conditions). Less intuitively, this works shows that the "decoupling" is dependent on appropriate training to avoid bias and insure generalization. / Master of Science / With the increase in spectrum congestion, intelligent spectrum sensing systems have become more important than ever before. In the field of Radio Frequency Machine Learning (RFML), techniques like deep neural networks and reinforcement learning have been used to develop more complex spectrum sensing systems that are not reliant on expert features. Architectures like Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN) have shown great promise for applications like automated modulation classification, signal detection, and specific emitter ID. Research in these areas has primarily focused on "one size fits all" networks that assume a fixed signal length in both training and inference. However, since some signals are more complex than others, due to channel conditions, transmitter/receiver effects, etc., being able to dynamically utilize just enough of the received symbols to make a reliable decision allows for more efficient decision making in applications such as electronic warfare and dynamic spectrum sharing. Additionally, the operator may want to get to the quickest possible decision. Recurrent neural networks have been shown to outperform other architectures when processing temporally correlated data, such as from wireless communication signals. However, compared to other architectures, such as CNNs, RNNs can suffer from drastically longer training and evaluation times due to their inherent sample-by-sample data processing. While traditional usage of both of these architectures typically assumes a fixed observation interval during both training and testing, the sample-by-sample processing capabilities of recurrent neural networks opens the door for "decoupling" these intervals. This is invaluable in real-world applications due to the relaxation of the typical requirement of a fixed time duration of the signals of interest. This work illustrates the benefits and considerations needed when "decoupling" these observation intervals for spectrum sensing applications. In particular, this work shows that, intuitively, recurrent neural networks can be leveraged to process less data (i.e. shorter observation intervals) for simpler inputs (less complicated signal types or channel conditions). Less intuitively, this works shows that the "decoupling" is dependent on appropriate training to avoid bias and insure generalization.

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.

Recently, automatic modulation classification techniques using deep neural networks on raw IQ samples have been investigated and show promise when compared to more traditional likelihood-based or feature-based techniques. While likelihood-based and feature-based techniques are effective, making classification decisions directly on the raw IQ samples removes the need for expertly crafted transformations and feature extractions. In practice, RF environments are typically very dense, and a receiver must first detect and isolate each signal of interest before classification can be performed. The errors introduced by this detection and isolation process will affect the accuracy of deep neural networks making automatic modulation classification decisions directly on raw IQ samples. The importance of defining upper limits on estimation errors in a detector is highlighted, and the negative effects of over-estimating or under-estimating these limits is explored. Additionally, to date, most of the published research has focused on synthetically generated data. While large amounts of synthetically generated data is generally much easier to obtain than real-world signal data, it requires expert knowledge and accurate models of the real world, which may not always be realistic. The experiments conducted in this work show how augmented real-world signal captures can be successfully used for training neural networks used in automatic modulation classification on raw IQ samples. It is shown that the quality and duration of real world signal captures is extremely important when creating training datasets, and that signal captures made from a single transmitter with one receiver can be broadly applicable to other radios through dataset augmentation. / Master of Science

One of the fundamental hurdles in realizing new spectrum sharing allocation policies is that of reliable spectrum sensing. In this thesis, three research thrusts are presented in order to further research in this critical area. The first of these research thrusts is the development of a novel asynchronous and noncoherent modulation classifier for PSK/QAM modulated signals in flat-fading channels. In developing this classifier, a novel estimator for the unknown channel gain and fractional time delay is proposed which uses a method-of-moments based estimation approach. For the second research thrust of this thesis, the developed method-of-moments based estimation approach is extended to estimate the signal-to-noise ratio of PSK/QAM modulated signals in flat-fading channels, in which no a priori knowledge of the modulation format and channel parameters is assumed. Finally, in the third research thrust, a distributed spectrum sensing approach is proposed in which a network of radios collaboratively detects the presence, as well as the modulation scheme, of a signal through the use of a combination of cyclic spectrum feature-based signal classification and an iterative algorithm for optimal data fusion. / Master of Science

Robust nonparametric spectral estimation includes generating an accurate estimate of the Power Spectral Density (PSD) for a given set of data while trying to minimize the bias due to data outliers. Robust nonparametric spectral estimation is applied in the domain of electrical communications and digital signal processing when a PSD estimate of the electromagnetic spectrum is desired (often for the goal of signal detection), and when the spectrum is also contaminated by Impulsive Noise (IN). Power Line Communication (PLC) is an example of a communication environment where IN is a concern because power lines were not designed with the intent to transmit communication signals. There are many different noise models used to statistically model different types of IN, but one popular model that has been used for PLC and various other applications is called the Middleton Class A model, and this model is extensively used in this thesis. The performances of two different nonparametric spectral estimation methods are analyzed in IN: the Welch method and the multitaper method. These estimators work well under the common assumption that the receiver noise is characterized by Additive White Gaussian Noise (AWGN). However, the performance degrades for both of these estimators when they are used for signal detection in IN environments. In this thesis basic robust estimation theory is used to modify the Welch and multitaper methods in order to increase their robustness, and it is shown that the signal detection capabilities in IN is improved when using the modified robust estimators. / Master of Science / One application of blind spectral estimation is blind signal detection. Unlike a car radio, where the radio is specifically designed to receive AM and PM radio waves, sometimes it is useful for a radio to be able to detect the presence of transmitted signals whose characteristics are not known ahead of time. Cognitive radio is one application where this capability is useful. Often signal detection is inhibited by Additive White Gaussian Noise (AWGN). This is analogous to trying to hear a friend speak (signal detection) in a room full of people talking (background AWGN). However, some noise environments are more impulsive in nature. Using the previous analogy, the background noise could be loud banging caused by machinery; the noise will not be as constant as the chatter of the crowd, but it will be much louder. When power lines are used as a medium for electromagnetic communication (instead of just sending power), it is called Power Line Communication (PLC), and PLC is a good example of a system where the noise environment is impulsive. In this thesis, methods used for blind spectral estimation are modified to work reliably (or robustly) for impulsive noise environments.

Face à une demande sans cesse croissante de haut débit et d’adaptabilité des systèmes existants, qui à son tour se traduit par l’encombrement du spectre, le développement de nouvelles solutions dans le domaine des communications sans fil devient nécessaire afin de répondre aux exigences des applications émergentes. Parmi les innovations récentes dans ce domaine, l’ultra large bande (UWB) a suscité un vif intérêt. La radio impulsionnelle UWB (IR-UWB), qui est une solution intéressante pour réaliser des systèmes UWB, est caractérisée par la transmission des impulsions de très courte durée, occupant une largeur de bande allant jusqu’à 7,5 GHz, avec une densité spectrale de puissance extrêmement faible. Cette largeur de bande importante permet de réaliser plusieurs fonctionnalités intéressantes, telles que l’implémentation à faible complexité et à coût réduit, la possibilité de se superposer aux systèmes à bande étroite, la diversité spatiale et la localisation très précise de l’ordre centimétrique, en raison de la résolution temporelle très fine.Dans cette thèse, nous examinons certains éléments clés dans la réalisation d'un système IR-UWB intelligent. Nous avons tout d’abord proposé le concept de radio UWB cognitive à partir des similarités existantes entre l'IR-UWB et la radio cognitive. Dans sa définition la plus simple, un tel système est conscient de son environnement et s'y adapte intelligemment. Ainsi, nous avons tout d’abord focalisé notre recherché sur l’analyse de la disponibilité des ressources spectrales (spectrum sensing) et la conception d’une forme d’onde UWB adaptative, considérées comme deux étapes importantes dans la réalisation d'une radio cognitive UWB. Les algorithmes de spectrum sensing devraient fonctionner avec un minimum de connaissances a priori et détecter rapidement les utilisateurs primaires. Nous avons donc développé de tels algorithmes utilisant des résultats récents sur la théorie des matrices aléatoires, qui sont capables de fournir de bonnes performances, avec un petit nombre d'échantillons. Ensuite, nous avons proposé une méthode de conception de la forme d'onde UWB, vue comme une superposition de fonctions B-splines, dont les coefficients de pondération sont optimisés par des algorithmes génétiques. Il en résulte une forme d'onde UWB qui est spectralement efficace et peut s’adapter pour intégrer les contraintes liées à la radio cognitive. Dans la 2ème partie de cette thèse, nous nous sommes attaqués à deux autres problématiques importantes pour le fonctionnement des systèmes UWB, à savoir la synchronisation et l’estimation du canal UWB, qui est très dense en trajets multiples. Ainsi, nous avons proposé plusieurs algorithmes de synchronisation, de faible complexité et sans séquence d’apprentissage, pour les modulations BPSK et PSM, en exploitant l'orthogonalité des formes d'onde UWB ou la cyclostationnarité inhérente à la signalisation IR-UWB. Enfin, nous avons travaillé sur l'estimation du canal UWB, qui est un élément critique pour les récepteurs Rake cohérents. Ainsi, nous avons proposé une méthode d’estimation du canal basée sur une combinaison de deux approches complémentaires, le maximum de vraisemblance et la décomposition en sous-espaces orthogonaux,d’améliorer globalement les performances. / Faced with an ever increasing demand of high data-rates and improved adaptability among existing systems, which inturn is resulting in spectrum scarcity, the development of new radio solutions becomes mandatory in order to answer the requirements of these emergent applications. Among the recent innovations in the field of wireless communications,ultra wideband (UWB) has generated significant interest. Impulse based UWB (IR-UWB) is one attractive way of realizing UWB systems, which is characterized by the transmission of sub nanoseconds UWB pulses, occupying a band width up to 7.5 GHz with extremely low power density. This large band width results in several captivating features such as low-complexity low-cost transceiver, ability to overlay existing narrowband systems, ample multipath diversity, and precise ranging at centimeter level due to extremely fine temporal resolution.In this PhD dissertation, we investigate some of the key elements in the realization of an intelligent time-hopping based IR-UWB system. Due to striking resemblance of IR-UWB inherent features with cognitive radio (CR) requirements, acognitive UWB based system is first studied. A CR in its simplest form can be described as a radio, which is aware ofits surroundings and adapts intelligently. As sensing the environment for the availability of resources and then consequently adapting radio’s internal parameters to exploit them opportunistically constitute the major blocks of any CR, we first focus on robust spectrum sensing algorithms and the design of adaptive UWB waveforms for realizing a cognitive UWB radio. The spectrum sensing module needs to function with minimum a-priori knowledge available about the operating characteristics and detect the primary users as quickly as possible. Keeping this in mind, we develop several spectrum sensing algorithms invoking recent results on the random matrix theory, which can provide efficient performance with a few number of samples. Next, we design the UWB waveform using a linear combination of Bsp lines with weight coefficients being optimized by genetic algorithms. This results in a UWB waveform that is spectrally efficient and at the same time adaptable to incorporate the cognitive radio requirements. In the 2nd part of this thesis, some research challenges related to signal processing in UWB systems, namely synchronization and dense multipath channel estimation are addressed. Several low-complexity non-data-aided (NDA) synchronization algorithms are proposed for BPSK and PSM modulations, exploiting either the orthogonality of UWB waveforms or theinherent cyclostationarity of IR-UWB signaling. Finally, we look into the channel estimation problem in UWB, whichis very demanding due to particular nature of UWB channels and at the same time very critical for the coherent Rake receivers. A method based on a joint maximum-likelihood (ML) and orthogonal subspace (OS) approaches is proposed which exhibits improved performance than both of these methods individually.