Specific emitter identification refers to the process of performing identification of radio frequency transmitters by exploiting unique variations in emitted signals, caused by hardware variations. In previous research, specific emitter identification was successfully performed on GSM handsets. However, no research has been done on the implementation of specific emitter identification of GSM handsets on an FPGA platform. This study focuses on feature extraction and identification algorithms, as well as the implementation of the identification algorithm on an FPGA. During this study, phase modulation error was used, as previous research indicated that phase modulation error is an effective feature set for identification purposes. As the implementation of a classification algorithm on an FPGA was required, a trade-off between complexity and feasibility needed to be made during the selection process. The artificial neural network was selected as the optimal classifier for implementation on an FPGA. The algorithm was first implemented in software and used as the basis for the design on an FPGA. A piece-wise linear approximation of a sigmoid function was used to approximate the activation function, where a look-up table was used to store the parameters. The off-line training of the artificial neural network was performed in software using the back-propagation gradient descent algorithm. Good results for the identification of GSM handsets on an FPGA were obtained, with a true acceptance ratio of 97.0%. This result is similar to the performance obtained in previous research performed in software. In this study, it was found that specific emitter identification of GSM handsets can be performed on an FPGA. Real-world applications for this technology include general cellular handset identification and access control. / MSc (Electrical and Electronic Engineering), North-West University, Potchefstroom Campus, 2015

Specific emitter identification refers to the process of performing identification of radio frequency transmitters by exploiting unique variations in emitted signals, caused by hardware variations. In previous research, specific emitter identification was successfully performed on GSM handsets. However, no research has been done on the implementation of specific emitter identification of GSM handsets on an FPGA platform. This study focuses on feature extraction and identification algorithms, as well as the implementation of the identification algorithm on an FPGA. During this study, phase modulation error was used, as previous research indicated that phase modulation error is an effective feature set for identification purposes. As the implementation of a classification algorithm on an FPGA was required, a trade-off between complexity and feasibility needed to be made during the selection process. The artificial neural network was selected as the optimal classifier for implementation on an FPGA. The algorithm was first implemented in software and used as the basis for the design on an FPGA. A piece-wise linear approximation of a sigmoid function was used to approximate the activation function, where a look-up table was used to store the parameters. The off-line training of the artificial neural network was performed in software using the back-propagation gradient descent algorithm. Good results for the identification of GSM handsets on an FPGA were obtained, with a true acceptance ratio of 97.0%. This result is similar to the performance obtained in previous research performed in software. In this study, it was found that specific emitter identification of GSM handsets can be performed on an FPGA. Real-world applications for this technology include general cellular handset identification and access control. / MSc (Electrical and Electronic Engineering), North-West University, Potchefstroom Campus, 2015

Radio Frequency Machine Learning (RFML) techniques for the classification tasks of Specific Emitter Identification (SEI) and Automatic Modulation Classification (AMC) have seen rapid improvements in recent years. The applications of SEI, a technique used to associate a received signal to an emitter, and AMC, a technique for determining the modulation scheme present within a transmission, are necessary for a variety of defense applications such as early warning systems and emitter tracking. Existing works studying SEI and AMC have sought to perform and improve classification through the use of various different machine learning (ML) model architectures. In ideal conditions, these efforts have shown strong classification results, however, when robust real-world data is applied to these models, performance notably decreases. Further efforts, therefore, are required to understand why each of these models fails in adverse conditions. With this understanding, robust architectures that are able to maintain performance in the presence of various data conditions can be created. The work presented in this thesis seeks to improve upon SEI and AMC models by furthering the understanding of how certain model architectures fail under varying data conditions, then applies Transfer Learning (TL) and Ensemble Learning techniques in an effort to mitigate discovered failures and improve the applicability of trained models to various types of data. Each of the approaches presented in this work utilize real-world datasets, collected in a way that emulate a variety of possible real life use conditions of RFML systems. Results show that existing AMC approaches are fairly robust to varying data conditions, while SEI approaches suffer a significant degradation in performance under conditions that differ than that used to train a given model. Further, TL and ensemble techniques can be utilized to improve the robustness of RFML models. This thesis helps isolate the rate and features of those SEI degradations, hopefully setting a foundation for future improvements. / Master of Science / Radio Frequency (RF) signals are produced by many different emitters encountered on a daily basis, including phones, networks, radar, and radios. These signals are used to transfer information from an emitter to a receiver, and contain a plethora of information that need be protected for defense practices in the RF domain. On the other hand, the information contained in these signals can be intercepted and utilized to discover information about potentially malicious transmissions. Two practices to determine information about received signals include Specific Emitter Identification (SEI), which relates an emitter to a received signal, and Automatic Modulation Classification (AMC), which determines the modulation scheme in which a signal is transmitted. A signal is made up of information, expressed in bits, and a modulation scheme is the method used to map those bits to express information. In recent years, Machine Learning (ML) techniques have been applied to SEI and AMC in an effort to improve the efficiency and accuracy results of classification. These ML approaches have shown high accuracy results when applied to data that is collected in the same environment as that used for training. When applied to data with different variables, however, model accuracy notably drops. This performance decrease motivates the need to discover more variables that negatively impact model performance, and further to create models that do not suffer from the same weaknesses. This work examines four different real-world variables that are common in deployed radio frequency machine learning (RFML) usage environments, and using the information learned about model failures, implements two approaches to create models that are more robust to variances in data. This work finds that model performance varies when exposed to variations in temperature, signal-to-noise ratio (SNR), training data quantity, and receiver hardware. Further, this work finds that Transfer Learning (TL) and Ensemble Learning can be used to create models that mitigate these discovered weaknesses.

Specific emitter identification is a technology used to uniquely identify a class of wireless devices, and in some cases a single device. Minute differences in the implementation of a wireless communication standard from one device manufacturer to another make it possi- ble to extract a wireless "fingerprint" from the transmitted signal. These differences may stem from imperfect radio frequency (RF) components such as filters and power amplifiers. However, the problem of identifying a wireless device through analysis of these key signal characteristics presents several difficulties from an algorithmic perspective. Given that the differences in these features can be extremely subtle, in general a high signal to noise ratio (SNR) is necessary for a sufficient probability of correct detection. If a sufficiently high SNR is not guaranteed, then some from of identification algorithm which operates well in low SNR conditions must be used. Cyclostationary analysis offers a method of specific emitter iden- tification through analysis of second order spectral correlation features which can perform well at relatively low SNRs. The eigenvector/eigenvalue decomposition (EVD) is capable of separating principal components from uncorrelated gaussian noise. This work proposes a technique of specific emitter identification which utilizes the principal components of the EVD of the spectral correlation function which has been arranged into a square matrix. An analysis of this EVD-based SEI technique is presented herein, and some limitations are identified. Analysis is constrained to orthogonal frequency division multiplexing (OFDM) using the IEEE 802.16 specification (used for WiMAX) as a guideline for a variety of pilot arrangements. / Master of Science

Radio Frequency Machine Learning (RFML) is the application of ML techniques to solve problems in the RF domain as an alternative to traditional digital-signal processing (DSP) techniques. Notable among these are the tasks of specific emitter identification (SEI), determining source identity of a received RF signal, and automated modulation classification (AMC), determining the modulation scheme of a received RF transmission. Both tasks have a number of algorithms that are effective on simulated data, but struggle to generalize to data collected in the real-world, partially due to the lack of available datasets upon which to train models and understand their limitations. This thesis covers the practical considerations for systems that can create high-quality datasets for RFML tasks, how variances from real-world effects in these datasets affect RFML algorithm performance, and how well models developed from these datasets are able to generalize and adapt across different receiver hardware platforms. Moreover, this thesis presents a proof-of-concept system for large-scale and efficient data generation, proven through the design and implementation of a custom platform capable of coordinating transmissions from nearly a hundred Software-Defined Radios (SDRs). This platform was used to rapidly perform experiments in both RFML performance sensitivity analysis and successful transfer between SDRs of trained models for both SEI and AMC algorithms. / Master of Science / Radio Frequency Machine Learning (RFML) is the application of machine learning techniques to solve problems having to do with radio signals as an alternative to traditional signal processing techniques. Notable among these are the tasks of specific emitter identification (SEI), determining source identity of a received signal, and automated modulation classification (AMC), determining the data encoding format of a received RF transmission. Both tasks have practical limitations related to the real-world collection of RF training data. This thesis presents a proof-of-concept for large-scale, efficient data generation and management, as proven through the design and construction of a custom platform capable of coordinating transmissions from nearly a hundred radios. This platform was used to rapidly perform experiments in both RFML performance sensitivity analysis and successful cross-radio transfer of trained behaviors.

Specific Emitter Identification (SEI) is the association of a received signal to an emitter, and is made possible by the unique and unintentional characteristics an emitter imparts onto each transmission, known as its radio frequency (RF) fingerprint. SEI systems are of vital importance to the military for applications such as early warning systems, emitter tracking, and emitter location. More recently, cognitive radio systems have started making use of SEI systems to enforce Dynamic Spectrum Access (DSA) rules. The use of pre-determined and expert defined signal features to characterize the RF fingerprint of emitters of interest limits current state-of-the-art SEI systems in numerous ways. Recent work in RF Machine Learning (RFML) and Convolutional Neural Networks (CNNs) has shown the capability to perform signal processing tasks such as modulation classification, without the need for pre-defined expert features. Given this success, the work presented in this thesis investigates the ability to use CNNs, in place of a traditional expert-defined feature extraction process, to improve upon traditional SEI systems, by developing and analyzing two distinct approaches for performing SEI using CNNs. Neither approach assumes a priori knowledge of the emitters of interest. Further, both approaches use only raw IQ data as input, and are designed to be easily tuned or modified for new operating environments. Results show CNNs can be used to both estimate expert-defined features and to learn emitter-specific features to effectively identify emitters. / Master of Science / When a device sends a signal, it unintentionally modifies the signal due to small variations and imperfections in the device’s hardware. These modifications, which are typically called the device’s radio frequency (RF) fingerprint, are unique to each device, and, generally, are independent of the data contained within the signal. The goal of a Specific Emitter Identification (SEI) system is to use these RF fingerprints to match received signals to the devices, or emitters, which sent the given signals. SEI systems are often used for military applications, and, more recently, have been used to help make more efficient use of the highly congested RF spectrum. Traditional state-of-the-art SEI systems detect the RF fingerprint embedded in each received signal by extracting one or more features from the signal. These features have been defined by experts in the field, and are determined ahead of time, in order to best capture the RF fingerprints of the emitters the system will likely encounter. However, this use of pre-determined expert features in traditional SEI systems limits the system in a variety of ways. The work presented in this thesis investigates the ability to use Machine Learning (ML) techniques in place of the typically used expert-defined feature extraction processes, in order to improve upon traditional SEI systems. More specifically, in this thesis, two distinct approaches for performing SEI using Convolutional Neural Networks (CNNs) are developed and evaluated. These approaches are designed to have no knowledge of the emitters they may encounter and to be easily modified, unlike traditional SEI systems

In this study, emitter identification techniques have been investigated and a schema has been proposed to solve the emitter identification problem in Electronic Warfare systems. Clustering technique, histogram based deinterleaving techniques and a continuous wavelet transform based deinterleaving technique have been reviewed. A receiver simulator software has been developed to test the performance of these techniques and to compare them against each other. To compensate the disadvantages of these techniques, a schema utilizing the beneficial points of them has been developed. With the modifications proposed a resultant schema has been obtained. Proposed schema uses clustering and deinterleaving together with other proposed modifications. Tests made through out this study have shown that this usage improves performance of emitter identification system. Hence, proposed schema can be used to identify the emitters in real EW systems.

The tremendous ongoing growth of wireless digital communications has raised spectrum shortage and security issues. In particular, the need for new spectrum is the main obstacle in continuing this growth. Recent studies on radio spectrum usage have shown that pre-allocation of spectrum bands to specific wireless communication applications leads to poor utilization of those allocated bands. Therefore, research into new techniques for efficient spectrum utilization is being aggressively pursued by academia, industry, and government. Such research efforts have given birth to two concepts: Cognitive Radio (CR) and Dynamic Spectrum Access (DSA) network. CR is believed to be the key enabling technology for DSA network implementation. CR based DSA (cDSA) networks utilizes white spectrum for its operational frequency bands. White spectrum is the set of frequency bands which are unoccupied temporarily by the users having first rights to the spectrum (called primary users). The main goal of cDSA networks is to access of white spectrum. For proper access, CR nodes must identify the right cDSA network and the absence of primary users before initiating radio transmission. To solve the cDSA network access problem, methods are proposed to design unique second-order cyclic features using Orthogonal Frequency Division Multiplexing (OFDM) pilots. By generating distinct OFDM pilot patterns and measuring spectral correlation characteristics of the cyclostationary OFDM signal, CR nodes can detect and uniquely identify cDSA networks. For this purpose, the second-order cyclic features of OFDM pilots are investigated analytically and through computer simulation. Based on analysis results, a general formula for estimating the dominant cycle frequencies is developed. This general formula is used extensively in cDSA network identification and OFDM signal detection, as well as pilot pattern estimation. CR spectrum awareness capability can be enhanced when it can classify the modulation type of incoming signals at low and varying signal-to-noise ratio. Signal classification allows CR to select a suitable demodulation process at the receiver and to establish a communication link. For this purpose, a threshold-based technique is proposed which utilizes cycle-frequency domain profile for signal detection and feature extraction. Hidden Markov Models (HMMs) are proposed for the signal classifier. The spectrum awareness capability of CR can be undermined by spoofing radio nodes. Automatic identification of malicious or malfunctioning radio signal transmitters is a major concern for CR information assurance. To minimize the threat from spoofing radio devices, radio signal fingerprinting using second-order cyclic features is proposed as an approach for Specific Emitter Identification (SEI). The feasibility of this approach is demonstrated through the identification of IEEE 802.11a/g OFDM signals from different Wireless Local Area Network (WLAN) card manufactures using HMMs. / Ph. D.

Radar technology is widely used in today´s society, whether it is the localisation and identification of aircraft in air traffic control systems, ships in harbour management systems, or the weather forecast presented on the news. In military applications, such as in fighter jets or missile lock-on systems, the speed at which the radar processes incoming data is essential to ensure a successful outcome. These applications need more powerful execution platforms, such as Application-Specific Integrated Circuits (ASICs) or Field-Programmable Gate Array (FPGAs) to run at the required speed. Radar technology manufacturers face both high development costs and long development cycles when designing, redesigning and developing these powerful execution platforms. A solution to these problems could be to design and develop the system in software before implementing it on the specialised ASIC and FPGA ships. As a step toward creating a software-based processing chain for radar systems, this thesis amide to investigate whether General-Purpose Graphical Processing Units (GPUs) could be used to develop and run beamforming applications in real-time at a sample rate of 100 MHz. To achieve this goal, the Bartlett and MVDR Beamforming algorithms were analysed and implemented in the Computer Unified Device Architecture (CUDA) programming framework on an NVIDIA GeForce RTX 2080 super GPU. The algorithms´ parallelisable elements were considered before implementation, and then implemented to fit with the CUDA programming model. In the end, both beamformers´ primary acceleration methods is using the cuBLAS library for linear algebra operations. The results of this study show that beamforming in real-time is not possible on a GPU when processing a signal sampled at 100 MHz. However, the GPU used just passed its third anniversary since release, and NVIDIA has released subsequent generations of GPUs with significant (>100%) performance increases. Future work on this area could include trying the implementations on more recent GPUs or implementing other beamforming algorithms on GPUs. / Radarteknik används inom flera områden idag. Bland annat för lokalisering och identifiering av flygplan och fartyg i flyglednings- och hamnledningssystem, eller för att skapa väderprognosen som presenteras på nyheterna. För militära tillämpningar, till exempel i stridsflygplan eller missillåsningssystem, är behandlingshastigheten av inkommande data avgörande för att korrekt information ska visas. Dessa användningsområden kräver kraftfulla chip, såsom ASICs eller FPGAer, för att uppnå den exekveringshastighet, kompakthet och energieffektivitet som krävs. Såväl höga utvecklingskostnader som långa utvecklingscykler uppstår för radartillverkare när de designar och utvecklar dessa kraftfulla chip. En lösning på de långa cyklerna och höga kostnaderna kan vara att designa och utveckla systemet i mjukvara på en högre abstraktionsnivå innan det implementeras på de specialiserade ASIC- och FPGA-chipen. I ett steg mot att skapa en mjukvarubaserad signalbehandligsskedja gör radarsystem, syftar detta examensarbete till att undersöka om GPUer skulle kunna användas för att utveckla och exekvera beamformingapplikationer i realtid med en samplingsfrekvens på 100 MHz. För att uppnå detta mål analyserades och  implementerades beamformingalgoritmerna Bartlett och MVDR i CUDA på en NVIDIA GeForce RTX 2080 Super GPU. De element av algoritmerna som går att parallellisera undersöktes, följt av implementation i CUDAs programmeringsmodell. Den primära accelerationsmetoden för båda beamformingalgoritmerna är att använda cuBLAS-biblioteket för linjär algebraoperationer. Resultaten av denna studie visa att beamforming i realtid inte är möjligt på en GPU vid behandling av en signal samplad i 100 MHz. Den GPU som användes har nyligen passerat sin tredje årsdag sedan lansering, och NVIDA har släppt efterföljande GPU-generationer med avsevärda (>100%) prestandaökningar- I framtida arbeten på området föreslås att testa implementationerna på nyare GPUer, implementera beamformers i t.ex. OpenGL, implementera andra beamformingalgoritmer i CUDA.