Search

We investigate the invertible and non-invertible symmetries of topological finite group gauge theories in general spacetime dimensions, where the gauge group can be Abelian or non-Abelian. We focus in particular on the 0-form symmetry. The gapped domain walls that generate these symmetries are specified by boundary conditions for the gauge fields on either side of the wall. We investigate the fusion rules of these symmetries and their action on other topological defects including the Wilson lines, magnetic fluxes, and gapped boundaries. We illustrate these constructions with various novel examples, including non-invertible electric-magnetic duality symmetry in 3+1d $\mathbb{Z}_2$ gauge theory, and non-invertible analogs of electric-magnetic duality symmetry in non-Abelian finite group gauge theories. In particular, we discover topological domain walls that obey Fibonacci fusion rules in 2+1d gauge theory with dihedral gauge group of order 8. We also generalize the Cheshire string defect to analogous defects of general codimensions and gauge groups and show that they form a closed fusion algebra., Comment: 55 pages, 20 figures. V2: fixed figures misprints and typos

We introduce a mirror twin Higgs model with spontaneous $\mathbb{Z}_2$ symmetry breaking that ameliorates the constraints in twin Higgs cosmology and, at the same time, generates the Standard Model neutrino masses. The model features an $SU(2)$ triplet with hypercharge $1$ alongside its twin counterpart. Spontaneous breaking of both $\mathbb{Z}_2$ and electroweak symmetry occurs in the scalar sector. The Standard Model neutrinos acquire small masses through the type-II seesaw mechanism. In contrast, their twin counterparts acquire large masses, effectively addressing the dark radiation problem in mirror twin Higgs scenarios. We study the impact of the model on the $N_{\rm eff.}$ constraints, as well as on collider phenomenology., Comment: 26 pages, 3 figures

This work presents an unsupervised method for automatically constructing and expanding topic taxonomies using instruction-based fine-tuned LLMs (Large Language Models). We apply topic modeling and keyword extraction techniques to create initial topic taxonomies and LLMs to post-process the resulting terms and create a hierarchy. To expand an existing taxonomy with new terms, we use zero-shot prompting to find out where to add new nodes, which, to our knowledge, is the first work to present such an approach to taxonomy tasks. We use the resulting taxonomies to assign tags that characterize merchants from a retail bank dataset. To evaluate our work, we asked 12 volunteers to answer a two-part form in which we first assessed the quality of the taxonomies created and then the tags assigned to merchants based on that taxonomy. The evaluation revealed a coherence rate exceeding 90% for the chosen taxonomies. The taxonomies' expansion with LLMs also showed exciting results for parent node prediction, with an f1-score above 70% in our taxonomies.

Objective To solve major clinical natural language processing (NLP) tasks using a unified text-to-text learning architecture based on a generative large language model (LLM) via prompt tuning. Methods We formulated 7 key clinical NLP tasks as text-to-text learning and solved them using one unified generative clinical LLM, GatorTronGPT, developed using GPT-3 architecture and trained with up to 20 billion parameters. We adopted soft prompts (i.e., trainable vectors) with frozen LLM, where the LLM parameters were not updated (i.e., frozen) and only the vectors of soft prompts were updated, known as prompt tuning. We added additional soft prompts as a prefix to the input layer, which were optimized during the prompt tuning. We evaluated the proposed method using 7 clinical NLP tasks and compared them with previous task-specific solutions based on Transformer models. Results and Conclusion The proposed approach achieved state-of-the-art performance for 5 out of 7 major clinical NLP tasks using one unified generative LLM. Our approach outperformed previous task-specific transformer models by ~3% for concept extraction and 7% for relation extraction applied to social determinants of health, 3.4% for clinical concept normalization, 3.4~10% for clinical abbreviation disambiguation, and 5.5~9% for natural language inference. Our approach also outperformed a previously developed prompt-based machine reading comprehension (MRC) model, GatorTron-MRC, for clinical concept and relation extraction. The proposed approach can deliver the ``one model for all`` promise from training to deployment using a unified generative LLM.

Scarcity of domain-specific data in the Portuguese financial domain has disfavored the development of Natural Language Processing (NLP) applications. To address this limitation, the present study advocates for the utilization of synthetic data generated through data augmentation techniques. The investigation focuses on the augmentation of a dataset sourced from the Central Bank of Brazil FAQ, employing techniques that vary in semantic similarity. Supervised and unsupervised tasks are conducted to evaluate the impact of augmented data on both low and high semantic similarity scenarios. Additionally, the resultant dataset will be publicly disseminated on the Hugging Face Datasets platform, thereby enhancing accessibility and fostering broader engagement within the NLP research community.

There is enormous enthusiasm and concerns in using large language models (LLMs) in healthcare, yet current assumptions are all based on general-purpose LLMs such as ChatGPT. This study develops a clinical generative LLM, GatorTronGPT, using 277 billion words of mixed clinical and English text with a GPT-3 architecture of 20 billion parameters. GatorTronGPT improves biomedical natural language processing for medical research. Synthetic NLP models trained using GatorTronGPT generated text outperform NLP models trained using real-world clinical text. Physicians Turing test using 1 (worst) to 9 (best) scale shows that there is no significant difference in linguistic readability (p = 0.22; 6.57 of GatorTronGPT compared with 6.93 of human) and clinical relevance (p = 0.91; 7.0 of GatorTronGPT compared with 6.97 of human) and that physicians cannot differentiate them (p < 0.001). This study provides insights on the opportunities and challenges of LLMs for medical research and healthcare.

Aiming at overcoming practical issues of successive interference cancellation (SIC), this paper proposes a dual-polarized rate-splitting multiple access (RSMA) technique for a downlink massive multiple-input multiple-output (MIMO) network. By modeling the effects of polarization interference, an in-depth theoretical analysis is carried out, in which we derive tight closed-form approximations for the outage probabilities and ergodic sum-rates. Simulation results validate the accuracy of the theoretical analysis and confirm the effectiveness of the proposed approach. For instance, under low to moderate cross-polar interference, our results show that the proposed dual-polarized MIMO-RSMA strategy outperforms the single-polarized MIMO-RSMA counterpart for all considered levels of residual SIC error., Comment: arXiv admin note: substantial text overlap with arXiv:2211.00851

The potential of Reconfigurable Intelligent Surfaces (RISs) for energy-efficient and performance-boosted wireless communications is recently gaining remarkable research attention, motivating their consideration for various $5$-th Generation (5G) Advanced and beyond applications. In this paper, we consider a Multiple-Input Multiple-Output (MIMO) Physical Layer Security (PLS) system with multiple data streams including one legitimate passive RIS and one malicious passive RIS, with the former being transparent to the multi-antenna eavesdropper and the latter's presence being unknown at the legitimate multi-antenna transceivers. We first present a novel threat model for the RIS-boosted eavesdropping system and design a joint optimization framework for the eavesdropper's receive combining matrix and the reflection coefficients of the malicious RIS. Focusing next on the secrecy rate maximization problem, we present an RIS-empowered PLS scheme that jointly designs the legitimate precoding matrix and number of data streams, the Artificial Noise (AN) covariance matrix, the receive combining matrix, and the reflection coefficients of the legitimate RIS. The proposed optimization algorithms, whose convergence to at least local optimum points is proved, are based on alternating maximization, minorization-maximization, and manifold optimization, including semi-closed form expressions for the optimization variables. Our extensive simulation results for two representative system setups reveal that, in the absence of a legitimate RIS, transceiver spatial filtering and AN are incapable of offering non-zero secrecy rates, even for malicious RISs with small numbers of elements. However, when an $L$-element legitimate RIS is deployed, confidential communication can be safeguarded against eavesdropping systems possessing even more than a $5L$-element malicious RIS., Comment: 33 pages, 8 figures, submitted to an IEEE Journal

Animations have become increasingly realistic with the evolution of Computer Graphics (CG). In particular, human models and behaviors were represented through animated virtual humans, sometimes with a high level of realism. In particular, gender is a characteristic that is related to human identification, so that virtual humans assigned to a specific gender have, in general, stereotyped representations through movements, clothes, hair and colors, in order to be understood by users as desired by designers. An important area of study is finding out whether participants' perceptions change depending on how a virtual human is visually presented. Findings in this area can help the industry to guide the modeling and animation of virtual humans to deliver the expected impact to the audience. In this paper, we reproduce, through CG, a perceptual study that aims to assess gender bias in relation to a simulated baby. In the original study, two groups of people watched the same video of a baby reacting to the same stimuli, but one group was told the baby was female and the other group was told the same baby was male, producing different perceptions. The results of our study with virtual babies were similar to the findings with real babies. First, it shows that people's emotional response change depending on the character gender attribute, in this case the only difference was the baby's name. Our research indicates that by just informing the name of a virtual human can be enough to create a gender perception that impact the participant emotional answer., Comment: 8 pages, 1 figure, 2 tables

Rate-splitting multiple access (RSMA) has recently appeared as a powerful technique for improving the downlink performance of multiple-input multiple-output (MIMO) systems. By flexibly managing interference, RSMA can deliver high spectral and energy efficiency, as well as robustness to imperfect channel state information (CSI). In another development, an intelligent reflecting surface (IRS) has emerged as a method to control the wireless environment through software-configurable, near-passive, sub-wavelength reflecting elements. This article presents the potential of synergy between IRS and RSMA. Three important improvements achievable by IRS-RSMA schemes are identified, supported by insightful numerical examples, and mapped to beyond-5G use cases, along with future research directions., Comment: IEEE Communications Magazine, May 2022

We propose and investigate a model for mate searching and marriage in large societies based on a stochastic matching process and simple decision rules. Agents have preferences among themselves given by some probability distribution. They randomly search for better mates, forming new couples and breaking apart in the process. Marriage is implemented in the model by adding the decision of stopping searching for a better mate when the affinity between a couple is higher than a certain fixed amount. We show that the average utility in the system with marriage can be higher than in the system without it. Part of our results can be summarized in what sounds like a piece of advice: don't marry the first person you like and don't search for the love of your life, but get married if you like your partner more than a sigma above average. We also find that the average utility attained in our stochastic model is smaller than the one associated with a stable matching achieved using the Gale-Shapley algorithm. This can be taken as a formal argument in favor of a central planner (perhaps an app) with the information to coordinate the marriage market in order to set a stable matching. To roughly test the adequacy of our model to describe existent societies, we compare the evolution of the fraction of married couples in our model with real-world data and obtain good agreement. In the last section, we formulate the model in the limit of an infinite number of agents and find an analytical expression for the evolution of the system., Comment: 34 pages, 21 figures

In this paper, the appealing features of a dual-polarized intelligent reflecting surface (IRS) are exploited to improve the performance of dual-polarized massive multiple-input multiple-output (MIMO) with non-orthogonal multiple access (NOMA) under imperfect successive interference cancellation (SIC). By considering the downlink of a multi-cluster scenario, the IRSs assist the base station (BS) to multiplex subsets of users in the polarization domain. Our novel strategy alleviates the impact of imperfect SIC and enables users to exploit polarization diversity with near-zero inter-subset interference. Our results show that when the IRSs are large enough, the proposed scheme always outperforms conventional massive MIMO-NOMA and MIMO-OMA systems even if SIC error propagation is present. It is also confirmed that dual-polarized IRSs can make cross-polar transmissions beneficial to the users, allowing them to improve their performance through polarization diversity., Comment: arXiv admin note: substantial text overlap with arXiv:2012.03639

Turbidity is a physical property related to the scattering of light by particles that are suspended in a liquid. Commercial turbidimeters are priced at the range of hundreds to thousands of dollars. Considering this scenario, it is proposed in this work the development of a low-cost portable turbidimeter for monitoring of turbidity in processes. An infrared LED was used as light emitter, and an infrared phototransistor as light receiver. The signal processing control unit was developed with the Arduino Uno platform. The calibration of the turbidimeter was done by means of a comparative test in triplicate, using as reference the commercial turbidimeter 2100P, HACH. The turbidimeter was able to perform analysis in the range of 100 to 1000 NTU, presenting an innovation character given its portability and computer communication via USB, and in a good price range for the prototype, costing US$ 46.30., Comment: 14 pages, 8 figures, 1 table

Despite the immense progress in the recent years, efficient solutions for monitoring remote areas are still missing today. This is especially notable in the context of versatile maritime and offshore use cases, owing to a broader span of operating regions and a lack of radio network infrastructures. In this article, we address the noted challenge by delivering a conceptual solution based on the convergence of three emerging technologies -- unmanned aerial vehicles (UAVs), battery-less sensors, and wireless powered communication networks (WPCNs). Our contribution offers a systematic description of the ecosystem related to the proposed solution by identifying its key actors and design dimensions together with the relevant resources and performance metrics. A system-level modeling-based evaluation of an illustrative scenario delivers deeper insights into the considered operation and the associated trade-offs. Further, unresolved challenges and perspective directions are underpinned for a subsequent study.

A dual-polarized intelligent reflecting surface (IRS) can contribute to a better multiplexing of interfering wireless users. In this paper, we use this feature to improve the performance of dual-polarized massive multiple-input multiple-output (MIMO) with non-orthogonal multiple access (NOMA) under imperfect successive interference cancellation (SIC). By considering the downlink of a multi-cluster scenario, the IRSs assist the base station (BS) to multiplex subsets of users in the polarization domain. Our novel strategy alleviates the impact of imperfect SIC and enables users to exploit polarization diversity with near-zero inter-subset interference. To this end, the IRSs are optimized to mitigate transmissions originated at the BS from the interfering polarization. The formulated optimization is transformed into quadratic constrained quadratic sub-problems, which makes it possible to obtain the optimal solution via interior-points methods. We also derive analytically a closed-form expression for the users' ergodic rates by considering large numbers of reflecting elements. This is followed by representative simulation examples and comprehensive discussions. The results show that when the IRSs are large enough, the proposed scheme always outperforms conventional massive MIMO-NOMA and MIMO-OMA systems even if SIC error propagation is present. It is also confirmed that dual-polarized IRSs can make cross-polar transmissions beneficial to the users, allowing them to improve their performance through diversity.

This paper presents a study regarding group behavior in a controlled experiment focused on differences in an important attribute that vary across cultures -- the personal spaces -- in two Countries: Brazil and Germany. In order to coherently compare Germany and Brazil evolutions with same population applying same task, we performed the pedestrian Fundamental Diagram experiment in Brazil, as performed in Germany. We use CNNs to detect and track people in video sequences. With this data, we use Voronoi Diagrams to find out the neighbor relation among people and then compute the walking distances to find out the personal spaces. Based on personal spaces analyses, we found out that people behavior is more similar, in terms of their behaviours, in high dense populations and vary more in low and medium densities. So, we focused our study on cultural differences between the two Countries in low and medium densities. Results indicate that personal space analyses can be a relevant feature in order to understand cultural aspects in video sequences. In addition to the cultural differences, we also investigate the personality model in crowds, using OCEAN. We also proposed a way to simulate the FD experiment from other countries using the OCEAN psychological traits model as input. The simulated countries were consistent with the literature., Comment: Computer Animation and Virtual Worlds, 2019

Massive multiple-input multiple-output (MIMO) and non-orthogonal multiple access (NOMA) are two key techniques for enabling massive connectivity in future wireless networks. A massive MIMO-NOMA system can deliver remarkable spectral improvements and low communication latency. Nevertheless, the uncontrollable stochastic behavior of the wireless channels can still degrade its performance. In this context, intelligent reflecting surface (IRS) has arisen as a promising technology for smartly overcoming the harmful effects of the wireless environment. The disruptive IRS concept of controlling the propagation channels via software can provide attractive performance gains to the communication networks, including higher data rates, improved user fairness, and, possibly, higher energy efficiency. In this article, in contrast to the existing literature, we demonstrate the main roles of IRSs in MIMO-NOMA systems. Specifically, we identify and perform a comprehensive discussion of the main performance gains that can be achieved in IRS-assisted massive MIMO-NOMA (IRS-NOMA) networks. We outline exciting futuristic use case scenarios for IRS-NOMA and expose the main related challenges and future research directions. Furthermore, throughout the article, we support our in-depth discussions with representative numerical results.

We construct anomaly-free $U(1)_1\times U(1)_2\times...\times U(1)_m$ gauge extensions of the Standard Model. To perform this construction we put together anomaly-free $U(1)$ extensions of one and two families of fermions. The availability of free parameters that enter linearly in the equations for the fermion charges and the large number of different classes of extensions may help other model builders interested in their use to solve problems of particle physics., Comment: 6 pages

Wireless communications empowered by Reconfigurable Intelligent (meta)Surfaces (RISs) are recently gaining remarkable research attention due to the increased system design flexibility offered by RISs for diverse functionalities. In this paper, we consider a Multiple Input Multiple Output (MIMO) physical layer security system with multiple data streams including one legitimate and one eavesdropping passive RISs, with the former being transparent to the eavesdropper and the latter's presence being unknown at the legitimate link. We first focus on the eavesdropping subsystem and present a joint design framework for the eavesdropper's combining vector and the reflection coefficients of the eavesdropping RIS. Then, focusing on the secrecy rate maximization, we propose a physical layer security scheme that jointly designs the legitimate precoding vector and the Artificial Noise (AN) covariance matrix, as well as the legitimate combining vector and the reflection coefficients of the legitimate RIS. Our simulation results reveal that, in the absence of a legitimate RIS, transceiver spatial filtering and AN are incapable of offering nonzero secrecy rates, even for eavesdropping RISs with small numbers of elements. However, when a L-element legitimate RIS is deployed, confidential communication can be safeguarded against cases with even more than a 5L-element eavesdropping RIS., Comment: 6 pages, 3 figures, submitted to an IEEE Conference

We construct chiral theories with the smallest number $n_\chi$ of Weyl fermions that form an anomaly-free set under various Abelian gauge groups. For the $U(1)$ group, where $n_\chi = 5$, we show that the general solution to the anomaly equations is a set of charges given by cubic polynomials in three integer parameters. For the $U(1) \times U(1)$ gauge group we find $n_\chi = 6$, and derive the general solution to the anomaly equations, in terms of 6 parameters. For $U(1) \times U(1) \times U(1)$ we show that $n_\chi = 8$, and present some families of solutions. These chiral gauge theories have potential applications to dark matter models, right-handed neutrino interactions, and other extensions of the Standard Model. As an example, we present a simple dark sector with a natural mass hierarchy between three dark matter components., Comment: 37 pages

As a typical example of bandwidth-efficient techniques, bit-interleaved coded modulation with iterative decoding (BICM-ID) provides desirable spectral efficiencies in various wireless communication scenarios. In this paper, we carry out a comprehensive investigation on tail-biting (TB) spatially coupled protograph (SCP) low-density parity-check (LDPC) codes in BICM-ID systems. Specifically, we first develop a two-step design method to formulate a novel type of constellation mappers, referred to as labeling-bit-partial-match (LBPM) constellation mappers, for SC-P-based BICM-ID systems. The LBPM constellation mappers can be seamlessly combined with high-order modulations, such as M-ary phase-shift keying (PSK) and M-ary quadrature amplitude modulation (QAM). Furthermore, we conceive a new bit-level interleaving scheme, referred to as variable node matched mapping (VNMM) scheme, which can substantially exploit the structure feature of SC-P codes and the unequal protection-degree property of labeling bits to trigger the wave-like convergence for TB-SC-P codes. In addition, we propose a hierarchical extrinsic information transfer (EXIT) algorithm to predict the convergence performance (i.e., decoding thresholds) of the proposed SC-P-based BICM-ID systems. Theoretical analyses and simulation results illustrate that the LBPM-mapped SC-P-based BICM-ID systems are remarkably superior to the state-of-the-art mapped counterparts. Moreover, the proposed SC-P-based BICM-ID systems can achieve even better error performance with the aid of the VNMM scheme. As a consequence, the proposed LBPM constellation mappers and VNMM scheme make the SC-P-based BICM-ID systems a favorable choice for the future-generation wireless communication systems., Comment: Accepted by IEEE Transactions on Vehicular Technology, Oct 2019

The main issue related to Software-Defined Network emulators is how to replicate real behavior in experiments. Mininet and others SDN emulators have an architecture that limits both the scope of experiments and the fidelity of networking tests. Consequently, the serialization, contention, and load of background processes may produce delays that compromise the operation of events such as transmitting a packet or completing a computation, possibly invalidating the performance evaluation of a network emulation. To address these problems, this paper presents vSDNEmul, a network emulator based on Docker container virtualization. Different from Mininet, vSDNEmul isolates each node in a container and interconnects the nodes through virtual or tunnel links. By using containers, vSDNEmul allows autonomous and flexible creation of independent network elements, resulting in more realistic emulations. This paper reports performance evaluations comparing vSDNEmul and Mininet. The results obtained with the vSDNEmul emulator are more realistic and present higher accuracy.

The anomaly cancellation equations for the $U(1)$ gauge group can be written as a cubic equation in $n-1$ integer variables, where $n$ is the number of Weyl fermions carrying the $U(1)$ charge. We solve this Diophantine cubic equation by providing a parametrization of the charges in terms of $n-2$ integers, and prove that this is the most general solution., Comment: 5 pages

Capitalizing on the recently proposed Fisher-Snedecor F composite fading model, in this letter, we investigate the sum of independent but not identically distributed (i.n.i.d.) Fisher-Snedecor F variates. First, a novel closed-form expression is derived for the moment generating function of the instantaneous signal-to-noise ratio. Based on this, the corresponding probability density function and cumulative distribution function of the sum of i.n.i.d. Fisher- Snedecor F variates are derived, which are subsequently employed in the analysis of multiple branch maximal-ratio combining (MRC). Specifically, we investigate the impact of multipath and shadowed fading on the outage probability and outage capacity of MRC based receivers. In addition, we derive exact closed-form expressions for the average bit error rate of coherent binary modulation schemes followed by an asymptotic analysis which provides further insights into the effect of the system parameters on the overall performance. Importantly, it is shown that the effect of multipath fading on the system performance is more pronounced than that of shadowing., Comment: 5 pages, 3 figures

Early results in using convolutional neural networks (CNNs) on x-rays to diagnose disease have been promising, but it has not yet been shown that models trained on x-rays from one hospital or one group of hospitals will work equally well at different hospitals. Before these tools are used for computer-aided diagnosis in real-world clinical settings, we must verify their ability to generalize across a variety of hospital systems. A cross-sectional design was used to train and evaluate pneumonia screening CNNs on 158,323 chest x-rays from NIH (n=112,120 from 30,805 patients), Mount Sinai (42,396 from 12,904 patients), and Indiana (n=3,807 from 3,683 patients). In 3 / 5 natural comparisons, performance on chest x-rays from outside hospitals was significantly lower than on held-out x-rays from the original hospital systems. CNNs were able to detect where an x-ray was acquired (hospital system, hospital department) with extremely high accuracy and calibrate predictions accordingly. The performance of CNNs in diagnosing diseases on x-rays may reflect not only their ability to identify disease-specific imaging findings on x-rays, but also their ability to exploit confounding information. Estimates of CNN performance based on test data from hospital systems used for model training may overstate their likely real-world performance.

Deep Learning methods are currently the state-of-the-art in many problems which can be tackled via machine learning, in particular classification problems. However there is still lack of understanding on how those methods work, why they work and what are the limitations involved in using them. In this chapter we will describe in detail the transition from shallow to deep networks, include examples of code on how to implement them, as well as the main issues one faces when training a deep network. Afterwards, we introduce some theoretical background behind the use of deep models, and discuss their limitations., Comment: Book chapter, in Portuguese, 31 pages

The paper presents a comprehensive closed-form performance analysis framework for multihop communications over Weibull fading channels. The analyzed scheme consists basically of multiple regenerative relays with generalized high-order quadrature amplitude modulation (M-QAM) transmissions. To take into consideration the channel fading in the mmWave range, we adopt the advocated Weibull model for its flexible ability to cover different channel conditions. The end-to-end performance is evaluated in terms of outage probability, bit error probability (BER), symbol error probability (SER), block error rate (BLER), ergodic capacity, and energy efficiency (EE). For all the metrics, we present exact closed-form expressions along with their asymptotic behavior, some in terms of generalized hypergeometric functions. Based on the obtained analytical results, we also present a practical application, we derive two BER- and EE-optimal transmit power allocation strategies, and we discuss the resulting performance gains. The exactness of our analysis is illustrated by numerical examples, and assessed via Monte-Carlo simulations for different system and channel parameters. Finally, as a secondary contribution, noting the increasing popularity of Fox's H and bivariate H functions, we provide new and generalized codes for computing these functions which are of practical utility in different contexts., Comment: Preprint V3, Submitted to IEEE Transactions on Wireless Communications

Image classification is one of the main research problems in computer vision and machine learning. Since in most real-world image classification applications there is no control over how the images are captured, it is necessary to consider the possibility that these images might be affected by noise (e.g. sensor noise in a low-quality surveillance camera). In this paper we analyse the impact of three different types of noise on descriptors extracted by two widely used feature extraction methods (LBP and HOG) and how denoising the images can help to mitigate this problem. We carry out experiments on two different datasets and consider several types of noise, noise levels, and denoising methods. Our results show that noise can hinder classification performance considerably and make classes harder to separate. Although denoising methods were not able to reach the same performance of the noise-free scenario, they improved classification results for noisy data.

The theory of wireless information and power transfer in energy constrained wireless networks has caught the interest of researchers due to its potential in increasing the lifetime of sensor nodes and mitigate the environment hazards caused by conventional cell batteries. Similarly, the advancements in areas of cooperative spectrum sharing protocols has enabled efficient use of frequency spectrum between a licensed primary user and a secondary user. In this paper, we consider an energy constrained secondary user which harvests energy from the primary signal and relays the primary signal in exchange for the spectrum access. We consider Nakagami-m fading model and propose two key protocols, namely time-splitting cooperative spectrum sharing (TS-CSS) and power-sharing cooperative spectrum sharing (PS-CSS), and derive expressions for the outage probabilities of the primary and secondary user in decode-forward and amplify-forward relaying modes. From the obtained results, it has been shown that the secondary user can carry its own transmission without adversely affecting the performance of the primary user and that PS-CSS protocol outperforms the TS-PSS protocol in terms of outage probability over a wide range of Signal to noise ratio(SNRs). The effect of various system parameters on the outage performance of these protocols have also been studied.

We assume a full-duplex (FD) cooperative network subject to hostile attacks and undergoing composite fading channels. We focus on two scenarios: \textit{a)} the transmitter has full CSI, for which we derive closed-form expressions for the \textit{average secrecy rate}; and \textit{b)} the transmitter only knows the CSI of the legitimate nodes, for which we obtain closed-form expressions for the \textit{secrecy outage probability}. We show that secure FD relaying is feasible, even under strong self-interference and in the presence of sophisticated multiple antenna eavesdropper.