Abstract Bidirectional generative adversarial networks (BiGANs) and cycle generative adversarial networks (CycleGANs) are two emerging machine learning models that, up to now, have been used as generative models, i.e., to generate output data sampled from a target probability distribution. However, these models are also equipped with encoding modules, which, after weakly supervised training, could be, in principle, exploited for the extraction of hidden features from the input data. At the present time, how these extracted features could be effectively exploited for classification tasks is still an unexplored field. Hence, motivated by this consideration, in this paper, we develop and numerically test the performance of a novel inference engine that relies on the exploitation of BiGAN and CycleGAN-learned hidden features for the detection of COVID-19 disease from other lung diseases in computer tomography (CT) scans. In this respect, the main contributions of the paper are twofold. First, we develop a kernel density estimation (KDE)-based inference method, which, in the training phase, leverages the hidden features extracted by BiGANs and CycleGANs for estimating the (a priori unknown) probability density function (PDF) of the CT scans of COVID-19 patients and, then, in the inference phase, uses it as a target COVID-PDF for the detection of COVID diseases. As a second major contribution, we numerically evaluate and compare the classification accuracies of the implemented BiGAN and CycleGAN models against the ones of some state-of-the-art methods, which rely on the unsupervised training of convolutional autoencoders (CAEs) for attaining feature extraction. The performance comparisons are carried out by considering a spectrum of different training loss functions and distance metrics. The obtained classification accuracies of the proposed CycleGAN-based (resp., BiGAN-based) models outperform the corresponding ones of the considered benchmark CAE-based models of about 16% (resp., 14%).

Abstract We present a probabilistic method for classifying chest computed tomography (CT) scans into COVID-19 and non-COVID-19. To this end, we design and train, in an unsupervised manner, a deep convolutional autoencoder (DCAE) on a selected training data set, which is composed only of COVID-19 CT scans. Once the model is trained, the encoder can generate the compact hidden representation (the hidden feature vectors) of the training data set. Afterwards, we exploit the obtained hidden representation to build up the target probability density function (PDF) of the training data set by means of kernel density estimation (KDE). Subsequently, in the test phase, we feed a test CT into the trained encoder to produce the corresponding hidden feature vector, and then, we utilise the target PDF to compute the corresponding PDF value of the test image. Finally, this obtained value is compared to a threshold to assign the COVID-19 label or non-COVID-19 to the test image. We numerically check our approach’s performance (i.e. test accuracy and training times) by comparing it with those of some state-of-the-art methods.

We consider an acquisition system where a continuous image is reconstructed from a set of irregularly distributed noisy samples. Moreover, the system is affected by a random pointing jitter which makes the actual sampling positions different from the nominal ones. We develop a model for the system and derive the optimal, minimum variance unbiased (MVU) estimate. Unfortunately, the latter estimate is not practical to compute when the data size is large. Therefore, we develop a simplified, low resolution model and derive the corresponding MVU estimate, which has a drastically lower complexity. Moreover, we analyze the estimators' performance by using both theoretical analysis and simulations. Finally, we discuss the application to the data of the Photodetector Array Camera and Spectrometer (PACS) instrument, which is an infrared photometer on board the Herschel satellite.

We consider an acquisition system where a continuous, band-limited image is reconstructed from a set of irregularly distributed, noisy samples. An optimal estimator can be obtained by exploiting Least Squares, but it is not practical to compute when the data size is large. A simpler, widely used estimate can be obtained by properly rounding off the pointing information, but it is suboptimal and is affected by a bias, which may be large and thus limits its applicability. To solve this problem, we develop a mathematical model for the acquisition system, which accounts for the pointing information round off. Based on the model, we derive a novel optimal estimate, which has a manageable computational complexity and is largely immune from the bias, making it a better option than the suboptimal one. Moreover, the model opens a new, fruitful point of view on the estimation performance analysis. Finally, we consider the application of the novel estimate to the data of the Photodetector Array Camera and Spectrometer instrument. In this paper, we discuss several implementation aspects and investigate the performance by using both true and simulated data.

We consider an acquisition system constituted by an array of sensors scanning an image. Each sensor produces a sequence of readouts, called a time series. In this framework, we discuss the image estimation problem when the time series are affected by noise and by a time shift. In particular, we introduce an appropriate data model and consider the least squares (LS) estimate, showing that it has no closed form. However, the LS problem has a structure that can be exploited to simplify the solution. In particular, based on two known techniques, namely, separable nonlinear LS and alternating LS, we propose and analyze several practical estimation methods. As an additional contribution, we discuss the application of these methods to the data of the photodetector array camera and spectrometer, which is an infrared photometer onboard the Herschel satellite. In this context, we investigate the accuracy and the computational complexity of the methods, using both true and simulated data.

The quality of astrophysical images produced by means of the Generalised Least Square (GLS) approach may be degraded by the presence of artificial structures, obviously not present in the sky. This problem affects in different degrees all images produced by the instruments onboard the European Space Agency (ESA) Herschel satellite. In this paper we analyse these artifacts and introduce a method to remove them. The method is based on a post-processing of GLS image that estimates and removes the artifacts subtracting them from the original image. We find that the only drawback of this method is a slight increase of the background noise which, however, can be mitigated by detecting the artifacts and by performing the subtraction only where they are detected. The efficiency of the approach is demonstrated and quantified using simulated and real data.

We study the problem of minimising the power consumption of wireless networks when the traffic is specified. Initially we consider single-hop networks that model currently existing frequency division multiplex (FDM) and code division multiple access (CDMA) networks and discuss the power minimisation problem for these networks. Next we consider networks where multi-hop transmission is deployed and where a source is allowed to partition its outgoing information flow into several sub-flows each delivered through different multi-hop routes. We present a simple, versatile, suboptimum algorithm for the power minimisation in these Multi-hop, load partitioning (MLP) networks. Finally we carry out a simulation campaign in order to quantify the power savings that are attainable by means of the.proposed algorithm.

The impact of narrowband interference (NBI) on two ultra wideband (UWB) systems is analyzed. The two systems are impulse radio (IR) and a variation of it, termed direct-sequence IR (DS-IR). The signal-to-noise ratio (SNR) at the decision device of a correlation receiver is computed for both systems, assuming that the NBI is wide sense stationary and that the channel is frequency-selective. The SNR is expressed by means of a simple equation involving the signal and the interference spectrum. Next, a statistical model for the interference is introduced, considering the interference as the sum of a given number of sinusoidal signals with random powers and frequencies. The bit-error rate of IR and DS-IR is derived. The results are then specialized and compared with simulations in three case studies of practical interest, where the NBI is a single jammer with deterministic power and frequency, a multitone signal with random frequencies, or a grid of interfering signals with random powers.

The performance of two ultra wideband communication systems is investigated. The first system is the classical impulse radio (IR). The second is obtained by adding a direct-sequence (DS) code to the IR system, and is termed direct-sequence impulse radio (DS-IR). For both systems, binary pulse amplitude modulation is considered. The performance of IR and DS-IR is analyzed in a multiuser scenario, assuming an ideal channel and a correlation receiver. The performance analysis is original and general. It yields simple and exact formulas relating the performance to the system parameters. The analysis shows that IR suffers a performance degradation with respect to DS-IR unless the system parameters are chosen carefully. An optimum shape for the monocycle is identified, yielding an upper bound to the performance. The optimum system achieving this upper bound is shown to be, in principle, feasible for both DS-IR and IR. The degradation experienced by real systems with respect to the optimum system is also evaluated. Furthermore, it is shown that DS-IR has practical advantages with respect to IR. Simulations are included to validate the theoretical findings.

The effects of the phase noise (PN) on orthogonal frequency-division multiplex modems are evaluated. Three receivers are studied: a coherent receiver, a common phase error correction receiver (which is a receiver specially designed to combat PN) and a differential receiver. The impact of the PN on the decision signal-to-noise ratio (SNR) of each of these receivers is computed as a function of the PN spectrum. The resulting formulas are extremely simple. The theory is applicable to a wide range of PN models, and unifies and extends previous results on the topic. The conditions under which the decision SNR yields correct symbol error rate predictions are discussed. Simulations are reported that confirm the results.

The effect of time-domain and frequency-domain synchronization errors is quantified in the context of various coherently and noncoherently detected 1, 2, and 4 bits/symbol OFDM constellations, in order to demonstrate the wide applicability of the techniques proposed for mitigating the bit error rate (BER) performance degradations inflicted. A reference symbol is proposed and a range of correlation techniques are suggested for coarse and fine synchronization. Their performance is studied over time-dispersive Rayleigh fading channels, with the conclusion that the proposed synchronization techniques result in virtually unimpaired BERs over the range of wideband channels investigated in comparison to a perfectly synchronized system.