Search

Reliable and rapid non-invasive testing has become essential for COVID-19 diagnosis and tracking statistics. Recent studies motivate the use of modern machine learning (ML) and deep learning (DL) tools that utilize features of coughing sounds for COVID-19 diagnosis. In this paper, we describe system designs that we developed for COVID-19 cough detection with the long-term objective of embedding them in a testing device. More specifically, we use log-mel spectrogram features extracted from the coughing audio signal and design a series of customized deep learning algorithms to develop fast and automated diagnosis tools for COVID-19 detection. We first explore the use of a deep neural network with fully connected layers. Additionally, we investigate prospects of efficient implementation by examining the impact on the detection performance by pruning the fully connected neural network based on the Lottery Ticket Hypothesis (LTH) optimization process. In general, pruned neural networks have been shown to provide similar performance gains to that of unpruned networks with reduced computational complexity in a variety of signal processing applications. Finally, we investigate the use of convolutional neural network architectures and in particular the VGG-13 architecture which we tune specifically for this application. Our results show that a unique ensembling of the VGG-13 architecture trained using a combination of binary cross entropy and focal losses with data augmentation significantly outperforms the fully connected networks and other recently proposed baselines on the DiCOVA 2021 COVID-19 cough audio dataset. Our customized VGG-13 model achieves an average validation AUROC of 82.23% and a test AUROC of 78.3% at a sensitivity of 80.49%. [ABSTRACT FROM AUTHOR]

Solar array management and photovoltaic (PV) fault detection is critical for optimal and robust performance of solar plants. PV faults cause substantial power reduction along with health and fire hazards. Traditional machine learning solutions require large, labeled datasets which are often expensive and/or difficult to obtain. This data can be location and sensor specific, noisy, and resource intensive. In this paper, we develop and demonstrate new semi supervised solutions for PV fault detection. More specifically, we demonstrate that a little-known area of semi-supervised machine learning called positive unlabeled learning can effectively learn solar fault detection models using only a fraction of the labeled data required by traditional techniques. We further introduce a new feedback enhanced positive unlabeled learning algorithm that can increase model accuracy and performance in situations such as solar fault detection when few sensor features are available. Using these algorithms, we create a positive unlabeled solar fault detection model that can match and even exceed the performance of a fully supervised fault classifier using only 5% of the total labeled data. [ABSTRACT FROM AUTHOR]

An Android application that realizes an audible version of echolocation has been implemented. Bundled within the app are videos, notes, problems, and discussion questions, thus creating an "eModule". The eModule contains material that is relevant to several educational levels including K-12 and undergraduate engineering. By using the echolocation features of the app, at the K-12 level, teachers are able to introduce labs, and science and engineering practices into their lesson plans. By using data collection and interpretation, undergraduate students can be exposed to various concepts in signals and systems, digital signal processing, and machine learning. Assessment instruments for the evaluation of the Reflections app, as well as the bundled content, have been developed. Pre/post assessment data and results of survey instruments have been collected and analyzed in an effort to assess the impact of the tool on student learning at the undergraduate level, as well as impacts on K-12 teacher interest in introducing science and engineering practices in their lesson plans. [ABSTRACT FROM AUTHOR]

The anniversary of a number of significant signal processing algorithms from the 1960s, including the least mean square algorithm and the Kalman filter, provided an opportunity at ICASSP 2019 to reflect on the links between education and innovation. This led ultimately to the proposal of some special sessions as well a panel session that would provide some insight, via a historical perspective, consideration of the current status, and an assessment of the emerging educational future. [ABSTRACT FROM AUTHOR]

In this paper, we present a unique Android-DSP (AJDSP) application which was built from the ground up to provide mobile laboratory and computational experiences for educational use. AJDSP provides a mobile intuitive environment for developing and running signal processing simulations in a user-friendly manner. It is based on a block diagram system approach to support signal generation, analysis, and processing. AJDSP is designed for use by undergraduate and graduate students and DSP instructors. Its extensive functions enable instructors and students to simulate DSP concepts including convolution, Fourier transforms, ztransform and filter design. We describe the AJDSP functions and the process of using them in computer exercises, classroom demonstrations, and undergraduate remote laboratories. We include quantitative and qualitative assessments of AJDSP at the end of this paper. We also include descriptions of outreach efforts where we used AJDSP in middle schools and high schools to demonstrate how DSP algorithms enable several applications. [ABSTRACT FROM AUTHOR]

A number of fundamental quantities in statistical signal processing and information theory can be expressed as integral functions of two probability density functions. Such quantities are called density functionals as they map density functions onto the real line. For example, information divergence functions measure the dissimilarity between two probability density functions and are useful in a number of applications. Typically, estimating these quantities requires complete knowledge of the underlying distribution followed by multidimensional integration. Existing methods make parametric assumptions about the data distribution or use nonparametric density estimation followed by high-dimensional integration. In this paper, we propose a new alternative. We introduce the concept of “data-driven basis functions”-functions of distributions whose value we can estimate given only samples from the underlying distributions without requiring distribution fitting or direct integration. We derive a new data-driven complete basis that is similar to the deterministic Bernstein polynomial basis and develop two methods for performing basis expansions of functionals of two distributions. We also show that the new basis set allows us to approximate functions of distributions as closely as desired. Finally, we evaluate the methodology by developing data-driven estimators for the Kullback-Leibler divergences and the Hellinger distance and by constructing empirical estimates of tight bounds on the Bayes error rate. [ABSTRACT FROM AUTHOR]

Distributed node counting in wireless sensor networks can be important in various applications, such as network maintenance and information aggregation. In this paper, a distributed consensus algorithm for estimating the number of nodes in a wireless sensor network in the presence of communication noise is introduced. In networks with a fusion center, counting the number of nodes can easily be done by letting each node to transmit a fixed constant value to the fusion center. In a network without a fusion center, where nodes do not know the graph structure, estimating the number of nodes is not straightforward. The proposed algorithm is based on distributed average consensus and norm estimation. Different sources of error are explicitly discussed; the Fisher information and the distribution of the final estimate are derived. Several design parameters and how they affect the performance of the algorithm are studied, which provide guidelines toward making the estimation error smaller. Simulation results corroborating the theory are also provided. [ABSTRACT FROM PUBLISHER]

A distributed consensus algorithm for estimating the maximum value of the initial measurements in a sensor network with communication noise is proposed. In the absence of communication noise, max estimation can be done by updating the state value with the largest received measurements in every iteration at each sensor. In the presence of communication noise, however, the maximum estimate will incorrectly drift and the estimate at each sensor will diverge. As a result, a soft-max approximation together with a non-linear consensus algorithm is introduced herein. A design parameter controls the tradeoff between the soft-max error and convergence speed. An analysis of this tradeoff gives a guideline toward how to choose the design parameter for the max estimate. We also show that if some prior knowledge of the initial measurements is available, the consensus process can converge faster by using an optimal step size in the iterative algorithm. A shifted non-linear bounded transmit function is also introduced for faster convergence when sensor nodes have some prior knowledge of the initial measurements. Simulation results corroborating the theory are also provided. [ABSTRACT FROM PUBLISHER]

This paper presents the use of nonlinear distributed estimation in a wireless system transmitting over channels with random gains. Specifically, we discuss the development of estimators and analytically determine their attainable variance for two conditions: 1) when full channel state information (CSI) is available at the transmitter and receiver; and 2) when only channel gain statistics and phase information are available. For the case where full CSI is available, we formulate an optimization problem to allocate power among each of the transmitting sensors while minimizing the estimate variance. We show that minimizing the estimate variance when the transmitter is operating in its most nonlinear region can be formulated in a manner very similar to optimizing sensor gains with full CSI and linear transmitters. Furthermore, we show that the solution to this optimization problem in most scenarios is approximately equivalent to one of two low-complexity power allocation systems. [ABSTRACT FROM AUTHOR]