Search

Anyone who has served as a technical program committee (TPC) chair for a conference (or program manager for a funding agency) understands that paper (or proposal panel) review assignment is a demanding job that takes a lot of time, and reviewers are rarely satisfied with the end results. This article presents signal processing tools for two critical ?mass assignment? tasks: assigning papers (or proposals) to reviewers in a way that matches reviewing expertise to scientific content while respecting the reviewers? capacity constraints and splitting accepted papers (or submitted proposals) to sessions (panels) while adhering to session (panel) capacity constraints. The basic idea is to use feature vectors to represent papers and reviewers. Features can be key words or phrases (e.g., optimization or sensor networks) or other types of attributes (e.g., timeliness). This viewpoint enables optimal assignment problem formulations that make sense from a scientific and practical point of view. While optimal solutions are hard to compute for a large number of papers and reviewers, high-quality approximate solutions of moderate complexity are developed here using familiar signal processing and optimization tools. These algorithmic solutions easily outperform days of expert manual work as demonstrated in experiments with real conference data. [ABSTRACT FROM PUBLISHER]

The scientific world is becoming more open to the public and fellow researchers. Open access publishing is becoming accepted, even if some publishers are resisting. The next step is the open code and data paradigm, which was briefly discussed in the "From the Editor" column in the November 2018 issue of IEEE Signal Processing Magazine (SPM) [1]. In this column, I follow up on this topic by sharing my experiences, best practices, and thoughts about reproducible research. [ABSTRACT FROM AUTHOR]

The articles in this special section focus on nonconvex optimization for signal processing and machine learning. Optimization is now widely recognized as an indispensable tool in signal processing (SP) and machine learning (ML). Indeed, many of the advances in these fields rely crucially on the formulation of suitable optimization models and deployment of efficient numerical optimization algorithms. In the early 2000s, there was a heavy focus on the use of convex optimization techniques to tackle SP and ML applications. This is largely due to the fact that convex optimization problems often possess favorable theoretical and computational properties and that many problems of practical interest have been shown to admit convex formulations or good convex approximations. [ABSTRACT FROM AUTHOR]

This paper provides an overview of distributed estimation-compression problems encountered with wireless sensor networks (WSN). A general formulation of distributed compression-estimation under rate constraints was introduced, pertinent signal processing algorithms were developed, and emerging tradeoffs were delineated from an information theoretic perspective. Specifically, we designed rate-constrained distributed estimators for various signal models with variable knowledge of the underlying data distributions. We proved theoretically, and corroborated with examples, that when the noise distributions are either completely known, partially known or completely unknown, distributed estimation is possible with minimal bandwidth requirements which can achieve the same order of mean square error (MSE) performance as the corresponding centralized clairvoyant estimators. [ABSTRACT FROM PUBLISHER]

This column summarizes the panel on open problems in signal processing, which took place on 5 March 2017 at the International Conference on Acoustics, Speech, and Signal Processing (ICASSP) in New Orleans, Louisiana. The goal of the panel was to draw attention to some of the challenges and open problems in various areas of signal processing and generate discussion on future research areas that can be of major significance and impact in signal processing. Five leading experts representing diverse areas within signal processing made up the panel. [ABSTRACT FROM PUBLISHER]

Digital signal processing (DSP) has been playing an increasingly important role in engine control and onboard diagnostics (OBD), a critical area of vehicle powertrain controls, to meet increasingly strict fuel efficiency requirements and emission regulations. This article gives an overview of DSP applications in this field from a practical perspective and identifies fruitful areas for DSP research and development in engine control and OBD. It also provides examples in the areas of engine misfire detection, individual cylinder fuel-air ratio imbalance detection, and engine knock detection. [ABSTRACT FROM PUBLISHER]

Fitting experiment data onto a curve is a common signal processing technique to extract data features and establish the relationship between variables. Often, we expect the curve to comply with some analytical function and then turn data fitting into estimating the unknown parameters of a function. Among analytical functions for data fitting, the Gaussian function is the most widely used one due to its extensive applications in numerous science and engineering fields. To name just a few, the Gaussian function is highly popular in statistical signal processing and analysis, thanks to the central limit theorem , and the Gaussian function frequently appears in the quantum harmonic oscillator, quantum field theory, optics, lasers, and many other theories and models in physics ; moreover, the Gaussian function is widely applied in chemistry for depicting molecular orbitals, in computer science for imaging processing, and in artificial intelligence for defining neural networks. [ABSTRACT FROM AUTHOR]

Sparse modeling for signal processing and machine learning, in general, has been at the focus of scientific research for over two decades. Among others, supervised sparsity-aware learning (SAL) consists of two major paths paved by 1) discriminative methods that establish direct input–output mapping based on a regularized cost function optimization and 2) generative methods that learn the underlying distributions. The latter, more widely known as Bayesian methods, enable uncertainty evaluation with respect to the performed predictions. Furthermore, they can better exploit related prior information and also, in principle, can naturally introduce robustness into the model, due to their unique capacity to marginalize out uncertainties related to the parameter estimates. Moreover, hyperparameters (tuning parameters) associated with the adopted priors, which correspond to cost function regularizers, can be learned via the training data and not via costly cross-validation techniques, which is, in general, the case with the discriminative methods. [ABSTRACT FROM AUTHOR]

This article discusses the contradiction between the exploding energy demand of artificial intelligence (AI) and the information and communication (ICT) industry as a whole and the parallel strong request for energy sobriety imposed by the need to mitigate the impact of climate change and the anticipated collapse of civilization as we know it. Under the form of an open reflection on the goods and evils of AI, the article raises the suggestion of a drastic change in the AI paradigm, more in phase with the vital obligation to design a more resilient society. [ABSTRACT FROM AUTHOR]

Fourier theory is the backbone of signal processing (SP) and communication engineering. It has been widely used in almost all branches of science and engineering in numerous applications since its inception. However, Fourier representations such as Fourier series (FS) and Fourier transform (FT) may not exist for some signals that fail to fulfill a predefined set of Dirichlet conditions (DCs). We note a subtle gap in explaining these conditions as available in popular SP literature. For example, original DCs require a signal to have bounded variations (BVs) over one time period for the convergence of FS, where there can be, at most, a countably infinite number of maxima and minima and, at most, a countably infinite number of discontinuities of finite magnitude. However, a large body of the literature replaces this statement with the requirements of a finite number of maxima and minima over one time period, and a finite number of finite discontinuities over one time period. Due to the latter, some signals fulfilling the original DCs are incorrectly perceived as not having convergent FS (CFS) representation. A similar problem holds in the description of DCs for the FT. Likewise, although it is easy to relate the first DC with the finite value of FS or FT coefficients, the relevance of second and third DCs, as specified in SP literature, is hard to assimilate. [ABSTRACT FROM AUTHOR]

In many modern data science problems, data are represented by a graph (network), e.g., social, biological, and communication networks. Over the past decade, numerous signal processing and machine learning (ML) algorithms have been introduced for analyzing graph structured data. With the growth of interest in graphs and graph-based learning tasks in a variety of applications, there is a need to explore explainability in graph data science. In this article, we aim to approach the issue of explainable graph data science, focusing on one of the most fundamental learning tasks, community detection, as it is usually the first step in extracting information from graphs. A community is a dense subnetwork within a larger network that corresponds to a specific function. Despite the success of different community detection methods on synthetic networks with strong modular structure, much remains unknown about the quality and significance of the outputs of these algorithms when applied to real-world networks with unknown modular structure. Inspired by recent advances in explainable artificial intelligence (AI) and ML, in this article, we present methods and metrics from network science to quantify three different aspects of explainability, i.e., interpretability, replicability, and reproducibility, in the context of community detection. [ABSTRACT FROM AUTHOR]

This article presents a systematic review on the challenges and recent progress of timing and carrier synchronization techniques for high-speed optical transmission systems using single-carrier-based coherent optical modulation formats. [ABSTRACT FROM PUBLISHER]

In this article, we describe advances in digital signal processing (DSP) techniques that enable Tb/s transmission, and software-defined flexible transponders that support adaptive modulation formats and elastic optical networks (EONs). [ABSTRACT FROM PUBLISHER]

Every person spends around one third of his/her life in bed. For an infant or a young toddler, this percentage can be much higher, and for bed-bound patients it can go up to 100% of their time. In-bed pose estimation is a critical step in many human behavior monitoring systems that are focused on prevention, prediction, and management of at-rest or sleep-related conditions in both adults and children. The topic of automatic noncontact human pose estimation has received a lot of attention/success, especially in the last few years in the computer vision community, thanks to the introduction of deep learning and its power in artificial intelligence (AI) modeling. However, the state-of-the-art (SOTA) vision-based AI algorithms in this field can hardly work under the challenges associated with in-bed human behavior monitoring, such as significant illumination changes (e.g., full darkness at night), heavy occlusion (e.g., covering by a sheet or blanket), as well as the privacy concerns that mitigate large-scale data collection, necessary for any deep learning-based model training. The data quality challenges and privacy concerns have hindered the use of advanced vision-based in-bed behavior monitoring systems at home, which during the recent COVID-19 pandemic could have been an effective way to control the spread of the virus by avoiding in-person visits to clinics. [ABSTRACT FROM AUTHOR]

Antenna array technology enables the directional transmission and reception of wireless signals for communication, localization, and sensing purposes. The signal processing algorithms that underpin it began to be developed several decades ago , but it was with the deployment of 5G wireless mobile networks that the technology became mainstream. The number of antenna elements in the arrays of 5G base stations (BSs) and user devices can be measured on the order of hundreds and tens, respectively. As networks shift toward using higher-frequency bands, more antennas fit into a given aperture. For communication purposes, the arrays are harnessed to form beams in desired directions to improve the signal-to-noise ratio (SNR) and multiplex data signals in the spatial domain (to one or multiple devices) and to suppress interference by spatial filtering. For localization purposes, these arrays are employed to maintain the SNR when operating across wider bandwidths, for angle-of-arrival estimation, and to separate multiple sources and scatterers. The practical use of these features requires that each antenna array is equipped with well-designed signal processing algorithms. [ABSTRACT FROM AUTHOR]

This paper gives emphasis on the importance of testing the performance of signal processing algorithms with real data and learning from the data. The discussion is made primarily within the context of a patient monitoring research. The author suggests six steps on how to test with and learn from real data. These steps are: (1) minimize measurement uncertainty so your physiological phenomenon is consistently observable, (2) understand your data, (3) acquire a significant number of appropriate training and testing sets, (4) construct high-performance validation criteria, (5) design algorithms that can generalize in the clinical environment, and (6) test and learn from your data. Results show that this process is also applicable to other fields as well. [ABSTRACT FROM PUBLISHER]

People suffering from hearing impairment often have difficulties participating in conversations in so-called cocktail party scenarios where multiple individuals are simultaneously talking. Although advanced algorithms exist to suppress background noise in these situations, a hearing device also needs information about which speaker a user actually aims to attend to. The voice of the correct (attended) speaker can then be enhanced through this information, and all other speakers can be treated as background noise. Recent neuroscientific advances have shown that it is possible to determine the focus of auditory attention through noninvasive neurorecording techniques, such as electroencephalography (EEG). Based on these insights, a multitude of auditory attention decoding (AAD) algorithms has been proposed, which could, combined with appropriate speaker separation algorithms and miniaturized EEG sensors, lead to so-called neurosteered hearing devices. In this article, we provide a broad review and a statistically grounded comparative study of EEG-based AAD algorithms and address the main signal processing challenges in this field. [ABSTRACT FROM AUTHOR]

Deep neural networks provide unprecedented performance gains in many real-world problems in signal and image processing. Despite these gains, the future development and practical deployment of deep networks are hindered by their black-box nature, i.e., a lack of interpretability and the need for very large training sets. An emerging technique called algorithm unrolling, or unfolding, offers promise in eliminating these issues by providing a concrete and systematic connection between iterative algorithms that are widely used in signal processing and deep neural networks. Unrolling methods were first proposed to develop fast neural network approximations for sparse coding. More recently, this direction has attracted enormous attention, and it is rapidly growing in both theoretic investigations and practical applications. The increasing popularity of unrolled deep networks is due, in part, to their potential in developing efficient, high-performance (yet interpretable) network architectures from reasonably sized training sets. [ABSTRACT FROM AUTHOR]

Capturing high-dimensional (HD) data is a long-term challenge in signal processing and related fields. Snapshot compressive imaging (SCI) uses a 2D detector to capture HD (≥3D) data in a snapshot measurement. Via novel optical designs, the 2D detector samples the HD data in a compressive manner; following this, algorithms are employed to reconstruct the desired HD data cube. SCI has been used in hyperspectral imaging, video, holography, tomography, focal depth imaging, polarization imaging, microscopy, and so on. Although the hardware has been investigated for more than a decade, the theoretical guarantees have only recently been derived. Inspired by deep learning, various deep neural networks have also been developed to reconstruct the HD data cube in spectral SCI and video SCI. This article reviews recent advances in SCI hardware, theory, and algorithms, including both optimizationbased and deep learning-based algorithms. Diverse applications and the outlook for SCI are also discussed. [ABSTRACT FROM AUTHOR]

Maneuvering target tracking is an important technology in engineering applications [1]-[3]. The traditional methodologies for designing it can be divided into two categories: modelbased and model-free algorithms. Almost all tracking algorithms are model based. The main idea behind modelbased tracking algorithms is to choose a representation that fits the actual state trajectories of the target movement and then to estimate the state based on the noisy observations recorded by sensors. The Kalman filter and its extensions are the most popular methods to estimate the state of a system. However, the stability and convergence rate of these algorithms depend directly on the accurate initial state estimation, unknown parameters, and covariance matrices of the process and measurement noise. [ABSTRACT FROM AUTHOR]

Many contemporary applications in signal processing and machine learning give rise to structured nonconvex nonsmooth optimization problems that can often be tackled by simple iterative methods quite effectively. One of the keys to understanding such a phenomenon-and, in fact, a very difficult conundrum even for experts-lies in the study of "stationary points" of the problem in question. Unlike smooth optimization, for which the definition of a stationary point is rather standard, there are myriad definitions of stationarity in nonsmooth optimization. In this article, we provide an introduction to different stationarity concepts for several important classes of nonconvex nonsmooth functions, discuss the geometric interpretations of these concepts, and further clarify their relationships. We then demonstrate the relevance of these constructions in some representative applications and indicate how they could affect the performance of iterative methods for addressing these applications. [ABSTRACT FROM AUTHOR]

Zeroth-order (ZO) optimization is a subset of gradient-free optimization that emerges in many signal processing and machine learning (ML) applications. It is used for solving optimization problems similarly to gradient-based methods. However, it does not require the gradient, using only function evaluations. Specifically, ZO optimization iteratively performs three major steps: gradient estimation, descent direction computation, and the solution update. In this article, we provide a comprehensive review of ZO optimization, with an emphasis on showing the underlying intuition, optimization principles, and recent advances in convergence analysis. Moreover, we demonstrate promising applications of ZO optimization, such as evaluating robustness and generating explanations from black-box deep learning (DL) models and efficient online sensor management. [ABSTRACT FROM AUTHOR]

Phase retrieval (PR), also sometimes referred to as quadratic sensing, is a problem that occurs in numerous signal and image acquisition domains ranging from optics, X-ray crystallography, Fourier ptychography, subdiffraction imaging, and astronomy. In each of these domains, the physics of the acquisition system dictates that only the magnitude (intensity) of certain linear projections of the signal or image can be measured. Without any assumptions on the unknown signal, accurate recovery necessarily requires an overcomplete set of measurements. The only way to reduce the measurements/sample complexity is to place extra assumptions on the unknown signal/image. A simple and practically valid set of assumptions is obtained by exploiting the structure inherently present in many natural signals or sequences of signals. [ABSTRACT FROM AUTHOR]

Deep learning has proven itself to be a powerful tool to develop datadriven signal processing algorithms for challenging engineering problems. By learning the key features and characteristics of the input signals instead of requiring a human to first identify and model them, learned algorithms can beat many human-made algorithms. In particular, deep neural networks are capable of learning the complicated features of nature-made signals, such as photos and audio recordings, and using them for classification and decision making. [ABSTRACT FROM AUTHOR]

Magnetic resonance imaging (MRI) is a noninvasive diagnostic tool that provides excellent soft-tissue contrast without the use of ionizing radiation. Compared to other clinical imaging modalities (e.g., computed tomography or ultrasound), however, the data acquisition process for MRI is inherently slow, which motivates undersampling; thus, there is a need for accurate, efficient reconstruction methods from undersampled data sets. In this article, we describe the use of plug-and-play (PnP) algorithms for MRI image recovery. We first describe the linearly approximated inverse problem encountered in MRI. Then, we review several PnP methods for which the unifying commonality is to iteratively call a denoising subroutine as one step of a larger optimization-inspired algorithm. Next, we describe how the result of the PnP method can be interpreted as a solution to an equilibrium equation, allowing convergence analysis from this perspective. Finally, we present illustrative examples of PnP methods applied to MRI image recovery. [ABSTRACT FROM AUTHOR]

The Gaussian function (GF) is widely used to explain the behavior or statistical distribution of many natural phenomena as well as industrial processes in different disciplines of engineering and applied science. For example, the GF can be used to model an approximation of the Airy disk in image processing, a laser heat source in laser transmission welding [1], practical microscopic applications [2], and fluorescence dispersion in flow cytometric deoxyribonucleic acid histograms [3]. In applied sciences, the noise that corrupts the signal can be modeled by the Gaussian distribution according to the central limit theorem. Thus, by fitting the GF, researchers can develop a sound interpretation of the corresponding process or phenomenon behavior. [ABSTRACT FROM AUTHOR]

This is an age of mobility. Phones, tablets, notebook computers, smart watches, and various other devices now supply people around the world with instant communication capabilities that were only dreamed of a generation ago. Mobility technologies are also transforming medicine, helping to improve the quality of care for people at all stages of life, giving both patients and healthcare providers deeper, more accurate insights into various medical conditions. An emerging generation of mobile and often wearable medical devices, enabled by sensors and signal processing, allow caregivers to noninvasively probe deeply inside the human body. The technologies are designed to analyze physical processes in real time and/or over defined periods to treat current conditions, detect hidden problems, and provide custom tailored treatment regimens. [ABSTRACT FROM AUTHOR]

Sequential Monte Carlo methods, also known as particle filtering, have seen an explosion of development both in theory and applications. The publication of [1] sparked huge interest in the area of sequential signal processing, particularly in sequential filtering. Ever since, the number of publications in which particle filtering plays a prominent role has continued to grow. An early reference of development is [2] and later tutorials include [3]-[9]. With particle filtering, we estimate probability density functions (pdfs) of interest by probability mass functions, whose masses are placed at randomly chosen locations (particles) and whose weights are assigned to the particles. The particle filter (PF) proposed in [1] is often called the bootstrap PF (BPF), and although it is not optimal, it is the most often used filter by practitioners. A filter that became also popular is known as the auxiliary PF (APF) and was proposed in [10]. With the APF, the objective is to generate better particles at each time step compared to those generated with the BPF, thereby improving filtering accuracy. In this article, we derive the APF from a new perspective, one based on interpreting the APF from the multiple importance sampling (MIS) paradigm. The derivation also shows its relationship with the BPF. [ABSTRACT FROM AUTHOR]

The majority of ICASSP presentations are given by professors and students. Perhaps resulting from its strategic location near a great deal of high-technology activity, the strong economy [the result of a focus on machine learning (ML) and in employment therein] and because of the energetic outreach by the organizing committee, ICASSP 2018 had an especially strong industry presence, both in attendance and sponsorship. Calgary, a cosmopolitan city in the Canadian province of Alberta, was the conference's site and provided the perfect opportunity for a panel focused on industry feedback. What spurs the interests of both industry and government, the entities who are actually making the products and know the competition? How can academia participate (and maybe even help)? [ABSTRACT FROM AUTHOR]

Presents corrections to the paper, ?Perfecting protection for interactive multimedia: A survey of forward errror correction for low-delay interactive applications,? (Badr, A. et al), IEEE Signal Process. Mag., vol. 34, no. 2, pp. 95?113, Mar. 2017. [ABSTRACT FROM PUBLISHER]

The knowledge of spatial distributions of physical quantities, such as radio-frequency (RF) interference, pollution, geomagnetic field magnitude, temperature, humidity, audio, and light intensity, will foster the development of new context-aware applications. For example, knowing the distribution of RF interference might significantly improve cognitive radio systems [1], [2]. Similarly, knowing the spatial variations of the geomagnetic field could support autonomous navigation of robots (including drones) in factories and/or hazardous scenarios [3]. Other examples are related to the estimation of temperature gradients, detection of sources of RF signals, or percentages of certain chemical components. As a result, people could get personalized health-related information based on their exposure to sources of risks (e.g., chemical or pollution). We refer to these spatial distributions of physical quantities as spatial fields. All of the aforementioned examples have in common that learning the spatial fields requires a large number of sensors (agents) surveying the area [4], [5]. [ABSTRACT FROM AUTHOR]

Determining the make and model of the camera that captured an image has been an important research area in information forensics for more than a decade [1]-[3] (see Figure 1). Information about which type of camera captured an image can be used to help determine or verify the origin of an image and can form an important piece of evidence in some scenarios, such as analyzing images involved in child exploitation investigations. While metadata may contain information about an image's source camera, metadata is both easy to falsify and frequently missing from an image. As a result, signal processing researchers have developed information forensic algorithms that can exploit traces intrinsically present in the image itself. [ABSTRACT FROM AUTHOR]

Full waveform inversion (FWI) is an advanced seismic processing method for high-resolution subsurface geophysical property reconstruction through a data-fitting procedure. When low-frequency seismic data are not available, conventional FWI often suffers from the cycle-skipping issue caused by the severe nonlinearity nature of the standard L2 norm objective function, inducing strong artifacts in the reconstructed geophysical property models. [ABSTRACT FROM PUBLISHER]

This article discusses how to estimate the Fourier spectra of irregularly sampled multidimensional functions in an approximate way, using current fast Fourier transform (FFT) algorithms. This estimate may be an alternative for more rigorous approaches when the inversion of huge matrices is prohibitively expensive. The approximation results from a Taylor expansion of the Fourier transform kernel, which is a series on the product of wavenumbers and displacements from centers of a grid where a regular, discrete Fourier transform (DFT) is defined. Convergence and efficiency, as well as some shortcuts for implementation, is indicated. The problem of finding a Fourier spectrum of a function given a finite set of irregular measurements is usually associated with the idea of regularization/interpolation. This, in turn, raises the question of representativeness of continuous functions via its discrete measurements. Although this question is central for the very Fourier spectrum estimation, the scope of this article will be restricted to the trigonometric interpolation, postponing representativeness issues. For the sake of simplicity, the proposed approximation is first derived for the one-dimensional (problem where most related aspects are more clearly stated. Then, extensions to higher dimensions are straightforwardly presented. [ABSTRACT FROM PUBLISHER]

Environmental monitoring is a topic of increasing interest, especially concerning the matter of natural hazards prediction. Regarding volcanic unrest, effective methodologies along with innovative and operational tools are needed to monitor, mitigate, and prevent risks related to volcanic hazards. In general, the current approaches for volcanoes monitoring are mainly based on the manual analysis of various parameters, including gas leaps, deformations measurements, and seismic signals analysis. However, due to the large amount of data acquired by in situ sensors for long-term monitoring, manual inspection is no longer a viable option. As in many big data situations, classic machinelearning approaches are now considered to automatize the analysis of years of recorded signals, thereby enabling monitoring on a larger scale. [ABSTRACT FROM PUBLISHER]

Deep reinforcement learning (DRL) is poised to revolutionize the field of artificial intelligence (AI) and represents a step toward building autonomous systems with a higherlevel understanding of the visual world. Currently, deep learning is enabling reinforcement learning (RL) to scale to problems that were previously intractable, such as learning to play video games directly from pixels. DRL algorithms are also applied to robotics, allowing control policies for robots to be learned directly from camera inputs in the real world. In this survey, we begin with an introduction to the general field of RL, then progress to the main streams of value-based and policy-based methods. Our survey will cover central algorithms in deep RL, including the deep Q-network (DQN), trust region policy optimization (TRPO), and asynchronous advantage actor critic. In parallel, we highlight the unique advantages of deep neural networks, focusing on visual understanding via RL. To conclude, we describe several current areas of research within the field. [ABSTRACT FROM PUBLISHER]

This article presents a new entry to the class of published algorithms for the fast computation of the arctangent of a complex number. Our method uses a look-up table (LUT) to reduce computational errors. We also show how to convert a large-sized LUT addressed by two variables to an equivalent-performance smaller-sized LUT addressed by only one variable. In addition, we demonstrate how and why the use of follow-on LUTs applied to other simple arctan algorithms produce unexpected and interesting results. [ABSTRACT FROM PUBLISHER]

Inferring information from a set of acquired data is the main objective of any signal processing (SP) method. The common problem of estimating the value of a vector of parameters from a set of noisy measurements is at the core of a plethora of scientific and technological advances in recent decades, including wireless communications, radar and sonar, biomedicine, image processing, and seismology. [ABSTRACT FROM PUBLISHER]

A fundamental problem in signal processing is the estimation of unknown parameters or functions from noisy observations. Important examples include localization of objects in wireless sensor networks [1] and the Internet of Things [2]; multiple source reconstruction from electroencephalograms [3]; estimation of power spectral density for speech enhancement [4]; or inference in genomic signal processing [5]. Within the Bayesian signal processing framework, these problems are addressed by constructing posterior probability distributions of the unknowns. The posteriors combine optimally all of the information about the unknowns in the observations with the information that is present in their prior probability distributions. Given the posterior, one often wants to make inference about the unknowns, e.g., if we are estimating parameters, finding the values that maximize their posterior or the values that minimize some cost function given the uncertainty of the parameters. Unfortunately, obtaining closed-form solutions to these types of problems is infeasible in most practical applications, and therefore, developing approximate inference techniques is of utmost interest. [ABSTRACT FROM PUBLISHER]

Matching theory is a powerful tool to study the formation of dynamic and mutually beneficial relations among different types of rational and selfish agents. It has been widely used to develop high performance, low complexity, and decentralized protocols. In this article, a comprehensive survey of matching theory, its variants, and their significant properties appropriate for the demands of wireless communications and network engineers is provided. Recent research progress in applying matching theory to wireless communications to address major technical opportunities and challenges is presented. A novel classification of matching models from the practical perspective is provided, and the properties and structure of each model are explained. This will enable a network designer to select an appropriate matching model for a specific application in wireless communications. Finally, the application of different matching models to various emerging wireless networks is discussed. [ABSTRACT FROM PUBLISHER]

A number of application areas such as biomedical engineering require solving an underdetermined linear inverse problem. In such a case, it is necessary to make assumptions on the sources to restore identifiability. This problem is encountered in brain-source imaging when identifying the source signals from noisy electroencephalographic or magnetoencephalographic measurements. This inverse problem has been widely studied during recent decades, giving rise to an impressive number of methods using different priors. Nevertheless, a thorough study of the latter, including especially sparse and tensor-based approaches, is still missing. In this article, we propose 1) a taxonomy of the algorithms based on methodological considerations; 2) a discussion of the identifiability and convergence properties, advantages, drawbacks, and application domains of various techniques; and 3) an illustration of the performance of seven selected methods on identical data sets. Directions for future research in the area of biomedical imaging are eventually provided. [ABSTRACT FROM PUBLISHER]

The Kalman filter and the least mean square (LMS) adaptive filter are two of the most popular adaptive estimation algorithms that are often used interchangeably in a number of statistical signal processing applications. They are typically treated as separate entities, with the former as a realization of the optimal Bayesian estimator and the latter as a recursive solution to the optimal Wiener filtering problem. In this lecture note, we consider a system identification framework within which we develop a joint perspective on Kalman filtering and LMS-type algorithms, achieved through analyzing the degrees of freedom necessary for optimal stochastic gradient descent adaptation. This approach permits the introduction of Kalman filters without any notion of Bayesian statistics, which may be beneficial for many communities that do not rely on Bayesian methods [1], [2]. [ABSTRACT FROM PUBLISHER]

Euclidean distance matrices (EDMs) are matrices of the squared distances between points. The definition is deceivingly simple; thanks to their many useful properties, they have found applications in psychometrics, crystallography, machine learning, wireless sensor networks, acoustics, and more. Despite the usefulness of EDMs, they seem to be insufficiently known in the signal processing community. Our goal is to rectify this mishap in a concise tutorial. We review the fundamental properties of EDMs, such as rank or (non)definiteness, and show how the various EDM properties can be used to design algorithms for completing and denoising distance data. Along the way, we demonstrate applications to microphone position calibration, ultrasound tomography, room reconstruction from echoes, and phase retrieval. By spelling out the essential algorithms, we hope to fast-track the readers in applying EDMs to their own problems. The code for all of the described algorithms and to generate the figures in the article is available online at http://lcav.epfl.ch/ivan.dokmanic. Finally, we suggest directions for further research. [ABSTRACT FROM PUBLISHER]

In this article, we review previous work on biometric security under a recent framework proposed in the field of adversarial machine learning. This allows us to highlight novel insights on the security of biometric systems when operating in the presence of intelligent and adaptive attackers that manipulate data to compromise normal system operation. We show how this framework enables the categorization of known and novel vulnerabilities of biometric recognition systems, along with the corresponding attacks, countermeasures, and defense mechanisms. We report two application examples, respectively showing how to fabricate a more effective face spoofing attack, and how to counter an attack that exploits an unknown vulnerability of an adaptive face-recognition system to compromise its face templates. [ABSTRACT FROM PUBLISHER]

Two decades ago, with the publication of [1], we witnessed the rebirth of particle filtering (PF) as a methodology for sequential signal processing. Since then, PF has become very popular because of its ability to process observations represented by nonlinear state-space models where the noises of the model can be non-Gaussian. This methodology has been adopted in various fields, including finance, geophysical systems, wireless communications, control, navigation and tracking, and robotics [2]. The popularity of PF has also spurred the publication of several review articles [2]?[6]. [ABSTRACT FROM PUBLISHER]

In pattern recognition and computer vision, one is often faced with scenarios where the training data used to learn a model have different distribution from the data on which the model is applied. Regardless of the cause, any distributional change that occurs after learning a classifier can degrade its performance at test time. Domain adaptation tries to mitigate this degradation. In this article, we provide a survey of domain adaptation methods for visual recognition. We discuss the merits and drawbacks of existing domain adaptation approaches and identify promising avenues for research in this rapidly evolving field. [ABSTRACT FROM PUBLISHER]

The evolution of cellular networks is driven by the dream of ubiquitous wireless connectivity: any data service is instantly accessible everywhere. With each generation of cellular networks, we have moved closer to this wireless dream; first by delivering wireless access to voice communications, then by providing wireless data services, and recently by delivering a Wi-Fi-like experience with wide-area coverage and user mobility management. The support for high data rates has been the main objective in recent years [1], as seen from the academic focus on sum-rate optimization and the efforts from standardization bodies to meet the peak rate requirements specified in IMT-Advanced. In contrast, a variety of metrics/objectives are put forward in the technological preparations for fifth-generation (5G) networks: higher peak rates, improved coverage with uniform user experience, higher reliability and lower latency, better energy efficiency (EE), lower-cost user devices and services, better scalability with number of devices, etc. These multiple objectives are coupled, often in a conflicting manner such that improvements in one objective lead to degradation in the other objectives. Hence, the design of future networks calls for new optimization tools that properly handle the existence of multiple objectives and tradeoffs between them. [ABSTRACT FROM PUBLISHER]

Cloud computing draws significant attention in the information technology (IT) community as it provides ubiquitous on-demand access to a shared pool of configurable computing resources with minimum management effort. It gains also more impact on the communication technology (CT) community and is currently discussed as an enabler for flexible, cost-efficient and more powerful mobile network implementations. Although centralized baseband pools are already investigated for the radio access network (RAN) to allow for efficient resource usage and advanced multicell algorithms, these technologies still require dedicated hardware and do not offer the same characteristics as cloud-computing platforms, i.e., on-demand provisioning, virtualization, resource pooling, elasticity, service metering, and multitenancy. However, these properties of cloud computing are key enablers for future mobile communication systems characterized by an ultradense deployment of radio access points (RAPs) leading to severe multicell interference in combination with a significant increase of the number of access nodes and huge fluctuations of the rate requirements over time. In this article, we will explore the benefits that cloud computing offers for fifth-generation (5G) mobile networks and investigate the implications on the signal processing algorithms. [ABSTRACT FROM PUBLISHER]

This article reviews recent advances in convex optimization algorithms for big data, which aim to reduce the computational, storage, and communications bottlenecks. We provide an overview of this emerging field, describe contemporary approximation techniques such as first-order methods and randomization for scalability, and survey the important role of parallel and distributed computation. The new big data algorithms are based on surprisingly simple principles and attain staggering accelerations even on classical problems. [ABSTRACT FROM PUBLISHER]