30 accepted papers
We establish a theoretical connection between wavelet transforms and spiking neural networks through scale-space theory. We rely on the scale-covariant guarantees in the leaky integrate-and-fire neurons to implement discrete mother wavelets that approximate continuous wavelets. A reconstruction expe…
This paper focuses on improving 3D sound mapping using acoustic simultaneous localization and mapping (SLAM), angle of arrival (AOA), and time difference of arrival (TDOA). We use quaternion for orientation tracking. Through simulations involving a drone equipped with microphone arrays and an inerti…
A log-likelihood gridless sparse direction-of-arrival (DOA) estimation is presented. The likelihood fit is optimized using the sample covariance matrix and a reconstructed covariance matrix constrained to a few atoms. This approach enables using Wirtinger gradients for DOA. The sensitivity to local…
Representing variable-length and continuous-indexed signals through a linear combination of basis functions poses a fundamental challenge in science and engineering. Current approaches resort to preprocessing steps, such as interpolation and extrapolation, to handle irregular and off-grid measuremen…
Predicting ocean acoustic fields in shallow water is challenging due to high spatial variability, with depth scales of 100 m and range scales of 1 km. Limited acoustic data further complicates this task. We propose a physics-informed neural network (PINN) with the Helmholtz equation as a physics con…
Large Language Model (LLMs) can be used to write or modify documents, presenting a challenge for understanding the intent behind their use. For example, benign uses may involve using LLM on a human-written document to improve its grammar or to translate it into another language. However, a document
This study investigates adversarial attacks on deep learning (DL)-enabled Wi-Fi sensing systems using channel state information (CSI) for privacy. This paper presents a technique to disturb the signal used for channel estimation transmitted from the user device when the classifier is located at the…
We propose a data-driven method for direction-of-arrival (DOA) trajectory estimation. We use a deep complex architecture which leverages complex-valued representations to capture both magnitude and phase information in the received sensor array data. The network is designed to output the DOA traject…
We present a sample-efficient Bayesian optimization (BO) method to estimate underwater source localization robust to unknown tilt in a vertical line array. Rather than conducting exhaustive search of parameter space to estimate localization and tilt, BO uses a Gaussian process (GP) surrogate model o…
We propose an approach to fuse multiresolution seismic tomography models with physics-informed probability graphical models (PIPGMs), which consider the physical information (ray-path density). To evaluate the efficacy of the PIPGM fusion method, we use both synthetic checkerboard models and real fa…
We consider uncertainty quantification (UQ) for multiple DOAs in an acoustic environment. The performance of DOA estimation is affected due to external uncertainty and yet the methods provide no UQ for the DOAs. Conformal prediction obtains statistically valid prediction intervals from an estimation…
Gridless direction-of-arrival (DOA) estimation with multiple frequencies can be applied to acoustic source localization. We formulate this as an atomic norm minimization (ANM) problem and derive a regularization-free semi-definite program (SDP) avoiding regularization bias. We also propose a fast SD…
Gaussian processes (GP’s) have been used to predict acoustic fields by interpolating under-sampled field observations. Using GP interpolation to predict fields is advantageous because of its ability to denoise measurements and for its prediction of likely field outcomes given a certain field coheren…
The physics-informed neural network (PINN) is able to identify partial differential equation (PDE) coefficients which are constant across the space directly from physical measurements. In this paper, we propose a modification of PINN, named as SD-PINN, which can recover the coefficients in spatially…
Recent investigations indicate that Sparse Bayesian Learning (SBL) is lacking in robustness. We derive a robust and sparse Direction of Arrival (DOA) estimation framework based on the assumption that the array data has a centered (zero-mean) complex elliptically symmetric (ES) distribution with fini…
We propose a data-driven partial differential equation (PDE) identification scheme based on ℓ <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</inf> -norm minimization which can identify spatially-dependent PDEs from measurements. Spatially-dependent P…
Direction of Arrival (DOA) estimation is widely applied in acoustic source localization. A multi-frequency model is suitable for characterizing the broadband structure in acoustic signals. In this work, we solve the continuous (gridless) line spectrum estimation problem by incorporating the multi-fr…
We present a sparsity-promoting method for the detection and estimation of the directions of arrival (DOAs) of source signals. The proposed method is based on the recently introduced variational Bayesian line spectral estimation (VALSE) approach, which is gridless. However, the performance of VALSE…
Echo State Networks (ESNs) are a variant of recurrent neural networks (RNNs). ESNs perform as nonlinear fading memory filters and excel in prediction of "chaotic" signals. Predictions are made using a forced nonlinear dynamical system called the "reservoir" which incorporates past information into n…
Machine learning (ML) approaches to source localization have demonstrated promising results in addressing reverberation. Even with large data volumes, the number of labels available for supervised learning in such environments is usually small. This challenge has recently been addressed using semi-s…
We present a gridless sparse iterative covariance-based estimation method based on alternating projections for direction-of-arrival (DOA) estimation. The gridless DOA estimation is formulated in the reconstruction of Toeplitz-structured low rank matrix, and is solved efficiently with alternating pro…
Reservoir computers are a fast training variant of recurrent neural networks, excelling at approximation of nonlinear dynamical systems and time series prediction. These machine learning models act as self-organizing nonlinear fading memory filters. While these models benefit from low overall comple…
This paper presents SSLIDE, Sound Source Localization for Indoors using DEep learning, which applies deep neural networks (DNNs) with encoder-decoder structure to localize sound sources with random positions in a continuous space. The spatial features of sound signals received by each microphone are…
This paper introduces compressive sensing (CS) based two-dimensional (2-D) off-grid direction-of-arrival (DOA) estimation approach which can output the azimuths and elevations of radiating sources for propeller tip vortex cavitation localization. With a discretized angular search-grid of the convent…
Sparse arrays such as co-prime and nested arrays can identify more sources than the number of sensors. This is because their difference co-arrays contain a uniformly spaced virtual array with more elements than the number of sensors in the array. In this paper we demonstrate this using two dimension…
An alternative method for solving the gridless direction-of-arrival (DOA) estimation problem is presented. Gridless DOA estimation involves solving the semi-definite characterization of a rank minimization problem. We show that the original non-convex formulation of the gridless DOA estimation probl…
We develop a 2D travel time tomography method which regularizes the inversion by modeling sparsely patches of slowness pixels from discrete slowness map, and adapts sparse dictionaries to the slowness data. This locally-sparse travel time tomography (LST) approach considers global and local behavior…
The paper considers direction of arrival (DOA) estimation from long-term observations in a noisy environment. In such an environment the noise source might evolve, causing the stationary models to fail. Therefore a heteroscedastic Gaussian noise model is introduced where the variance can vary across…
Densely sampled dynamic geophysical data are often modeled using principal components analysis (PCA, a.k.a. empirical orthogonal function or EOF analysis) to provide constraints for their inversion with remote sensing techniques. We show that overcomplete sparsifying dictionaries, generated using di…
Sparse Bayesian learning is a sparse processing method used for solving high-dimensional, underdetermined linear equations. Often the sensing matrix in the system of equations is assumed known and in presence of perturbations in this matrix performance of sparse processing degrades. We develop a spa…