Sorry, you need to enable JavaScript to visit this website.

Deep neural networks have recently demonstrated the traffic prediction capability with the time series data obtained
by sensors mounted on road segments. However, capturing spatio-temporal features of the traffic data often requires a
significant number of parameters to train, increasing computational burden. In this work we demonstrate that embedding

Categories:
8 Views

Deep neural networks have recently demonstrated the traffic prediction capability with the time series data obtained by sensors mounted on road segments. However, capturing spatio-temporal features of the traffic data often requires a significant number of parameters to train, increasing computational burden. In this work we demonstrate that embedding topological information of the road network improves the process of learning traffic features.

Categories:
27 Views

Many problems in statistical signal processing involve tracking the state of a dynamic system that evolves on a Stiefel manifold. To this aim, we introduce in this paper a novel particle filter algorithm that approximates the optimal importance function on the Stiefel manifold and is capable of handling nonlinear observation functions. To sample from the required importance function, we develop adaptations of previous MCMC algorithms.

Categories:
64 Views

We consider the sequential change-point detection problem of detecting changes that are characterized by a subspace structure. Such changes are frequent in high-dimensional streaming data altering the form of the corresponding covariance matrix. In this work we present a Subspace-CUSUM procedure and demonstrate its first-order asymptotic optimality properties for the case where the subspace structure is unknown and needs to be simultaneously estimated.

Categories:
17 Views

In recent years approximate message passing algorithms have gained a lot of attention and different versions have been proposed for coping with various system models. This paper focuses on vector approximate message passing (VAMP) for generalized linear models. While this algorithm is originally derived from a message passing point of view, we will review it from an estimation theory perspective and afterwards adapt it for a quantized compressed sensing application. Finally, numerical results are presented to evaluate the performance of the algorithm.

Categories:
30 Views

We investigate the performance of the generalized Spectral Kurtosis (SK) estimator in detecting and discriminating natural and artificial very short duration transients in the 2-bit sampling time domain Very-Long-Baseline Interferometry (VLBI) data. We demonstrate that, after a 32-bit FFT operation is performed on the 2-bit time domain voltages, these two types of transients become distinguishable from each other in the spectral domain.

Categories:
31 Views

This paper presents an empirical Bayesian method to estimate regularisation parameters in imaging inverse problems. The method calibrates regularisation parameters directly from the observed data by maximum marginal likelihood estimation, and is useful for inverse problems that are convex. A main novelty is that maximum likelihood estimation is performed efficiently by using a stochastic proximal gradient algorithm that is driven by two proximal Markov chain Monte Carlo samplers, intimately combining modern optimisation and sampling techniques.

Categories:
19 Views

In this paper, we study the capacity of UAV networks for high-accuracy localization of targets. We address the problem of designing a distributed control scheme for UAV navigation and formation based on an information-seeking criterion maximizing the target localization accuracy. Each UAV is assumed to be able to communicate and collaborate with other UAVs that are within a neighboring region, allowing for a feasible distributed solution which takes into account a trade-off between localization accuracy and speed of convergence to a suitable localization of the target.

Categories:
29 Views

Recently, we developed a robust generalization of the Gaussian quasi-likelihood ratio test (GQLRT). This generalization, called measure-transformed GQLRT (MT-GQLRT), operates by selecting a Gaussian model that best empirically fits a transformed probability measure of the data. In this letter, a plug-in version of the MT-GQLRT is developed for robust detection of a random signal in nonspherical noise. The proposed detector is derived by plugging an empirical measure-transformed noise covariance, ob- tained from noise-only secondary data, into the MT-GQLRT.

Categories:
16 Views

Pages