This paper presents a new method for estimating the nonstationary noise power spectral density given a noisy signal. The method is based on averaging the noisy speech power spectrum using time and frequency dependent smoothing factors. These factors are adjusted based on signal-presence probability in individual frequency bins. Signal presence is determined by computing the ratio of the noisy s...

Needless to say, the signal detection is one of the most important problems in the signal processing area for a long time, and a great deal of investigations has been done up to the present time. Most of the conventional approaches are based on the (binary) hypothesis-testing, and treat the corrupting (additive) noise as a stationary random process because stationary process is rather easy to h...

We introduce Latent Gaussian Process Regression which is a latent variable extension allowing modelling of non-stationary processes using stationary GP priors. The approach is built on extending the input space of a regression problem with a latent variable that is used to modulate the covariance function over the input space. We show how our approach can be used to model non-stationary process...

Amodified Karhunen-Loève Decomposition/Proper Orthogonal Decomposition method, named Smooth Decomposition (SD) (also named smooth Karhunen-Loève decomposition), was recently introduced to analyze stationary random signal. It is based on a generalized eigenproblem defined from the covariance matrix of the random process and the covariance matrix of the associated time-derivative random process. ...

This paper addresses the source-filter separation problem in the context of causal/anticausal linear filter model of voice production. An algorithm based on standard signal processing tools is proposed for the class of quasi-periodic signals (voiced sounds with quasi-stationary pitch). At first, a one-period frame of an equivalent stationary infinitely periodic signal is built. A particular att...

S-transform is a new time-frequency analysis method, which is deduced from short-time Fourier transform and continue Wavelet transform. It has much better performance than traditional time-frequency method. Therefore, in this paper, the basic principle of is briefly introduced and the relationships between is analyzed by theoretical derivation. According to the simulation experiments, the time-...

Degraded signal has an adverse effect on the performance of the system and therefore it must be removed before further processing of signals. Spectral analysis through the Fourier transform is the simplest way to remove the noise, but it is limited to only stationary signals. Non stationary signals require continuous analysis of the signal in time and frequency domain. Wavelets provide better t...

In this paper, we consider sparsity-driven joint localization of stationary and moving targets inside enclosed structures using a reduced set of spatial-fast time-slow time observations in ultrawideband (UWB) pulsed radar platforms. We exploit the compact temporal support of the UWB signal to suppress the front wall clutter through time gating. The resulting enhancement in the signal-to-clutter...

Cyclostationary models for communications signals have been shown in recent years to ooer many advantages over stationary models. Stationary models are adequate in many situations, but they cause important features of the signal to be overlooked. One such important feature is the correlation between spectral components that many signals exhibit. Cyclostationary models allow this spectral correl...

A cloned esterase and several cytoplasmic proteins which lack a classical cleavable signal-peptide were secreted during late stationary phase in B. subtilis. Several lines of evidence indicate that secretion of enolase, SodA, and Est55 is not due to cell lysis. The extent of possible release of these proteins mediated by membrane vesicles into the medium was also found to be minimal. We have id...