Noise reduction

About: Noise reduction is a research topic. Over the lifetime, 25121 publications have been published within this topic receiving 300815 citations. The topic is also known as: denoising & noise removal.

Vibration/noise reduction device for electrical apparatus

Abstract: A vibration/noise reducing device for applying vibration/noise-reducing vibrations/sound waves of substantially opposite phase to vibrations/noises generated from an electrical apparatus is disclosed, in which the vibrations/noises from the electrical apparatus is sensed by a sensor (12, 70), the sensed analog time-domain signal (14) is A/D converted, the resulting digital time-domain signal (18) is then Fourier-transformed, the resulting Fourier-transformed digital frequency-domain signal (22) is modified in its amplitude/phase to produce a second digital frequency-domain signal (26) for generating a vibration/noise reducing control signal, the second digital frequency-domain signal (26) is inverse-Fourier-transformed, the resulting inverse-Fourier transformed second digital time-domain signal (30) is D/A converted, and the resulting analog time-domain signal (34) is used as the control signal to generate the vibration/noise-reducing vibrations/sound waves to be applied to the virbrations/noises.

Analysis of the Decision-Directed SNR Estimator for Speech Enhancement With Respect to Low-SNR and Transient Conditions

Abstract: Because of their many applications and their relative ease of implementation, single-channel speech enhancement algorithms have received much attention. As a consequence, a vast amount of publications on estimation procedures and their implementation in noise reduction systems exists. However, there has been little systematic research on the theoretic performance of such estimators. In this paper, we provide a systematic analysis of the performance of noise reduction algorithms in low signal-to-noise ratio (SNR) and transient conditions, where we consider approaches using the well-known decision-directed SNR estimator. We show that the smoothing properties of the decision-directed SNR estimator in low SNR conditions can be analytically described and that the limits of noise reduction for widely used spectral speech estimators based on the decision-directed approach can be predicted. We also illustrate that achieving both a good preservation of speech onsets in transient conditions on one side and the suppression of musical noise on the other can be especially problematic when the decision-directed SNR estimation is used.

Noise Levels of Superconducting Gravimeters: Updated Comparison and Time Stability

Abstract: Since the beginning of the Global Geodynamics Project (GGP) in 1997, the number of superconducting gravimeters (SGs) has increased to reach 30 operating sites today. Data from this network allow a comparison of the noise levels of the different contributing stations. The knowledge of the noise levels at each site is important in any combination of data to determine global Earth parameters, for example, the stacking of the data in the search for elusive signals, like the gravity variations associated with the translational mode of the inner core. We use a standardized procedure based on the computation of the residual power spectral densities (PSDs) over a quiet time period in order to evaluate the combined instrument plus site noise in the long-period seismic band (0.3 mHz–1 mHz). The experience at Strasbourg (France) has shown some improvements from the TT70-T005 full-size instrument to the C026 compact model in terms of noise reduction, while the most recent Observatory SG types, OSG044 at Bad Homburg (Germany) and OSG052 at Sutherland (South Africa), do not show any further improvement with respect to the compact models, respectively CD30 and CD037, operating at the same stations. At Black Forest Observatory (BFO) in Germany, the experience of the dual-sphere OSG with a lower sphere heavier than usual has shown that the instrumental and site conditions make this station the least noisy one at frequencies larger than 0.1 mHz. The noise analysis using the longest time-series available has shown that the noise level at these sites is mostly stable (within 1 σ ) over the years. The comparison with some seismological noise models shows that the best SG sites are less noisy than long-period seismometers below 1 mHz. However, the noise level of the best SGs is still at the limit of detection of the subseismic translational mode of the inner core.