The Experts below are selected from a list of 17676 Experts worldwide ranked by ideXlab platform

H.l. Van Trees - One of the best experts on this subject based on the ideXlab platform.

  • ICASSP - A spectrally-based signal subspace approach for speech enhancement
    1995 International Conference on Acoustics Speech and Signal Processing, 1
    Co-Authors: Yariv Ephraim, H.l. Van Trees
    Abstract:

    The signal subspace approach for enhancing speech signals degraded by uncorrelated additive noise is studied. The underlying principle is to decompose the vector space of the noisy signal into a signal plus noise subspace and a noise subspace. Enhancement is performed by removing the noise subspace and estimating the clean signal from the remaining signal subspace. The decomposition can theoretically be performed by applying the Karhunen-Loeve transform to the noisy signal. Linear estimation of the clean signal is performed using a perceptually meaningful estimation criterion. The estimator is designed by minimizing signal distortion for a fixed Desired Spectrum of the residual noise. This criterion enables masking of the residual noise by the speech signal. The filter is implemented as a gain function which modifies the KLT components corresponding to the signal subspace. The gain function is solely dependent on the Desired Spectrum of the residual noise. Listening tests indicate that 14 out of 16 listeners strongly preferred the proposed approach over the spectral subtraction approach.

Yariv Ephraim - One of the best experts on this subject based on the ideXlab platform.

  • ICASSP - A spectrally-based signal subspace approach for speech enhancement
    1995 International Conference on Acoustics Speech and Signal Processing, 1
    Co-Authors: Yariv Ephraim, H.l. Van Trees
    Abstract:

    The signal subspace approach for enhancing speech signals degraded by uncorrelated additive noise is studied. The underlying principle is to decompose the vector space of the noisy signal into a signal plus noise subspace and a noise subspace. Enhancement is performed by removing the noise subspace and estimating the clean signal from the remaining signal subspace. The decomposition can theoretically be performed by applying the Karhunen-Loeve transform to the noisy signal. Linear estimation of the clean signal is performed using a perceptually meaningful estimation criterion. The estimator is designed by minimizing signal distortion for a fixed Desired Spectrum of the residual noise. This criterion enables masking of the residual noise by the speech signal. The filter is implemented as a gain function which modifies the KLT components corresponding to the signal subspace. The gain function is solely dependent on the Desired Spectrum of the residual noise. Listening tests indicate that 14 out of 16 listeners strongly preferred the proposed approach over the spectral subtraction approach.

B. Srinivasan - One of the best experts on this subject based on the ideXlab platform.

  • Spectral tailoring of fiber Bragg gratings through scanning beam fabrication technique
    Journal of Optics, 2015
    Co-Authors: Sonali Prava Dash, A. Pandiarajan, Alessandra Rinaldo, B. Srinivasan
    Abstract:

    In this paper, we discuss the importance of spectral tailoring of fiber Bragg gratings (FBG) for applications ranging from optical sensors to optical communication, and demonstrate a simple technique for fabricating FBGs with a Desired Spectrum. The underlying principle that is exploited in our work is the dependence of the FBG Spectrum on the average index profile along the grating. In the first part of the paper, we simulate the grating Spectrum for various refractive index profiles including the raised cosine profile and the Gaussian profile. In the later part, we discuss the fabrication of such gratings using a scanning beam technique. We demonstrate the control of the grating Spectrum and the reflectivity through the scanning beam fabrication.

Zhang Ying-hui - One of the best experts on this subject based on the ideXlab platform.

  • New peak searching method in subspace Spectrum
    Journal of Computer Applications, 2009
    Co-Authors: Zhang Ying-hui
    Abstract:

    The orthogonality between the signal subspace and the noise subspace is widely employed to implement the Direction of Arrival(DOA)estimation.However,it is hard to seek the Spectrum peak in real time due to the large computing load.The new algorithm configured a threshold firstly so that all intervals including the peaks could be located by large searching step.Furthermore,small searching step was proposed to find out the Desired Spectrum peaks in these intervals.The principle,implementation and performance of the new approach were addressed.Last but not least,the computer simulations illustrate that the new strategy reduces the computing load by 40% to 50%.

Periklis Petropoulos - One of the best experts on this subject based on the ideXlab platform.

  • Nonlinear Generation of Ultra-Flat Broadened Spectrum Based on Adaptive Pulse Shaping
    Journal of Lightwave Technology, 2012
    Co-Authors: Xin Yang, David J. Richardson, Periklis Petropoulos
    Abstract:

    This paper investigates the nonlinear generation of ultra-flat broadband spectra suited to communication applications. The technique is based on the inclusion of a linear pulse shaping element prior to a nonlinear fiber, where highly controllable spectral broadening is enabled. An adaptive pulse shaping system exploiting optimization algorithms (evolution strategy-ES) allows for automatic convergence to the Desired Spectrum. Using this technique, a Spectrum exhibiting a 3-dB bandwidth of ~12 nm and a 0.5-dB bandwidth of ~8 nm is reported. The technique can be used to generate even broader spectra or indeed spectra exhibiting more exotic shapes.