The Experts below are selected from a list of 270 Experts worldwide ranked by ideXlab platform

Alexander Jung - One of the best experts on this subject based on the ideXlab platform.

  • On the Sample Complexity of Graphical Model Selection From Non-Stationary Samples
    IEEE Transactions on Signal Processing, 2020
    Co-Authors: Nguyen Tran, Oleksii Abramenko, Alexander Jung
    Abstract:

    We study conditions that allow accurate graphical model selection from non-stationary data. The observed data is modelled as a vector-valued zero-mean Gaussian Random Process whose samples are uncorrelated but have different covariance matrices. This model contains as special cases the standard setting of i.i.d. samples as well as the case of samples forming a stationary time series. More generally, our approach applies to any data for which efficient decorrelation transforms, such as the Fourier transform for stationary time series, are available. By analyzing a conceptually simple model selection method, we derive a sufficient condition on the required sample size for accurate graphical model selection based on non-stationary data.

  • Learning conditional independence structure for high-dimensional uncorrelated vector Processes
    2017 IEEE International Conference on Acoustics Speech and Signal Processing (ICASSP), 2017
    Co-Authors: Nguyen Tran Quang, Alexander Jung
    Abstract:

    We formulate and analyze a graphical model selection method for inferring the conditional independence graph of a high-dimensional nonstationary Gaussian Random Process (time series) from a finite-length observation. The observed Process samples are assumed uncorrelated over time but having a time-varying marginal distribution. The selection method is based on testing conditional variances obtained for small subsets of Process components. This allows to cope with the high-dimensional regime, where the sample size can be (much) smaller than the Process dimension. We characterize the required sample size such that the proposed selection method is successful with high probability.

  • ICASSP - Learning conditional independence structure for high-dimensional uncorrelated vector Processes
    2017 IEEE International Conference on Acoustics Speech and Signal Processing (ICASSP), 2017
    Co-Authors: Nguyen Tran Quang, Alexander Jung
    Abstract:

    We formulate and analyze a graphical model selection method for inferring the conditional independence graph of a high-dimensional nonstationary Gaussian Random Process (time series) from a finite-length observation. The observed Process samples are assumed uncorrelated over time but having a time-varying marginal distribution. The selection method is based on testing conditional variances obtained for small subsets of Process components. This allows to cope with the high-dimensional regime, where the sample size can be (much) smaller than the Process dimension. We characterize the required sample size such that the proposed selection method is successful with high probability.

W. Greblicki - One of the best experts on this subject based on the ideXlab platform.

  • Nonlinearity recovering in Wiener system driven with correlated signal
    IEEE Transactions on Automatic Control, 2004
    Co-Authors: W. Greblicki
    Abstract:

    The characteristic of the nonlinear part of the Wiener system is estimated. The system is driven by a Gaussian Random Process which may not be white. Three algorithms are presented, two semirecursive and one of the offline type. Their pointwise convergence in probability is shown and results of numerical simulation are given.

  • Continuous-time Wiener system identification
    IEEE Transactions on Automatic Control, 1998
    Co-Authors: W. Greblicki
    Abstract:

    A continuous-time Wiener system is identified. The system consists of a linear dynamic subsystem and a memoryless nonlinear one connected in a cascade. The input signal is a stationary white Gaussian Random Process. The system is disturbed by stationary white Random Gaussian noise. Both subsystems are identified from input-output observations taken at the input and output of the whole system. The a priori information is very small and, therefore, resulting identification problems are nonparametric. The impulse impulse of the linear part is recovered by a correlation method, while the nonlinear characteristic is estimated with the help of the nonparametric kernel regression method. The authors prove convergence of the proposed identification algorithms and examine their convergence rates.

Nguyen Tran Quang - One of the best experts on this subject based on the ideXlab platform.

  • Learning conditional independence structure for high-dimensional uncorrelated vector Processes
    2017 IEEE International Conference on Acoustics Speech and Signal Processing (ICASSP), 2017
    Co-Authors: Nguyen Tran Quang, Alexander Jung
    Abstract:

    We formulate and analyze a graphical model selection method for inferring the conditional independence graph of a high-dimensional nonstationary Gaussian Random Process (time series) from a finite-length observation. The observed Process samples are assumed uncorrelated over time but having a time-varying marginal distribution. The selection method is based on testing conditional variances obtained for small subsets of Process components. This allows to cope with the high-dimensional regime, where the sample size can be (much) smaller than the Process dimension. We characterize the required sample size such that the proposed selection method is successful with high probability.

  • ICASSP - Learning conditional independence structure for high-dimensional uncorrelated vector Processes
    2017 IEEE International Conference on Acoustics Speech and Signal Processing (ICASSP), 2017
    Co-Authors: Nguyen Tran Quang, Alexander Jung
    Abstract:

    We formulate and analyze a graphical model selection method for inferring the conditional independence graph of a high-dimensional nonstationary Gaussian Random Process (time series) from a finite-length observation. The observed Process samples are assumed uncorrelated over time but having a time-varying marginal distribution. The selection method is based on testing conditional variances obtained for small subsets of Process components. This allows to cope with the high-dimensional regime, where the sample size can be (much) smaller than the Process dimension. We characterize the required sample size such that the proposed selection method is successful with high probability.

Pengju Yang - One of the best experts on this subject based on the ideXlab platform.

  • Imaging of object in the presence of rough surface using scattered electromagnetic field data
    2013 Proceedings of the International Symposium on Antennas & Propagation, 2013
    Co-Authors: Pengju Yang
    Abstract:

    Electromagnetic (EM) scattering from a perfect electrically conducting object above lossy dielectric half-space with rough surface is investigated, for both TE and TM polarizations, employing a parallel fast multiple method. Then, based on the scattered electromagnetic field data at multiple-incidence angles and frequencies, a back-projection tomography technique is applied to generate two-dimensional (2-D) synthetic aperture radar images. The Randomly rough surfaces are modeled as realizations of a Gaussian Random Process with the Gaussian spectrum, while the tapered incident wave is chosen to reduce the truncation error.

  • Electromagnetic scattering from 1-D dielectric sea surface based on second-order small slope approximation
    2013 5th IEEE International Symposium on Microwave Antenna Propagation and EMC Technologies for Wireless Communications, 2013
    Co-Authors: Hongmei Miao, Pengju Yang
    Abstract:

    In this paper, electromagnetic scattering from one dimensional dielectric sea surface is investigated, for both TE and TM polarizations, employing the second-order small slope approximation (SSA-II). The second order terms of the small slope approximation method have been numerically implemented in order to obtain accurate results for a large range of slope. The sea surfaces are modeled as realizations of a Gaussian Random Process with the Pierson-Moskowitz spectrum, while the tapered incident wave is chosen to reduce the truncation error. The numerical results obtained by SSA-II are presented by comparison with benchmark numerical method.

  • Electromagnetic scattering from PEC object embedded between two dielectric rough interfaces
    2013 5th IEEE International Symposium on Microwave Antenna Propagation and EMC Technologies for Wireless Communications, 2013
    Co-Authors: Nengxun Yang, Pengju Yang
    Abstract:

    Electromagnetic (EM) scattering from perfect electric conductor (PEC) object embedded between two rough interfaces is investigated utilizing method of moments (MoM) for both TE and TM polarizations. Theoretical formulations of EM scattering model considered in this paper is presented and the total fields and their derivatives are solved numerically by employing MoM. The rough surfaces are modeled as realizations of a Gaussian Random Process with the Gaussian spectrum, while the tapered incident wave is chosen to reduce the truncation error. The influences of characteristic parameters of the rough surfaces, the relative permittivity of the medium, as well as the average height between the two rough surfaces, on the bistatic scattering coefficient (BSC) are discussed.

Dennis Goeckel - One of the best experts on this subject based on the ideXlab platform.

  • Convergence of the Complex Envelope of Bandlimited OFDM Signals
    IEEE Transactions on Information Theory, 2010
    Co-Authors: Dennis Goeckel, Patrick A. Kelly
    Abstract:

    Orthogonal frequency division multiplexing (OFDM) systems have been used extensively in wireless communications in recent years; thus, there is significant interest in analyzing the properties of the transmitted signal in such systems. In particular, a large amount of work has focused on analyzing the variation of the complex envelope of the transmitted signal and on designing methods to minimize this variation. In this paper, it is established that the complex envelope of a bandlimited uncoded OFDM signal converges weakly to a Gaussian Random Process as the number of subcarriers goes to infinity. This shows that the properties of the OFDM signal will asymptotically approach those of a Gaussian Random Process over any finite time interval. The convergence proof is then extended to two important cases, namely, coded OFDM systems and systems with an unequal power allocation across subcarriers.

  • the complex envelope of a bandlimited ofdm signal converges weakly to a Gaussian Random Process
    IEEE Transactions on Information Theory, 2008
    Co-Authors: Dennis Goeckel
    Abstract:

    Orthogonal frequency division multiplexing (OFDM) systems have been used extensively in wireless communications applications in recent years; thus, there is significant interest in analyzing the properties of the transmitted signal in such systems. In particular, a large amount of recent work has focused on analyzing the variation of the complex envelope of the transmitted signal and on designing methods to minimize this variation. In this paper, it is established that the complex envelope of a bandlimited uncoded OFDM signal converges weakly to a Gaussian Random Process as the number of subcarriers goes to infinity. This establishes that the properties of the OFDM signal will asymptotically approach those of a Gaussian Random Process over any finite time interval. The symbol length in a bandlimited OFDM system will eventually exceed any finite time interval as the number of subcarriers approaches infinity; however, practical interest is in how asymptotic approximations apply for a finite number of carriers, and, hence, the convergence proof is reasonable motivation for considering how the extremal value theory of Gaussian Random Processes might provide accurate approximations for the distribution of the peak-to-mean envelope power ratio (PMEPR) of practical OFDM systems. Indeed, numerical results are presented that indicate that the resulting simple expressions are accurate for a wide range of the distribution for moderate numbers of subcarriers. The important extensions of the analytical and numerical results to coded OFDM systems, as well as systems with unequal power allocation across subcarriers, are also presented.