Computation of higher-order cross moments based on matrix multiplication

Abstract The demand for efficient architectural designs for fast computation of higher-order cross moments is becoming increasingly significant as these statistics grow in importance as a signal processing tool. In this paper, we present a new approach to the computation of third- and fourth-order cross moments based on matrix multiplication. The results presented in this paper has the advantage of putting cross moments computation in a form that can benefit from previous (and possibly future) work on architectural designs for matrix multiplication.

[1]  Saleh A. Alshebeili,et al.  A new approach for the design of linear systolic arrays for computing third-order cumulants , 1997, Integr..

[2]  I. V. Ramakrishnan,et al.  Synthesis of an Optimal Family of Matrix Multiplication Algorithms on Linear Arrays , 1986, IEEE Transactions on Computers.

[3]  Rana Ejaz Ahmed,et al.  A concurrent system for the computation of higher-order moments , 1999 .

[4]  Tanguy Risset.  Linear systolic arrays for matrix multiplication: comparisons of existing synthesis methods and new results , 1991, Algorithms and Parallel VLSI Architectures.

[5]  Elias S. Manolakos,et al.  An array of processors for the real-time estimation of fourth- and lower-order moments , 1994, Signal Process..

[6]  Chrysostomos L. Nikias,et al.  Bearing estimation in the bispectrum domain , 1990, Fifth ASSP Workshop on Spectrum Estimation and Modeling.

[7]  Chrysostomos L. Nikias,et al.  Multichannel adaptive blind deconvolution using the complex cepstrum of higher order cross-spectra , 1993, IEEE Trans. Signal Process..

[8]  Steve A. Billings,et al.  Performance monitoring in non-linear adaptive noise cancellation , 1992 .

[9]  Elias S. Manolakos,et al.  Adaptive computation of higher-order moments and its systolic realization , 1996 .

[10]  Raghuveer M. Rao,et al.  Nonparametric bispectrum-based time-delay estimators for multiple sensor data , 1991, IEEE Trans. Signal Process..

[11]  Chrysostomos L. Nikias,et al.  The cross-bicepstrum: definition, properties, and application for simultaneous reconstruction of three nonminimum phase signals , 1993, IEEE Trans. Signal Process..

[12]  Patrice Quinton,et al.  Algorithms and Parallel VLSI Architectures , 1992, Algorithms and Parallel VLSI Architectures.

[13]  Rana Ejaz Ahmed,et al.  VLSI architecture for computing third-order cumulants , 1994 .

[14]  Jitendra K. Tugnait,et al.  On time delay estimation with unknown spatially correlated Gaussian noise using fourth-order cumulants and cross cumulants , 1991, IEEE Trans. Signal Process..

[15]  A. Lohmann,et al.  Triple correlations , 1984, Proceedings of the IEEE.

[16]  Alessandro Neri,et al.  Bandpass nonlinear systems identification by higher order cross correlation , 1991, IEEE Trans. Signal Process..

[17]  Viktor K. Prasanna,et al.  On Synthesizing Optimal Family of Linear Systolic Arrays for Matrix Multiplication , 1991, IEEE Trans. Computers.

[18]  Anastasios N. Venetsanopoulos,et al.  Cumulant-based parametric multichannel FIR system identification methods , 1994 .