Detection in multiple disparate systems using multi-channel coherence analysis

This paper presents a coherence-based detection method for multiple disparate sensing systems using the multi-channel coherence analysis (MCA) framework. MCA provides an optimal coordinate system for multi-channel detection problems as it finds sets of one-dimensional mapping vectors that maximize the sum of the cross-correlations among all pair-wise combinations of channels. The standard detector for Gaussian random vectors is then cast into the MCA framework by developing the log-likelihood ratio and J-divergence measure. The proposed detection method is then tested on a data set consisting of sets of four side-scan sonar images coregistered over the same region on the seafloor and the results are compared with those of a multi-channel generalized likelihood ratio (GLR) detector.

[1]  A. Leshem,et al.  Multichannel detection of Gaussian signals with uncalibrated receivers , 2001, IEEE Signal Processing Letters.

[2]  J. Kettenring,et al.  Canonical Analysis of Several Sets of Variables , 2022 .

[3]  J D Tucker,et al.  Coherence-Based Underwater Target Detection From Multiple Disparate Sonar Platforms , 2011, IEEE Journal of Oceanic Engineering.

[4]  M.R. Azimi-Sadjadi,et al.  Buried underwater target classification using frequency subband coherence analysis , 2008, OCEANS 2008.

[5]  Louis L. Scharf,et al.  Detection of Spatially Correlated Gaussian Time Series , 2010, IEEE Transactions on Signal Processing.

[6]  H. Hotelling.  Relations Between Two Sets of Variates , 1936 .

[7]  Robert H. Halstead,et al.  Matrix Computations , 2011, Encyclopedia of Parallel Computing.

[8]  R. M. Johnson.  On a theorem stated by eckart and young , 1963 .

[9]  Allan Aasbjerg Nielsen,et al.  Multiset canonical correlations analysis and multispectral, truly multitemporal remote sensing data , 2002, IEEE Trans. Image Process..

[10]  Ignacio Santamaría,et al.  A learning algorithm for adaptive canonical correlation analysis of several data sets , 2007, Neural Networks.

[11]  Louis L. Scharf,et al.  Canonical coordinates and the geometry of inference, rate, and capacity , 2000, IEEE Trans. Signal Process..

[12]  Louis L. Scharf,et al.  Low rank detectors for Gaussian random vectors , 1987, IEEE Trans. Acoust. Speech Signal Process..

[13]  Mahmood R. Azimi-Sadjadi,et al.  Buried Underwater Object Classification Using a Collaborative Multi-Aspect Classifier , 2007, 2007 International Joint Conference on Neural Networks.

[14]  Herbert Gish,et al.  A geometric approach to multiple-channel signal detection , 1995, IEEE Trans. Signal Process..

[15]  Gene H. Golub,et al.  Matrix computations (3rd ed.) , 1996 .

[16]  John K. Thomas,et al.  Canonical Coordinates are the Right Coordinates for Low-Rank Gauss–Gauss Detection and Estimation , 2006, IEEE Transactions on Signal Processing.

[17]  Mahmood R. Azimi-Sadjadi,et al.  Iterative multi-channel coherence analysis with applications , 2008, Neural Networks.