STUDY OF BEAMFORMER TECHNIQUE WITH REDUCED TIME WINDOW FOR MOVING CURRENT SOURCE RECONSTRUCTION
In this paper, a beamformer technique with reduced time window for moving current dipole sources reconstruction is described. The linearly constrained minimum variance (LCMV) beamformer technique uses spatial filters as a weighted sum of the signal on a set of array measurement sites. Based on the assumption that the magnetoencephalography (MEG) is generated by a stationary source distribution in the brain, the covariance matrix of measured signals was calculated with an entire recording. However, we notice that the approach is unsuitable for moving current dipole source reconstruction such as the magnetocardiography (MCG). Therefore, we use a set of temporal highly-correlated measurement vectors to estimate the covariance matrix of measured signals, and to calculate the weight vectors piecewise. The effectiveness of the approach is illustrated with three moving current sources reconstruction in simulation. We applied this reduced time window beamformer technique to two sets of MCG data. Finally, the results and future work are discussed.
beamformer bio-magnetism source reconstruction MCG inverse problem signal processing
W. Y. Wang D. F. Zhou S. Q. Jiang
Department of Control Science and Engineering School of Electronics and Information Engineering, Tongji University,Shanghai, P. R. China
国际会议
2012 International Conference on System Simulation(2012年国际系统仿真学术会议)
上海
英文
294-297
2012-04-06（万方平台首次上网日期，不代表论文的发表时间）