基于统计信号处理的电磁层析成像解调方法

Demodulation Method for Electromagnetic Tomography Based on Statistical Signal Processing

中文关键词:  电磁层析成像  统计信号处理  幅值特征提取  数字解调  有色噪声减弱

英文关键词:EMT  statistical signal processing  amplitude feature extraction  digital demodulation  colored noise reduction

中文摘要:

      在电磁层析成像（Electromagnetic Tomography,EMT）领域，幅值信号解调的准确性直接决定了图像重建的精度。针对传统EMT解调方法受环境中有色噪声影响大的问题，基于统计信号处理的理论分析，探索使用了更适应有色噪声环境的广义最小二乘法（Generalized Least Squares,GLS）信号幅值解调方法，并在EMT硬件系统中实现，将其与快速傅里叶变换（Fast Fourier Transform,FFT）和其他统计方法进行幅值对比，验证了该方法在白噪声和有色噪声中的抗噪性。通过对比成像指标，得出该方法可解决因噪声引起的频谱混叠问题，平均相关系数提升了15.32%，相对误差减少了27.93%。与传统EMT成像算法相比，该算法提高了EMT的成像精度。

英文摘要:

      In the field of electromagnetic tomography(EMT),the accuracy of amplitude signal demodulation directly determines the accuracy of image reconstruction.Aiming at the problem that the traditional EMT demodulation method is greatly affected by coloured noise of the environment,based on the theoretical analysis of statistical signal processing,the generalized least squares(GLS) signal amplitude demodulation method,which is more adapted to the environment of coloured noise,is adopted,and implemented in EMT hardware system.The method is compared with the fast Fourier transform(FFT) and other statistical methods to validate the methods noise immunity in white noise and coloured noise.By comparing the imaging metrics,it is concluded that the method can solve the problem of spectral aliasing due to noise,the average correlation coefficient is improved by 15.32%,and the relative error is reduced by 27.93%.Compared with the traditional EMT imaging algorithm,the algorithm improves the imaging accuracy of EMT.