A DSP-Based EIT System With Adaptive Boundary Voltage Acquisition

Abstract

Electrical Impedance Tomography (EIT), a valuable technique modality for estimating electrical properties of an object by electrode measurements. During the measurement process, high-frequency excitation signal and the large dynamic range of boundary voltage both cause negative effects on system detection accuracy. Focusing on the induced signal detection, we proposed an adaptive boundary voltage acquisition method, which thoroughly modeled the EIT system by LTSPICE and MATLAB, dynamically adjusted the boundary voltage amplitude gain according to the induced signal voltage range, and changed sampling period for high frequency component of excitation signal. A DSP-based EIT hardware experimental platform under development was constructed, including a current source integrated direct digital synthesis and data acquisition circuit. The data acquisition circuit composed of programmable differential amplifier, low-pass filter and offset adjustment circuit is used for signal acquisition. Comparative experiments are systematically implemented for evaluating the proposed method. Quantitatively, the SNR difference of each measurement channel is reduced from 26.42dB to 6.33dB. The maximum SNR of the system can reach 60.67dB. The maximum image error (IE) is reduced from 28.17% to 15.60%. Accordingly, the proposed adaptive boundary voltage acquisition method is proved to be great beneficial to improve EIT system accuracy.