An Accurate Deep Neural Network Based Inversion Approach for Multichannel Microwave Interferometer Characterization of Plasma Wake Electron Density Distribution

In on-ground characterization of plasma wake of hypersonic target, the multichannel microwave interferometer is commonly used with the Abel inversion to retrieve the distribution of electron density. Although the Abel inversion exhibits excellent calculation efficiency over the numerical method, it suffers from poor accuracy due to the neglect of electromagnetic wave refraction in the characterization of high electron density plasma. Recently, neural network based inversion methods have been extensively explored in other research fields due to their high accuracy and excellent calculation efficiency. Therefore, in this work a inversion approach based on deep neural network is proposed to retrieve the distribution of plasma electron density in multichannel microwave interferometer measurement. The validation of the proposed inversion approach with and without the presence of noise is approved and the accuracy improvement is also verified by comparing the inversion results with those achieved from the Abel inversion. This method can be used for the accurate electron density distribution characterization of plasma in multichannel microwave interferometer, especially when the electron density is high.