作者: J. Monzo-Cabrera , J.L. Pedreno-Molina , A. Lozano-Guerrero , A. Toledo-Moreo
DOI: 10.1109/TMTT.2008.2007318
关键词:
摘要: In this study, a novel ten-port waveguide microwave sensor is designed, implemented, calibrated and tested in order to obtain the reflection coefficient magnitude phase. This reflectometer based on well known six-port structure but number of detectors has been increased eight improve sampling procedure standing wave present within waveguide. addition, learning method neural networks' usage implemented for autonomous calibration from data collected by vector network analyzer. An automated consisting moving sample multimode cavity enabled different coefficients be obtained. Neural networks have employed learn relationship between actual parameter acquired signals power detectors. device with architecture radial basis functions error measurements analyzed. new design incorporated allow one avoid problems caused fault or nonlinearity detectors, get robustness, flexibility adaptability characteristics presented device.