Microwave Dielectric Property Retrieval From Open-Ended Coaxial Probe Response With Deep Learning

This work presents a technique for dielectric property retrieval through Debye parameter reconstruction from open-ended coaxial probe (OECP) response. parameters were obtained with the application of deep learning (DL) model to reflection coefficient response OECP when terminated material under test. The was modelled well-known admittance 0.5 6 GHz 20 MHz resolution. A dataset generated using and data utilized design DL model. As part standard procedure, separated train, validate, test parts by allocating 80%, 10%, 10% each section, respectively. Obtained percent relative error 1.86±3.01%, 3.33±9.52%, 2.07±7.42% $\epsilon _{s}$ , _\infty $ notation="LaTeX">$\tau To further constructed model, responses measured at same frequency band it five different liquids, four mixtures, gel-like material. Reconstructed used retrieve complex properties results compared literature data. mean ranging 1.21±0.06 10.89±0.08 within interest.