Multi-layer Permittivity Measurement Based on Complementary Split-Ring Resonator and Neural Networks

A neural network-based method for multi-layer permittivity measurement is proposed in this paper. This method uses the multiple-square concentric complementary split-ring resonator (CSRR) to take multiple non-identical resonance frequency measurement, and a scalable, iterative neural network approach is applied to estimate for dielectric property measurement. Instead of the tedious development and establishment of analytic formulas, neural network engine solver can simplify this step and still have acceptable accuracy. The dual-layer MUTs measurement had an average error of 8.78% for ε1 and an average error of 8.9% for ε2. It can be extended to the measurement of more than two layers substrate.