Customization Theory of a Set of Optimal Calibration Standards for the Six-Port Reflectometer Based on Power Measurement Errors

A design theory of optimal calibration standards (C.S.) for a six-port reflectometer based on the power measurement error is proposed here. Rather than general, the C.S. are specially customized corresponding to the practically used six-port reflectometer system itself. The basic logic is the objective stability of the six-port reflectometer in representing a reflection coefficient (Γ ). Treated as devices under test, the C.S. are customized as optimal corresponding to a power measurement condition during long-term and repeated usage. The availability of the customization is first verified in simulation. The coefficient Γ around the entire Smith chart is calculated after different sets of C.S., respectively. The customized calibration is more accurate and distributes more concentratedly. The effect is especially obvious for Γ with 0 to -15 dB in magnitude. A practical six-port reflectometer system is built and performed as the proposed method. By the replacement of the optimal C.S., the average of the worst-case range under long-term power measurement error is reduced from ±5.22 to ±3.50 dB in magnitude and from ±31.17° to ±19.18° in phase for Γ at -13 dB. More importantly, the probability with error larger than 15% in magnitude ( Γ < -11.8 dB or Γ > -14.4 dB) is decreased from 27% to 16%. The proposed method is theoretically verified for a general working bandwidth of the built six-port (11.7% at 2.55GHz). The theory can be a plug-and-play and supplementary rule for most instruments based on multiport measurement to enhance the reliability and extend the lifespan.