“…To tackle this issue, a sixteen-term error model (Fig. 1a) and several calibration methods based on the model have been developed [21], [22]. In the sixteen-term error model, eight errors (e00, e01, e10, e11, e22, e23, e32, e33) are the same as those of the traditional eight-term error model but the remaining eight errors represent the crosstalk between probes (e21 and e12), the crosstalk between receivers in the vector network analyzer (VNA) (e30 and e03), and the crosstalk between the microwave probe on one side of the DUT and the receiver of the VNA at the other side (e02, e20, e31, and e13).…”