Sensitivity analysis of S-parameters Including port variations using the Transfinite Element Method

An Impedance Transition Method to Verify the Reference Impedance of Multiline TRL Calibration

“…The expected response and the validation bounds were determined by 2d EM simulation using the full-wave solver ANSYS HFSS. The values listed in Table I were used for the simulation and all necessary derivatives were calculated directly from the EM simulator [30]. From Fig.…”

“…The derivatives in the Jacobian matrix can be determined from analytical models of the transmission line or, more accurately, directly from the EM solver [30]. The standard uncertainty of the absolute value of the reflection coefficient is determined similarly by propagating the covariance matrices of the characteristic impedance of both calibrations from (32) through the absolute value of ( 16), which is determined by…”

Near-field detection at microwave frequencies based on self-adjoint response sensitivity analysis

“…In self-adjoint formulations, the adjoint system analysis is eliminated since the adjoint-problem solution Ê is derived from the solution Ē of the original problem by simple mathematical transformations without solving (13). Self-adjoint approaches have been developed for both time-and frequency-domain problems; see, for example, [38,41,42,44,50,51]. The self-adjoint formulation becomes possible in problems where (i) the constitutive tensors are symmetric (reciprocity holds), i.e.…”

“…In imaging, the number of parameters may be on the order of thousands. Another prohibitive constraint is due to the limited accuracy of the simulated responses, which causes catastrophic cancellations in the finite-difference estimates of the response derivatives [36][37][38].…”

“…Self-adjoint sensitivity formulas have been proposed where the second (adjoint) system analysis is eliminated [22,33,34,38,[41][42][43][44]. This is achieved by formulating an adjointproblem solution which can be obtained from the solution of the original problem by simple mathematical transformations.…”