Behavioral Modeling of Quadrature Modulators for Characterization of Nonlinear Distortion

Nonlinear behavioral model for quadrature modulators and compensators

“…Second, the identified different behavioral models are validated using the original QMOD input signals, and then the NMSE with different models can be calculated using (8), the results are shown in Table I. Comparison results show that the proposed nonlinear behavioral model has much better modeling accuracy than the models proposed in [8] and [9], and it also validates the correctness of the proposed nonlinear behavioral model.…”

“…where c n =αa n +βb n (n=1,…, 9), and they are all complex numbers. From (4) we can see that the QMOD nonlinear behavioral model can be simplified as a complex polynomial about I and Q parts of the input signal.…”

“…Compared with the complex power series model proposed in [9], this paper proposed model adds cross terms about I and Q channels to improve the modeling accuracy, and it also can be considered as the simplified version of the Volterra series model proposed in [10].…”

“…where y(n) is the experimental measured feedback complex signal from the QMOD, y'(n) is the output signal of the nonlinear behavioral model, and N is the number of samples. For comparison, the QMOD nonlinear behavioral models proposed in [8] and [9] are also presented. It takes the following two steps to evaluate the modeling accuracy of different behavioral models.…”

“…A nonlinearity compensation method is proposed in [8], but only third order nonlinearity is considered. In [9], complex power series are used to character the nonlinearity of QMOD. A dual-input nonlinear model based on a real-valued Volterra series is proposed in [10], which can compensate the frequency-dependent nonlinear distortions.…”