Research On Microwave Power Amplifier Behavioral Model

With the advancement of simulation tools and simulation technology,China has made great progress in the field of digital/virtual prototypes of radar electronics system in recent years.However,because accurate modeling of active microwave devices is difficult and time-consuming,ideal models are generally used in radar systems to characterize the behaviors of microwave devices,making the characteristics of noise,nonlinearity,spurs,and delay in actual systems not accurately described.This results in a large difference between the simulation results and the measurement results.As most of the current radar system simulations focus on algorithm level verification,it is difficult to accurately describe the performance of the entire system.Therefore,constructing active microwave device models quickly and accurately has become an urgent problem to be solved in system simulation.Due to the strong nonlinearity of the power amplifier,it has the biggest impact on the system performance,which is the focus and difficulty of active microwave device modeling.From the perspective of system simulation,the model is only required to feedback the corresponding output according to the input signal,so the behavioral model can be used to represent the power amplifier.At present,the researchers at home and abroad focus on microwave power amplifier behavioral model research using high-order mathematical expressions to describe the device characteristics,so that the model has a more comprehensive characterization capability,thereby obtaining higher accuracy.However,this requires a complex parameter extraction platform and extraction process,making behavioral models extraction difficult to popularize.In this regard,the K-parameter model and its extraction method are proposed.This method can not only describe the nonlinear response of the microwave power amplifier when there is a certain mismatch at the input / output,but also can measure and model the device at low cost and quickly,which can speed up the usage of the behavioral model.The main research contents of this article are as follows:1.K-parameter behavioral model and its parameter extraction technology.When the behavioral model with load-pull capability is extracted,in addition to the signal acquisition instrument,the additional load-pull equipment is required,which increases the cost and complexity of model extraction.Considering that the main influencing factor of the power amplifier’s large signal operating point is the fundamental impedance of load,the fundamental incident wave at the output port is taken as one of the input variables of the model to derive the K-parameter.In the process of K-parameter extraction,the amplitude and the phase of the fundamental incident wave on the output will change,which can be equivalent to the change of the fundamental load impedance,so that the extracted behavioral model have a certain load-pull capability.Further,in view of the complexity of the existing behavioral model extraction platform and the high cost of modeling and testing,a K-parameter modeling method combining simulation and measurement is proposed.This method can build a K-parameter extraction platform based on commercial microwave simulation software.Moreover,using a commercial nonlinear vector network analyzer and a few equipments,nonlinear testing and parameter extraction can be achieved.At the same time,because the test platform does not contain narrow-band devices such as isolators,and the driver amplifier only needs to cover the fundamental frequency band of the device under test,the platform is suitable for the extraction of the broadband power amplifier.2.Vector signal superposition effect and incident wave recovery algorithm.The scene of vector calibration is that one port of the test equipment outputs the excitation signal,and the other ports are connected to the matched loads.In the K-parameter extraction,the input and output ports will simultaneously stimulate large signals.Due to factors such as impedance mismatch and limited isolation of the coupler,the incident wave recorded by the coupler is a vector superposition of the excitation signal and the interference signal,and it cannot be recovered using vector calibration.After the principle analysis of the generation of products when working with large signals,the incident wave signal recovery algorithm is proposed according to the characteristics of K-parameter extraction process,and the incident wave signals at the input and output ends can be recovered without increasing the test steps.3.Cellular point-taking method for load-pull.Aiming at the problem that the existing load-pull sector point-taking method has poor uniformity of points in the Smith chart,a cellular structure is used to propose point-taking method.The new method can uniformly cover the whole Smith chart area without increasing the number of points to be taken,thereby improving the accuracy of the behavioral model.The new method is applied to load-dependent X-parameter and K-parameter model extraction.Compared with the sector point-taking method,the normalized mean square error of the model obtained by new method is reduced to a certain extent.4.K-parameter model cascade characteristics analysis.In system simulation,the power amplifier will be connected to other devices,which requires the model to have cascade simulation capabilities.In view of the current situation that most model studies only analyze the characteristics of the device itself,K-parameter models are theoretically analyzed from two aspects of source traction and load traction.And it is deduced that K-parameter can make correct response according to the impedance of front and rear cascade circuit,that is,it has the capability of cascade simulation.Then simulation and testing methods are used to verify the cascade simulation capability of K-parameters under the excitation of continuous wave and modulated signal.The accuracy of the cascaded K-parameter models is comparable to the cascaded load-dependent X-parameter models,while the K-parameter extraction platform is simpler and the model extraction cost is lower.5.K-parameter characterization method for two-port devices.Existing limiters,frequency multipliers and other two-port devices based on the nonlinear characteristics of amplifiers are characterized mostly by compact models,which are difficult and time consuming to extract.By analyzing the system simulation requirements for these two-port device models,the key indicators are selected for different devices,and the expansion order of K-parameter output reflected wave is adjusted to accurately characterize the device model.The fundamental K-parameter model for the limiter chip is extracted and verified.For the terahertz frequency tripler circuit,the K-parameter is extracted and compared with the simulation results of the circuit model,and the average relative error of the reflected wave is 1.5%.Using the K-parameter to model a two-port device,the key characteristics of the device can be extracted quickly and accurately.