Study On Rf Behavioral Models Of Power Amplifier And Circuit Simulation Technolgy

With the rapid development of wireless communication systems, for the radio-frequency microwave field has a much shorter product design cycles and higher precision. Propose higher demand on RF microwave circuit CAD key techniques. But there are a number of RF circuit simulation software existence problems of models and simulation engine algorithm's accuracy is not high, especially in the communication system of the most critical components - power amplifier system-level simulation, its behavior model accuracy is an important factor affect the simulation results. Therefore, in order to enable RF circuit simulation system can more accurately simulate the real system, to study the amplifier behavior model and circuit simulation technology is of great significance.Under the above context, with the support of the National Natural Science Foundation, "Study on nonlinearity and behavioral models for broadband RF power amplifier " and Doctoral Research in Liaoning Province to start fund "RF microwave circuit CAD simulation technology and radio frequency reader research", this article is intended to develop some techniques for RF power amplifier behavioral models and RF circuit simulation technology.As the neural network has strong learn function, can simulate the complex relationship of the nonlinear function. On basis of analyses and comparison of some PA behavioral models, proposes a kind of method that apply neural network modeling behavioral model, using MATLAB to validate and compare the performance of models, results show that the PA model based on RBF neural network compared with other models has higher accuracy and better performance. At the same time, based on the study on the past RF linear and nonlinear circuit analysis method, improve S-matrix method and harmonic balance method, and validate the effectiveness of the improved method. This has important reference value of RF microwave circuit CAD simulation technology.