论文部分内容阅读
针对神经网络和支持向量机在射频功率放大器建模领域存在的优缺点,提出一种利用PSO_SVM算法对射频功率放大器进行建模的方法。从理论上分析了支持向量机(SVM)及粒子群优化(PSO)算法的相关原理,并将PSO_SVM算法应用到功放器件建模中。仿真结果表明,基于PSO_SVM的射频功放模型在模型精度、小样本学习和逼近能力方面均优于传统SVM模型和BP神经网络(BPNN)模型。
In view of the advantages and disadvantages of neural networks and support vector machines in the field of RF power amplifier modeling, a method of modeling the RF power amplifier by using the PSO_SVM algorithm is proposed. The theory of Support Vector Machine (SVM) and Particle Swarm Optimization (PSO) algorithm is theoretically analyzed. The PSO_SVM algorithm is applied to power amplifier device modeling. The simulation results show that the RF power amplifier model based on PSO_SVM is superior to the traditional SVM model and BP neural network (BPNN) model in model accuracy, small sample learning and approximation ability.