论文部分内容阅读
神经网络具有模拟任意非线性系统的优势。考虑到射频功放的非线性和记忆效应,在BP神经网络模型的基础上,提出一种基于PSO的BP神经网络射频功放行为模型。利用飞思卡尔(Freescale)半导体晶体管MRF6S21140器件模型及设计的电路,从ADS中导出输入输出数据,对模型进行了仿真实现,得出输出电压幅度的拟合曲线以及均方根误差,并与BP神经网络模型进行比较。仿真结果表明,所提模型具有较高的精度和较好的逼近能力,可以精确模拟功率放大器的特性,对系统仿真的构建具有重要的应用价值。
Neural networks have the advantage of simulating any nonlinear system. Considering the nonlinearity and memory effect of RF power amplifiers, a BP neural network model of RF power amplifier based on PSO is proposed based on BP neural network model. Using Freescale semiconductor transistor MRF6S21140 device model and circuit design, the input and output data are derived from ADS, the model is simulated, the fitting curve of output voltage amplitude and root mean square error are obtained, Neural network model for comparison. The simulation results show that the proposed model has high accuracy and good approximation ability, and can accurately simulate the characteristics of the power amplifier. It has an important application value for the system simulation.