计算机应用与软件Issue(12)：60-63,4.DOI:10.3969/j.issn.1000-386x.2013.12.016
基于 GAPSO＿BP 神经网络的 Doherty功放行为模型
DOHERTY POWER AMPLIFIER BEHAVIOURAL MODEL BASED ON GAPSO_BP NEURAL NETWORK
摘要
Abstract
In the communication system-level simulation, it is extremely important for the design and optimisation of RF power amplifiers to build accurate behavioural models .Based on the BP neural network model , we use a hybrid algorithm which combines the genetic algorithm with particle swarm algorithm to optimise the network , build GAPSO_BP amplifier behavioural model , and simulate the model by using Doher-ty structure amplifier input and output voltage data .Through the comparison between the root mean square error of voltage and the conver-gence rate, it eventually comes to a conclusion that the model based on GAPSO has a better fit than the model based on original two algo -rithms, the mean square error between the PA actual output and the modal output reaches 0.0011, and thus it is more accurate to describe the nonlinear characteristics of RF PA .关键词
功率放大器/行为模型/GAPSO算法/BP神经网络Key words
Power amplifier/Behavioural model/GAPSO/BP neural network分类
信息技术与安全科学引用本文复制引用
许璟,南敬昌..基于 GAPSO＿BP 神经网络的 Doherty功放行为模型[J].计算机应用与软件,2013,(12):60-63,4.基金项目
国家自然科学基金项目（60971048）；辽宁省博士科研启动基金项目（20091033）。 （）
