Research On Irregular Matching Network Optimization Method For High Efficiency Power Amplifier

With the development of computer and intelligent computing technology,multi-objective evolutionary algorithm has been fully applied in the field of practical engineering design due to its strong global search capability and high degree of optimization freedom,and related cross-application research has become the current research hotspot and development trend.The power amplifier is the core functional component of the wireless communication network system,and its performance is a key factor affecting the data transmission rate and energy efficiency of the entire system.Future wireless communication systems will place higher demands on the bandwidth and efficiency of the power amplifiers.Therefore,the optimizations of the power amplifier and the output matching networks are the important researches in the field of optimization.However,due to the complexity of the power amplifiers,there are certain difficulties in directly applying multi-objective evolutionary algorithms to optimization of the power amplifier and its matching network topology.Related algorithms and strategies need to be further studied and improved.Therefore,to improve bandwidth and back-off efficiency,this dissertation studies the optimization method of irregular matching network structure for high-efficiency power amplifier design based on the multi-objective evolutionary algorithm.The main works are as follows:1.For the multi-objective optimization of irregular matching networks,the algorithm strategies such as grid coding,grid generation and grid division for fragment-type discrete structures are designed,and multi-objective evolutionary algorithms based on decomposition and modified genetic operators is used to optimize output matching network examples and achieve complete impedance matching over broad frequency band.Compared with the traditional multi-objective evolutionary algorithm,the method used has better convergence and population diversity.Based on it,a multi-objective optimization design framework for irregular matching networks is proposed,which can be used for the optimization design of high-efficiency continuous class power amplifiers and Doherty power amplifiers in this dissertation.2.To extend the bandwidth of high-efficiency continuous class power amplifiers,a multi-objective optimization algorithm based on edge filtering for irregular filtering matching network is proposed.The weighted edge filter operator is used to improve the connectivity of the grid-type discrete structure to improve optimization efficiency and success rate.Using this method,an irregular output matching network for continuous class power amplifiers is optimally designed,which has a steeper frequency characteristic than the traditional Chebyshev method.This method solves the problems of poor out-of-band rejection and insufficient filtering characteristics.With it,a power amplifier is optimized to achieve a relative operating bandwidth of about 67% and a saturation efficiency exceeding 65%.Compared with the traditional design method,the relative working bandwidth is expanded by 15%,which meets the network convergence requirements of various wireless communication systems in the 1.5-3GHz frequency band.3.For the improvement of efficiency and bandwidth in Doherty power amplifiers,an impedance optimization objective function algorithm based on the reflection coefficient circle is proposed,which fully utilizes the target impedance region obtained by load-pull and improves the optimal impedance space of the matching network.Then,the problem of low optimization efficiency in the traditional optimization objective function using the impedance error value can be solved.Considering the two states of low power and saturation for Doherty power amplifiers,an output matching network optimization method based on two-state reflection coefficient circle is proposed,which can simultaneously meet the impedance matching requirements in two states and is used for irregular matching in broadband Doherty power amplifier optimization design.Compared with the traditional method,the relative bandwidth is expanded by 33% with high efficiency at 6dB back-off power,which verifies the effectiveness of the optimized design method in expanding the bandwidth of the Doherty power amplifier.4.For high-power back-off applications,based on a symmetric Doherty power amplifier with integrated enhanced reactance,an optimization method for an output matching network with irregular structure with phase constraints is proposed using an impedance and phase mixed objective function algorithm.By reducing the phase dispersion of the output matching network,the efficiency difference of the existing high-power back-off Doherty power amplifier can be reduced.For verification,an irregular structure high-power back-off Doherty power amplifier is designed and optimized.Compared with the existing regular structure design,the efficiency difference of the optimized power amplifier over wide frequency band is reduced from 10% to 5%.With the 9dB power back-off efficiency of higher than 45%,the relative operating bandwidth is widened by 26%,which can be used to amplify the modulated signals with a peak-to-average power ratio exceeding 9dB in the 1.7-2.5GHz band.