The Research On Digital Predistortion Method Of Concurrent Dual-band Power Amplifier

With the rapid development of communication technology and people’s increasing demand for communication,multi-band and multimode technology has gradually become a development trend.Power Amplifier(PA)is one of the indispensable components of mobile communication system,and its inherent nonlinear characteristics will inevitably cause nonlinear distortion of the signal.Therefore,the linearity of PA directly affects the signal transmission quality of the whole communication system.Digital Predistortion(DPD)technology can effectively compensate the nonlinear distortion caused by power amplifiers in dual-band communication system.It is one of the research hotspots at home and abroad.This paper mainly studies the digital predistortion method of concurrent dual-band power amplifier.The main work is as follows:1.A segmented 2D-U-DVR dual-band predistortion behavior model with low complexity is proposed.The new model is derived from the 2DDVR(Dualband-Decomposed Vector Rotation)model.The product of the envelope amplitude of each frequency band input signal term and the other frequency band input signal is used to represent the correlation of two frequency bands,replacing the cross term in the original 2D-DVR model which has little influence on the overall performance.Then,the magnitudeselective Affine(MSA)function is used as the nonlinear basis function for piecewise processing,which greatly reduces the complexity of the model.The improved segmented 2D-DVR(2D-U-DVR)model takes into account the in-band and inter-band intermodulation components of two frequency bands,and can effectively compensate the nonlinear distortion in concurrent dual-band amplifiers.The experimental results show that the linearization performance of the proposed model is the best and the complexity of the model is the lowest.2.In view of the problem of large number and high complexity of behavior model systems of concurrent dual-band power amplifier,A model clipping method based on Adaptive Regularized Subspace Pursuit based on Variable step-size(ARSP-VS)was proposed.This method can be used to clip the behavior model of power amplifier based on Volterra series type.Experimental results show that the proposed method can obtain a higher model accuracy with a lower number of coefficients,and compared with other Compressed Sensing(CS)greedy algorithms applied to DPD,the linearization effect of the ARSP-VS algorithm is better.This method is applied to the concurrent dual-band digital predistortion technology,which provides a solution to reduce the model complexity.