| New communication systems are emerging with the development of wireless communication demand and technology.However,due to the scarcity of spectrum resources,the spectrum of different communication systems is usually discontinuous.The development of a concurrent multi-band communication transmitter can ensure the simultaneous use of different communication systems and reduce the complexity and cost of the multi-system/dual-band integrated communication system.Various technologies of the concurrent dual-band communication transmitter include a dual-band filtering technology,a dual-band power amplifier matching technology,a dual-band antenna technology and a dual-band power amplifier distortion compensation technology.However,the combined peak-to-average ratio(PAPR)of concurrent dual-band signals is higher than that of single-band signals,which reduces the efficiency of dual-band communication transmitters.When passing through the dual-band communication transmitter,the signal will produce more serious nonlinear distortion,including not only in-band signal distortion,but also intermodulation distortion between different bands.Combined with distortion of a transmitter modulator,it greatly increases the difficulty of distortion compensation and the calculation complexity of distortion compensation.It is difficult to improve system performance and reduce computation complexity at the same time.In order to optimize peak clipping efficiency,distortion compensation performance and computational complexity at the same time,the input signal characteristics and inherent nonlinear characteristics of the transmitter are used in distortion compensation methods.The combination effect of dual-band signals is taken into account in PAPR suppression.In distortion compensation,the inherent distortion characteristics of the transmitter and the probability distribution characteristics of input signals are taken into consideration,which greatly reduces the computational complexity with the same distortion compensation performance.The main researches of this thesis are as follows.1)An optimized two-dimensional PAPR suppression algorithm based on input dual-band signals is proposed.Concurrent dual-band signals are more complicated than traditional single-band signals.The dual-band combination effect leads to over-clipping and under-clipping issue in traditional methods.To address this issue,a two-dimensional PAPR suppression method is proposed based on amplitude match of input dual-band signals.The algorithm avoids the over-clipping and under-clipping and reduces the clipping distortion by amplitudes match of the input signal.The simulation results show that this algorithm improves the PAPR suppression of dual-band signals and the clipping distortion compared with the traditional algorithm.2)A low-complexity two-dimensional peak cancelling method is proposed.In the traditional clipping and filtering method,the calculation complexity of filtering is very high,and at the same time,peak regeneration reduces the effect PAPR suppression.To address this problem,this dissertation proposes a low-complexity two-dimensional peak cancelling method.This method regards the clipping noise as a combination of dual-band parabola,so as to construct clipping signals based on the combined effect of dual-band signals.The simulation results show that the proposed two-dimensional peak clipping method is suitable for the moderate clipping threshold.Since no filtering or correlation processing is required,the calculation complexity is greatly reduced with better PAPR suppression and approximate bit error rate(BER)performance.3)A real-value polynomial model that can compensate for the phase modulated-phase modulated(PM-PM)and the phase modulated-amplitude modulated(PM-AM)distortion of the transmitter is proposed.The traditional method uses the memory polynomial to model the power amplifier,while neglecting the fact that the phase variation of the wideband signal will result in distortion in the transmitter linear broadband filter and the power amplifier.In this dissertation,the laws of PM-PM and PM-AM distortion in the transmitter are analyzed and a real-value polynomial model is used to compensate for PM-PM and PM-AM distortion and the strong nonlinear characteristics of the transmitter.The proposed model takes the real variable of IQ component as the input of the model,which contains the phase information as well as the amplitude information of the input signal.The proposed model can characterize and compensate for PM-PM and PM-AM distortion generated under wideband signal input,making balance between accuracy and amount of calculation.The test results and related analysis show that the real-value polynomial model can characterize the distortion,and can more accurately characterize the distortion characteristics of the transmitter.Moreover,the calculation complexity of real variable is lower than that of complex variable and reduce the calculation amount.This model takes into account the problems of distortion compensation performance and computational complexity of the transmitter.4)A lookup-table(LUT)model of a power amplifier with a multi-order interpolation function is proposed.The traditional power amplifier models do not combine the inherent nonlinear input and output characteristics of the power amplifier,and are characterized using general models such as memory polynomials or LUT models,resulting in unnecessary calculations complexity.In fact,the inherent nonlinearity of the power amplifier is linear when being driven by a small signal.As the amplitude increases,the nonlinearity gradually increases.In response to this phenomenon,this dissertation proposes a LUT model of a transmitter based on a multi-order interpolation function.In the case of a small signal,the linear interpolation is used to fit the nonlinear characteristics of the power amplifier.As the input signal amplitude increases,the order of the interpolation function increases to provide sufficient degrees of freedom to fit the increased nonlinearity.At the same time,according to the continuity and smoothness of the fitting function,the relationship between the function coefficients is analyzed in advance,thereby reducing the number of model parameters to be extracted,and further greatly reducing the complexity of model extraction.The simulation results show that the LUT model based on the multi-order interpolation function proposed in this dissertation greatly reduces the model extraction complexity under approximate performance compared with the traditional nonlinear compensation model.5)A weighted memory polynomial model based on the characteristics of the signal and the power amplifier is proposed.The distortion characteristics of the power amplifier depend on the amplitude of the input signal,while the traditional power amplifier model take the distribution characteristics of the input signal into consideration,and the used general equivalent accuracy model fails to achieve the non-linear compensated optimal effect.In this dissertation,in combination with the characteristics of the input signal,the traditional memory polynomial model is weighted according to the influence of its amplitude on the distortion and the distribution of the amplitude.While the amount of calculation is kept unchanged,the large signal and the high-probability signal are weighted to reduce its resulting nonlinearity,thereby improving the overall nonlinear compensation effect.The test and analysis results show that,compared with the traditional general memory polynomial model,the weighted memory polynomial model proposed in this dissertation can take into account the complexity of the algorithm and the performance of nonlinear compensation. |
PDF Full Text Request