基于Kronecker积的无线通信系统参量分离方法

针对无线通信系统中发射机非线性与多径衰落信道问题，提出了一种根据接收无线信号联合估计无线设备功率放大器（PA）非线性与无线信道单位脉冲响应的方法。首先，根据通信帧训练符号及发射机非线性模型构造卷积矩阵；接着，采用一次最小二乘（LS），根据接收信号估计PA的非线性模型系数与实际发送符号；然后，再采用一次LS，得到无线信道的单位脉冲响应估计；两估计可迭代或直接加取平均进行积累。理论推导与实验结果显示，采用过采样技术，所提方法可应用于单载波或多载波通信，实现PA非线性与无线信道单位脉冲响应的有效分离。新方法在无线通信的物理层射频指纹认证与信号可靠传输中具有应用价值。

A Kronecker product based parameter separation method for wireless communication systems

For the problem of transmitter nonlinearity and wireless multi-path fading channel in wireless communication systems,a combined estimation of the nonlinearity of power amplifier (PA) and impulse response of wireless channel with the received signal is proposed. Firstly,a convolution matrix is constructed based on the training symbols and model structure of the nonlinearity,then once least squares (LS) is used to estimate the coefficients of the model of the PA and actual transmitted symbols through the PA. Secondly,the estimation of the impulse response of the wireless channel is obtained with another LS. And the integration of the two estimations can be iterative or with direct summation and mean. Theoretical derivation and experimental results demonstrate that the method can be used in single-carrier or multi-carrier communication with oversampling technology,and can separate the nonlinearity of PA and impulse response of wireless channel effectively. The novel method can be used for the physical-layer radio frequency(RF) fingerprint authentication and reliable signal transmission.