论文部分内容阅读
分振幅光偏振仪(DOAP)是一种高速测量光波偏振态的传感器。提出了一种基于人工神经网络(ANN)的分振幅光偏振仪的数据处理方法,将分振幅光偏振仪中电路系统输出的电信号作为神经网络的输入,入射光的斯托克斯参数作为神经网络的输出,建立一个前向多层神经网络模型。通过网络训练,使该网络确立了电路系统输出电信号与入射光斯托克斯参数之间的映射关系。由测量时得到的电信号,利用训练后的神经网络可以计算出待测的入射光的斯托克斯参数。测试结果表明,在测量精度方面,该方法获得的斯托克斯参数的总均方根偏差为1.9%,略优于基于矩阵运算的数据处理方法。
Split Amplitude Polarimeter (DOAP) is a sensor that measures the polarization of light waves at high speeds. A data processing method based on Artificial Neural Network (ANN) is proposed. The electrical signal output from the circuit system in the polarization amplitude polarimeter is taken as the input of neural network. The Stokes parameters of the incident light are taken as Neural network output, a forward multi-layer neural network model. Through network training, the network establishes the mapping relationship between the output electrical signal of the circuit system and the incident Stokes parameters. From the measured electrical signals, the Stokes parameters of the incident light to be measured can be calculated by using the trained neural network. The test results show that the total root mean square deviation of the Stokes parameters obtained by this method is 1.9% in terms of measurement accuracy, which is slightly better than the data processing method based on the matrix operation.