在发射机上，经过适当的数字信号处理器数据通过数模转换器（dac）转换成

在发射机上，经过适当的数字信号处理器数据通过数模转换器（dac）转换成双通道模拟信号，dac对IQ组件进行调制射频载波。另一方面，在接收器处，发送的RF信号与LO混合，IQ分量是在基带获得。这样的智商信号被转换成通过模数转换器（ADC）的数字域，以及<br>符号通过数字信号处理器解码。这种类型的接收器称为直接转换接收器。如果图（a）中的射频调制器和图（b）中的射频混频器替换为光学对应物，即光学IQ调制器和相位分集零差接受者，我们可以很容易地想象基于数字信号处理器的光发射机和接收机。基于dsp的发射机，包括奈奎斯特谱等功能成形、信号预失真/补偿和前向纠错（FEC）编码在商业上是重要的，但远不止这些本文的研究范围。高速数字集成电路的最新发展为在DSP核中处理高速电信号和以非常稳定的方式从零差检测信号中提取光载波复振幅的IQ分量提供了可能。然后解调20gb/s的QPSK信号相位分集零差接收机后接数字中的载波相位估计，尽管误码率测量仍在脱机执行。由于载波相位是在利用数字信号处理器进行零差检测后恢复的，所以这种类型的接收机现在通常被称为“数字相干接收机”尽管一个将LO相位锁定到信号相位的OPLL由于环路延迟问题，仍然难以实现，数字信号处理器电路变得越来越快，为我们提供了估计载波相位的简单而有效的方法。非常与OPLL方案相比，载波相位的快速跟踪大大提高了系统的稳定性。可以引入任何类型的多级调制格式使用相干接收机。而二进制调制格式的频谱效率限制在1位/s/Hz/极化，称为奈奎斯特极限，每个符号有mb位信息的调制格式可以达到最高mb位/秒/赫兹/极化的频谱效率。虽然已经使用光延迟检测来解调QPSK信号（mb＝2），但在多重性方面进一步增加这样的计划很难实现。数字相干接收机的另一个重要优点是后信号处理功能。数字相干接收机的IQ解调是完全线性的处理；因此，即使在检测和光信号处理功能作用于检测后的电级可进行光学滤波和色散补偿等光载波。因此，电气GVD补偿消除了色散补偿光纤。在检测到后也可以进行偏振对准通过将偏振分集方案引入到零差接收机中。利用数字信号处理器同时测量和处理水平极化和垂直极化的复振幅。PMD的偏振解复用和补偿也已被数字相干接收机所证明，其中笨重和慢偏振控制器和光延迟删除行。

在发射机上，经过适当的数字信号处理器数据通过数模转换器（dac）转换成双通道模拟信号，dac对IQ组件进行调制射频载波。 另一方面，在接收器处，发送的RF信号与LO混合，IQ分量是在基带获得。 这样的智商信号被转换成通过模数转换器（ADC）的数字域，以及<br>符号通过数字信号处理器解码。 这种类型的接收器称为直接转换接收器。 如果图（a）中的射频调制器和图（b）中的射频混频器替换为光学对应物，即光学IQ调制器和相位分集零差接受者，我们可以很容易地想象基于数字信号处理器的光发射机和接收机。 基于dsp的发射机，包括奈奎斯特谱等功能成形、信号预失真/补偿和前向纠错（FEC）编码在商业上是重要的，但远不止这些本文的研究范围。 高速数字集成电路的最新发展为在DSP核中处理高速电信号和以非常稳定的方式从零差检测信号中提取光载波复振幅的IQ分量提供了可能。 然后解调20gb/s的QPSK信号相位分集零差接收机后接数字中的载波相位估计，尽管误码率测量仍在脱机执行。 由于载波相位是在利用数字信号处理器进行零差检测后恢复的，所以这种类型的接收机现在通常被称为“数字相干接收机”尽管一个将LO相位锁定到信号相位的OPLL由于环路延迟问题，仍然难以实现，数字信号处理器电路变得越来越快，为我们提供了估计载波相位的简单而有效的方法。 非常与OPLL方案相比，载波相位的快速跟踪大大提高了系统的稳定性。 可以引入任何类型的多级调制格式使用相干接收机。 而二进制调制格式的频谱效率限制在1位/s/Hz/极化，称为奈奎斯特极限，每个符号有mb位信息的调制格式可以达到最高mb位/秒/赫兹/极化的频谱效率。 虽然已经使用光延迟检测来解调QPSK信号（mb＝2），但在多重性方面进一步增加这样的计划很难实现。 数字相干接收机的另一个重要优点是后信号处理功能。 数字相干接收机的IQ解调是完全线性的处理；因此，即使在检测和光信号处理功能作用于检测后的电级可进行光学滤波和色散补偿等光载波。 因此，电气GVD补偿消除了色散补偿光纤。 在检测到后也可以进行偏振对准通过将偏振分集方案引入到零差接收机中。 利用数字信号处理器同时测量和处理水平极化和垂直极化的复振幅。 PMD的偏振解复用和补偿也已被数字相干接收机所证明，其中笨重和慢偏振控制器和光延迟删除行。

On the transmitter, the data is converted into two channel analog signal by a digital to analog converter (DAC), which modulates the RF carrier of the IQ module. On the other hand, at the receiver, the RF signal transmitted is mixed with the lo, and the IQ component is obtained at the baseband. Such I.Q. signals are converted into a digital domain via an analog-to-digital converter (ADC), and<br>Symbols are decoded by a digital signal processor. This type of receiver is called a direct conversion receiver. If the RF modulator in figure (a) and the RF mixer in figure (b) are replaced by the optical counterpart, i.e. the optical IQ modulator and the phase diversity zero difference receiver, we can easily imagine the optical transmitter and receiver based on the digital signal processor. The transmitter based on DSP, including the function shaping of Nyquist spectrum, signal predistortion / compensation and forward error correction (FEC) coding are important in business, but they are far beyond the scope of this paper. The latest development of high-speed digital integrated circuits makes it possible to process high-speed electrical signals in DSP core and extract IQ components of complex amplitude of optical carrier from zero difference detection signals in a very stable way. Then demodulate the carrier phase estimation in the subsequent digital of the 20 GB / s QPSK signal phase diversity zero difference receiver, although the bit error rate measurement is still offline. Since the carrier phase is recovered after zero difference detection by digital signal processor, this type of receiver is now commonly called "digital coherent receiver". Although an OPLL which locks the lo phase to the signal phase is still difficult to realize due to the loop delay problem, the digital signal processor circuit becomes faster and faster, which provides us with estimation of carrier phase Simple and effective method. Compared with OPLL scheme, the fast tracking of carrier phase greatly improves the stability of the system. Any type of multistage modulation format can be introduced to use coherent receivers. The spectrum efficiency of binary modulation format is limited to 1 bit / S / Hz / polarization, which is called Nyquist limit. The modulation format with MB bit information per symbol can achieve the highest spectrum efficiency of MB bit / S / Hz / polarization. Although optical delay detection has been used to demodulate QPSK signals (MB = 2), it is difficult to achieve such a plan in terms of multiplicity. Another important advantage of digital coherent receiver is its post signal processing function. The IQ demodulation of the digital coherent receiver is a completely linear process; therefore, even after the detection and optical signal processing functions act on the detected electrical stage, optical carriers such as optical filtering and dispersion compensation can be carried out. Therefore, the electric GVD compensation eliminates the dispersion compensation fiber. After detection, polarization alignment can also be carried out by introducing polarization diversity scheme into the zero difference receiver. The complex amplitudes of horizontal polarization and vertical polarization are measured and processed simultaneously by digital signal processor. The polarization demultiplexing and compensation of PMD has also been proved by the digital coherent receiver, in which the cumbersome and slow polarization controllers and optical delay delete lines are included.<br>