| The rapid increase of IP data traffic has dirven the single channel rate increase from 10Gbit/s to 40Gbit/s and higher rates. The increase of the rate induces a reduction in the duration period of the symbol, which leads high-speed transmission systems to have a worse tolerance to chromatic dispersion, polarization mode dispersion and nonlinearity. At the same time, higher OSNR is required, and intrachannel nonlinearity becomes the dominating nonlinear effect in high-speed transmission systems due to its narrow pulse transmission. Various technologies have to be used to face these challenges, and one of the most effective methods is the use of novel modulation formats, such as QPSK, DQPSK and OFDM. Novel modulation formats can significantly increase the performance of high-speed transmission systems. However, it still exists or brings some other problems to high-speed transmission systems by novel modulation formats, when it is used to solve some challenges. For example, OFDM have the ability to compensate for CD and PMD. However, it is very susceptible to nonlinearities due to narrow frequency interval and high peak to average power ratio of the OFDM signal. Moreover, OFDM with coherent receiver is sensitive to I/Q (inphase/quadrature phase) imbalance. As to DQPSK, it has high toleranc to CD and PMD, and can effectively cancel nonlinearity, but it has limited effectiveness of intra-channel nonlinearity, which is the dominating nonlinear effect in high-speed transmission systems, cancellation. Concerning QPSK, it has high receiver sensitivity when the coherent detection is used, but it is also susceptible to I/Q imbalance.This thesis focuses on the existing problems and key technologies of novel modulation formats in high-speed transmission systems. The main contributions of this doctoral dissertation are summarized as follows.1. The nonlinearity of optical OFDM is investigated. Sinse OPC has limited effectiveness of nonlinearity cancellation in transmission links using lumped amplification, an optimized OPC configuration is proposed, and it is used in optical OFDM systems for nonlinearity compensation. Some proper length of fiber and EDFA are added in the link to optimize the OPC configuration. The optimized OPC configuration has symmetrical nonlinear regions distribution with respect to zero accumulated dispersion, which leads to an increase in the effectiveness of nonlinearity cancellation of OPC. To test the optimized OPC configuration, numerical simulations are performed in a 40-Gb/s OFDM system. Simulation results show that, the OPC can partially compensate for nonlinear distortions. However, the nonlinearity can be significantly cancelled by the optimized OPC configuration, which leads to effective improvement of system performance. Compared with OPC, the maximal Q factor, nonlinear threshold, and transmission distance of optimized OPC configuration increase by over 1.6 dB,2 dB, and 1 times, respectively.2. The intra-channel nonlinearity of DQPSK is investigated. A compensation scheme based on alternate polarization is proposed for intra-channel nonlinearity cancellation in 40-Gb/s DPQSK system. The alternate polarization modulates the adjacent symbols of DQPSK to different polarization by using polarization modulator, which produces orthogonal polarization of neighboring symbols and effective cancellation of intra-channel nonlinear impairments. Numerical simulations are performed in 40-Gb/s DQPSK system with alternate polarization, and the conventional DQPSK system is also simulated. The simulation results show that, the intra-channel nonlinearity has significant effect on the conventional DQPSK, which is effectively cancelled by alternate polarization. The results also show that the DQPSK with alternate polarization has about 2dB and 4dB increase in the maximal Q factor and the nonlinear threshold, respectively.3. The receiver I/Q imbalance in QPSK and OFDM system with coherent detection is investigated. A novel compensation algorithm based on the statistical properties of the In-phase (I) and Quadrature-phase (Q) of the received signals is proposed. This approach firstly calculates the mean and variance of I and Q components of the received signals. Simultaneously, the correlation coefficient of I and Q signal is also calculated. Then, the amplitude and phase imbalance parameters of the receiver are estimated by using the calculated parameters. Finally, the estimated imbalance parameters are adopted to compensate for the receiver I/Q imbalance. To test the compensation algorithm, numerical simulations are performed in 40-Gb/s OFDM system and QPSK system with coherent receiver, respectively. Simulation results show that, the proposed approach can effectively cancel receiver I/Q imbalance, and increase the system performance. The Q factor can increase, by using the compensation algorithm, to the value which is obtained in the system without I/Q imbalance. |
PDF Full Text Request