Research On Peak-to-Average Power Ratio Reduction Technologies Based On Slm In Optical OFDM Systems

Optical orthogonal frequency division multiplexing?O-OFDM?system has been adopted as a standard technology in many high-speed data transmission systems because it has the high spectral efficiency,anti-intersymbol interference?ISI?,anti-channel fading,anti-dispersion and other advantages.However,O-OFDM technology also has many shortcomings.One of the mainly shortcomings is that the transmitted signal has a relatively high peak-to-average power ratio?PAPR?.In order to recovery O-OFDM signals with high PAPR correctly,it is necessary to provide power amplifiers with high linearity at the transmitting end,otherwise it will lead to serious in-band distortion and unwanted out-of-band radiation,but there is no doubt that it will increase the operating costs of system.So it is of great significance and value for the research of PAPR reduction technology.Based on the PAPR reduction technology in O-OFDM system,this thesis analyzes the clipping technique in the non-distorting technology and the partial transmit sequences?PTS?and selective mapping?SLM?techniques in the pre-distortion technology,and studies the peak-to-average power ratio reduction technologies emphatically based on SLM.The main work is summarized as follows:1.The concept and basic definition of peak-to-average power ratio are expounded,and the three major peak-to-average power reduction techniques are analyzed comparatively in detail,and their advantages and disadvantages are also probed.The clipping technology,the PTS technology and the SLM technology are simulated and analyzed.The simulation results show that the clipping technology is relatively simple,and the drawbacks are that it will cause the signal distortion and the higher bit error rate.The PTS and SLM technologies belong to the probability technology,but the complexity of the SLM technology is lower than that of the PTS technology.In the same simulation case,the SLM technology has a slight PAPR reduction performance of about 0.5dB compared to the PTS technology at complementary cumulative distribution function?CCDF?of 10^-4(CCDF?28?10^-4).Thus,this thesis chooses the SLM technology as the object of the further study.2.The basic principle of the conventional SLM?C-SLM?scheme is analyzed,and the factors that affect the performance of SLM technology are further discussed.Aiming at the shortcomings of the C-SLM scheme with the high computational complexity,a low complexity SLM?LC-SLM?scheme is proposed in this thesis.By dividing a complex signal into two real signals,the real and imaginary parts of the signal are processed separately,and then the two real signals are reconstructed after inverse fast Fourier transform?IFFT?to obtain a new candidate signal.The computational complexity of the proposed scheme that generates M candidate signals is equivalent to that of the C-SLM scheme that generatesM² candidate signals.When the number of candidate signals is the same,the complexity of the proposed scheme is reduced by about 35%to 75%compared to the C-SLM scheme.In the same simulation case,the PAPR reduction performance of the proposed scheme is about 1dB higher than that of the C-SLM scheme atCCDF?28?10^-4.Thus,the LC-SLM scheme proposed in this thesis can significantly reduce the computational complexity and does not impair PAPR reduction performance.3.In view of the shortcomings of the poor PAPR reduction performance for the low complexity SLM scheme,a joint improvement scheme based on sub-optimal selection is proposed in this thesis.In order to make the scheme have low computational complexity and good PAPR reduction performance,it combines the sub-optimal selection and the modified widely linear SLM?MWL-SLM?scheme on the basis of the LC-SLM scheme,and improves the LC-SLM scheme.When the number of candidate signals is the same,the complexity is equal to that of the MWL-SLM scheme,but is about 70%to 80%lower than that of the C-SLM scheme.In the same simulation case,when the number of sub-blocks is the same,the PAPR reduction performance of the scheme is about 2.7dB and 0.7dB higher than that of the C-SLM scheme and the MWL-SLM scheme atCCDF?28?10^-4.When the number of candidate signals is the same,the PAPR reduction performance of the scheme is about 0.8dB and 0.6dB higher than that of the C-SLM scheme and the MWL-SLM scheme4.The transmitting side-band information in the SLM scheme is necessary,but it will lead to the decrease of the transmission efficiency of the system,so in view of this shortcomings of the SLM scheme,a joint improvement scheme based on no side-band information transmission technology is proposed in this thesis.On the basis of the LC-SLM scheme,the scheme that combines the side-band information processing technology and the sub-optimal selection of ideas is introduced to make a joint improvement.When the number of candidate signals is the same,the complexity is reduced by about 30%to 75%compared with the C-SLM scheme.In the same simulation case,the PAPR reduction performance of the scheme,compared with the C-SLM,the improved SLM?I-SLM?and the LC-SLM scheme,is improved by about1.3dB,1dB and 0.8dB respectively atCCDF?28?10^-4.The simulation results show that the scheme does not increase the bit error rate of the system,furthermore,the low complexity and the good PAPR reduction performance can be obtained.