| Coherent optical orthogonal frequency division multiplexing(CO-OFDM)technique combines coherent optical communication and OFDM technique,and meets the requirements of high speed data transmission and become a main technical standards for many ultra high speed communication systems,which because of its obvious advantages in spectrum efficiency and anti-dispersion capacity.However,the high peak-to-average power ratio(PAPR)is a intrinsic problem of OFDM that the CO-OFDM have to inherit.This problem may increase the operating cost and degenerate the operating efficiency of the systems,it will also affect the correct recovery of the signal by receiving end.Thus,PAPR reduction is essential to improve the efficiency and stability of the systems.Therefore,the problem of high PAPR in the CO-OFDM systems are studied in this thesis at first,and then the techniques of PAPR reduction are compared and analyzed.Finally,the partial transmit sequences(PTS)technique with better performance is selected as the object of further research.The main contents are as follows:1.For the high PAPR in the systems,the three classification of PAPR reduction techniques are simply introduced in this thesis.Then the Clipping technique,selective mapping(SLM)technique and the PTS technique are probed,furthermore,their performances are compared and analyzed.The simulation results show that,the clipping technique is relatively easy to implement and has superior performance,but this technique will cause signals distortion.The SLM technique and the PTS technique both perform linear process on the signals,which will not distort the signals,and the performance of PTS reduction technique is better than SLM technique if the amount of computational complexity is the same.Therefore,the PTS technique is selected for further research.2.The basic theory of traditional PTS technique and several factors to affect its performance are analyzed.Aiming at the shortcomings of the high complexity in the traditional PTS scheme,a joint improved PAPR suppression algorithm,adopts the idea of hierarchical inverse fast Fourier transform and combines with the clipping technique,is proposed in this thesis.The simulation results show that,the proposed algorithm can reduce the computational complexity by 78% under the circumstance of the original portioned into 4 subblocks at the remaining stage n-d=5,and the number of sub-carriers is 256.And the PAPR reduction performance improved about 4.82 d B when compared with the original signals at the 4CCDF10-=.At the same time,the bit error ratio(BER)of the joint improvement algorithm is obviously better than the clipping algorithm.Therefore,the proposed joint improvement algorithm significantly improve the PAPR performance of the systems,at the same time,the computation of the system is reduced while the BER is considered.3.The FPGA design and implementation on the joint improved algorithm based on PTS technique are presented in this thesis.Firstly,basic structure and development process of FPGA are introduced.Secondly,the overall implementation plan of the improved algorithm is given,and the implementation plan of each module is introduced in detail.Finally,the FPGA data is imported into Matlab for joint simulation analysis,so as to verify the consistency of the FPGA and Matlab results.The consistency of simulation result indicates that the improved algorithm is feasible in hardware implementation,and establishes the basis for practical application. |
