Research On The Reduction Algorithm Of PAPR Of OFDM In Visible Light Communication Systems

Because of its fast transmission speed,high security and easy implementation,the visible light communication(VLC)has become one of the hot topics in the field of communication.The VLC system with the orthogonal frequency division multiplexing(OFDM)is able to effectively resist multipath interference and improve the data transmission rate and spectrum efficiency,but the resulting high peak-to-average power ratio(PAPR)signals will lead to serious nonlinear distortion.On the basis of the nonlinearity transmission characteristics of direct current-biased optical orthogonal frequency division multiplexing(DCO-OFDM)system,improved PAPR reduction algorithms are proposed and better PAPR reduction performance is obtained.Specifically,the major contents are as follows.For the DCO-OFDM system,the branch and bound methods based tone injection(BBM-TI)scheme is proposed with the purpose of reducing the dimension and peak power of the optimization problem.An improved constellation extension tone injection(TI)scheme is first presented,in which the number of the average degrees of freedom is less than 2.Then,the PAPR reduction is formulated as a particular integer combinatorial optimization problem.Next,the BBM method is applied to solve the integer combinatorial optimization problem.By splitting the superior branches and pruning the inferior branched,the close optimal solution is obtained.The simulation results and theoretical analysis reveal that the BBM-TI scheme not only obtains the better PAPR reduction performance,but also has a lower average power.Furthermore,the computational complexity of the BBM-TI scheme has been greatly reduced when compared with the semidefinite relaxation based TI scheme.The tone reservation(TR)scheme for reducing the PAPR is studied in the DCO-OFDM system,and the tone reservation based on the time domain kernel matrix(TKM-TR)scheme and the enhanced TKM-TR scheme are applied to the VLC system.It is found that the TKM-TR scheme has the drawback of severe tailing.Although the enhanced TKM-TR scheme can overcome this drawback to some extent,it cannot achieve the satisfied results.Therefore,an improved TKM-TR scheme based on the optimization of the scaling factors is proposed,in which the scaling factor is set to be zero in the order of clipping noise until the peak regrowth being eliminated.The simulation results show that the improved TKM-TR scheme achieves the more PAPR reduction than the enhanced TKM-TR,which efficiently addresses the tailing problem in TKM-TR.In addition,since the appropriate clipping ratio(CR)is very important in the reduction of PAPR,the particle swarm optimization based TR scheme is proposed to obtain a better CR,which correspondingly achieves the better reduction of PAPR and the lower BER performance.