Research On Underwater Optical Communication Channel Estimation For DCO-OFDM System

With the increasing demand for reliable,fast and efficient communication in underwater applications,underwater visible light communication has attracted more and more attention in recent years.Nonetheless LED-based visible light communication,is limited in modulation bandwidth.The introduction of OFDM modulation technology with high communication rate and spectrum utilization rate can improve this shortcoming.In OFDM-based underwater optical communication,accurate channel state information(CSI)is required for channel equalization.However,accurate underwater channel estimation remains a challenging problem due to the complex channel conditions underwater environment.Therefore,based on the DCOOFDM system,this paper conducts related research on underwater visible light communication channel estimation algorithm.The main work contents are as follows:(1)Firstly,this paper introduces the technical principle of DCO-OFDM system in detail,and then analyzes the influence of absorption and scattering on the underwater propagation of light.According to the characteristics of underwater transmission of light,the attenuation model of underwater optical communication channel was established in combination with Beer-Lambert's law.Finally,based on this model,linear processing is performed to obtain the corresponding channel frequency domain response model.(2)Research on channel estimation algorithm of DCO-OFDM system based on pilot is conducted.First,the traditional channel estimation algorithms based on the pilot-based DCO-OFDM system are studied,including: LS algorithm,LMMSE algorithm,SVD algorithm.Subsequently the simulation analysis and performance comparison of these several algorithms are performed.Then,according to the characteristics of DCO-OFDM system,a DFT channel estimation algorithm is given based on the conjugate symmetry of channel frequency response.In addition,the DFT underwater optical channel estimation algorithm is improved,which uses the noise power to set the threshold value.As a result,it can eliminate the influence of the noise in the cyclic prefix,reduce the complexity of the traditional estimation algorithm and improve the estimation accuracy.(3)Research on underwater optical communication channel estimation based on compressed sensing is conducted.Aiming at the problems of large pilot overhead and low frequency band utilization of DCO-OFDM system in traditional LS channel estimation algorithms,an underwater optical channel estimation method based on compressed sensing multipath matching pursuit(MMP)is given for DCO-OFDM system.By using the frequency domain response of the pilot of the DCO-OFDM system to construct a compressed sensing model,the MMP algorithm is used in underwater optical channel estimation.Then the channel time domain impulse response is estimated,and the smallest residual is selected from multiple candidate sets as the final.The support set can effectively improve the reconstruction performance.Simulation results show that,in DCO-OFDM system,compared with the LS algorithm,the algorithm saves the number of pilots and has better estimation performance;when the number of pilots is less,its estimation accuracy is better than the traditional orthogonal matching tracking algorithm.