Performance Research On Multiuser MIMO Visible Light Communication System Based On BD Precoding

Visible light communication(VLC)uses light-emitting diodes(LED)as the emission light source,and combines lighting and communication technology,which has advantages of good confidentiality,no spectrum license,and low cost.In a typical indoor environment,multiple LED arrays are often installed to ensure sufficient lighting requirements.There are usually multiple user terminals distributed in the room,which together with the LED array constitute a multi-user,multi-input and multi-output(MIMO)visible light communication system that can effectively improve the VLC transmission rate.However,multiple users will cause interference between users,which will reduce the bit error rate performance of users in the system.The precoding algorithm can effectively eliminate the multi-user interference(MUI).For each user terminal equipped with multiple optical receivers,the commonly used precoding algorithm in the multi-user MIMO visible light communication system is the block-diagonal(BD)precoding based on singular value decomposition(SVD).However,due to the high correlation of user optical channels,the VLC system based on this precoding algorithm has limitations that the bit error rate performance of different optical receivers of the same user terminal is quite different.To solve this problem,this paper proposes three precoding systems.Based on this,the main work done in this paper is as follows:(1)Firstly,this paper briefly summarizes the VLC system,which mainly includes three parts: optical transmitter,optical channel and optical receiver.In the optical transmitter part,the types,optical properties and modulation properties of white LEDs are introduced.In the optical channel part,two link models in optical wireless communication are introduced.In the optical receiver part,the signal receiving range of the optical receiver and the two types of noise that affect the the optical receiver are introduced.(2)In this paper,BD precoding based on geometric mean decomposition(GMD)is applied to the indoor VLC system,and a decision feedback equalizer is used at the user receiver to eliminate the interference between data streams,and the principle of the system is analyzed in detail.The final simulation results show that compared with the SVD-BD precoding system,the system can enable different receivers of the same user to obtain similar bit error rate performance and reduce the bit error rate of the user terminal.(3)In this paper,THP-BD evolved from nonlinear Tomlinson-Harashima precoding(THP)is applied to VLC system.However,the optical receiver noise in this system is amplified to different degrees.To solve this problem,the improved THP-BD precoding is applied to the VLC system.Finally,the simulation results shows that compared with the SVD-BD precoding system,both nonlinear precoding systems have a certain improvement in the bit error rate performance of the user terminal.But the performance of the two systems is not ideal,limited by the strong correlation between the optical receivers of the same user.(4)To solve the problem of strong channel correlation,this paper adopts the model of optical receiver with different field of view(FOV)for the same user terminal to reduce the correlation of user terminal channels,and finds the optimal FOV combination for the same user terminal optical receiver.Based on this FOV combination,the BER performance of the above three precoding systems is studied.The simulation results show that this FOV combination can effectively improve the bit error rate performance of each precoding system,and the GMD-BD precoding system has the best performance.On this basis,the influence of user terminal location on the bit error rate performance of GMD-BD precoding system is analyzed.