Channel Estimation Method Based On Compressed Sensing In MIMO-VLC

Combining multiple input multiple output（MIMO）technology with visible light communication（VLC）systems can effectively achieve spatial resource reuse and improve the communication performance of VLC systems.Accurate channel estimation is required to obtain channel state information for reliable communication in MIMO-VLC systems.The commonly used Least Squares（LS）algorithm and Linear Minimum Mean Square Error（LMMSE）algorithm require a large amount of pilot overhead and estimation error,making it difficult to meet the requirements of MIMO-VLC for reliable communication.This paper applies compressed sensing（CS）technology to MIMO-VLC channel estimation to improve system channel estimation performance and further ensure system communication quality.The specific research content includes the following aspects:1)Based on the Lambertian radiation model,theoretical analysis is conducted on the direct and non-direct links of the VLC channel,and the channel gains of these two different links are modeled to construct the MIMO-VLC channel impulse response.Simulation experiments are conducted to analyze the multipath characteristics and sparsity of the VLC channel.2)In response to the low communication rate caused by insufficient LED modulation bandwidth,MIMO technology and orthogonal frequency division multiplexing（OFDM）technology are combined and applied to the VLC system to increase system capacity and spectral utilization,and improve system transmission rate.The MIMO-VLC system model is established,and the channel estimation method is studied.Finally,LS channel estimation algorithm and discrete Fourier transform（DFT）channel estimation algorithm are analyzed.3)CS channel estimation methods are studied,and CS technology is used to reduce the pilot overhead of channel estimation while improving channel estimation performance.To address the problem of excessive iteration times and slow operation efficiency of the sparsity adaptive matching pursuit（SAMP）channel estimation algorithm,this paper proposes a DFT sparsity prediction and observation matrix optimization based prediction-sparsity adaptive matching pursuit（DFT-OSAMP）algorithm.Based on the SAMP algorithm,a method of predicting the sparsity of the channel using DFT is introduced to quickly approach the actual sparsity,improve algorithm efficiency,and an observation matrix optimization method is also introduced to improve channel estimation performance.Two-input two-output indoor MIMO-VLC system experimental platform was established,and the channel estimation method based on compressive sensing was experimentally verified in an actual indoor environment,and the system error rate performance was analyzed.Theoretical research and experimental results show that the channel estimation based on CS algorithm can effectively improve communication reliability.When the bit error rate is 10^-3,the SAMP algorithm performance is 4.5d B better than the LS algorithm.Moreover,with only 16pilots,the SAMP algorithm outperforms the LS algorithm with 32 pilots,reducing 50%of pilot overhead.The proposed algorithm in the paper further improves the performance of the SAMP algorithm using sparsity prediction and observation matrix optimization.At a bit error rate of 10^-3,the performance of the proposed algorithm in the paper is 0.3d B better than that of the optimized SAMP algorithm,and the channel estimation efficiency is improved by 69%compared with the SAMP algorithm.