Research On Adaptive Equalization Technology For Indoor Visible Light Communication

Visible light communication has the advantages of unrestricted spectrum and high communication security.It has broad application prospects in indoor LiFi and indoor positioning.Indoor visible light communication systems usually use channel equalization technology to solve the delay spread and inter-symbol interference caused by multipath effect.The adaptive equalization method based on the least mean square(LMS)algorithm cannot reduce the steady-state error while improving the convergence speed,and requires the use of training sequences,resulting in a decrease in communication efficiency of system.The double-mode blind equalization algorithm based on decision directed least mean square(DD-LMS)algorithm and constant modulus(CMA)algorithm converges slowly and there is unreliable communication under the condition of low signal-to-noise ratio.In the paper,aimed at the above problems,the following studies are carried out:(1)The inverse hyperbolic sinusoidal function is introduced into the fixed-step LMS adaptive equalization algorithm,making the step size factor in the algorithm follow the change of error signal dynamically,so the contradiction between convergence speed and steady-state error in fixed-step LMS algorithm can solved effectively.The error in the iteration formula of the weight coefficient of the algorithm is replaced by the expected value of the error,which reduces the effect of additive Gauss white noise on steady-state error.The simulation result shows that compared with the traditional variable step size LMS adaptive equalization algorithm,the improved LMS adaptive equalization algorithm with variable step size improves the convergence speed and anti-noise performance,reduces the steady-state error after convergence.(2)Combining CMA+DD-LMS dual-mode blind equalization algorithm with fractional spaced equalizer,the channel distortion is effectively compensated and the bit error rate of the system is reduced.Two blind equalization algorithms,DD-LMS and CMA,are improved by using variable step size algorithm,the convergence speed of CMA+DD-LMS dual-mode blind equalization algorithm is improved obviously.The difference between adjacent k iteration errors of variable step size CMA blind equalization algorithm is used as the switching condition of improved CMA+DD-LMS double-mode blind equalization algorithm,which reduces the number of false switching.The simulation result shows that the improved CMA+DD-LMS dual-mode blind equalization algorithm can converge stably in low SNR environment,the convergence speed and adaptability of the algorithm are improved while reducing the steady-state error and the bit error rate.