Indoor visible light communication channel equalization method based on improved K-means

The invention provides an indoor visible light communication channel equalization method based on improved K-means, and the method comprises the steps: 1, enabling a photoelectric detector to receive an optical signal transmitted through an optical wireless channel, and converting the optical signal into an electric signal; 2, sequentially performing analog-to-digital conversion (A/D), serial-to-parallel conversion (S/P), CP deletion and Fourier transform (FFT) on the electric signal; 3, mapping the data subjected to FFT to a three-dimensional space by using a BP neural network; step 4, clustering is carried out on the mapped data by using an IC-Kmeans algorithm; and 5, demodulating and recovering the original binary sequence. According to the invention, an IC-Kmeans and NNIC-Kmeans balancing technology is provided; an auxiliary training sequence is introduced into the IC-Kmeans, so that the K-means equalization complexity is reduced; according to the method, the NNIC-Kmeans is introduced into the neural netw