Symbol decision method of color-independent visual-MIMO system using a dynamic palette

Jai-Eun Kim, Ki-Doo Kim

Abstract: In color-independent visual multiple-input multiple-output (hereinafter, “visual-MIMO”) systems that use color-space-based modulation techniques, determining the color of the received symbols using image sensors is difficult. In this study, we apply the K-means based color quantization method at the receiving end to minimize the influence of the wireless channel noise. In addition, we propose a method to determine the symbol color using a dynamic palette that was generated based on the characteristics of a color-independent visual-MIMO system. Finally, by comparing the position of the received symbol with the position of the transmitted symbol in the color coordinate space, we prove that the proposed method improves the accuracy of symbol decision (i.e., symbol error rate performance).