The shadowing effect caused by the human body can have a considerable influence on indoor visible light communications (VLC) channel characteristics. A shadowing ray tracing (SRT) algorithm to investigate the human shadowing channel characteristics is proposed. To validate the effectiveness of the SRT model, an extensive indoor human shadowing channel measurement has been presented. Both scenarios, the line-of-sight (LOS) and the non-line-of-sight (NLOS), have been measured. Theoretical results from the SRT model are validated against the measurement results for channel path loss. The results from measurements and simulations show that receiver rotations can mitigate human shadowing effect under the LOS scenario. Under the NLOS scenario, the VLC system becomes more sensitive to human shadowing when the distance between the LED and the receiver is large. Finally, more simulation results on the channel root mean square delay spread are also presented to analyze its relation with human shadowing distance to the receiver.