Intra-body communications based on near-field radio signals is a short range communication solution suitable for Body Area Networks wherein the human body is used as the transmission medium. This paper proposes a statistical model for the intra-body propagation channel based on experimental data of the human body transmission characteristics. The channel models are presented over three measurements configurations with several transmitter and receiver locations and considering two common scenarios: the human body not moving and walking. We define the statistical models in terms of most fitting probability density function to the experimental data of the received signal power among 15 families of probability distributions. These general models should be useful for simulating propagation effects in the laboratory and, therefore, we believe that these models will play a significant role in the design and evaluation of intra-body communication systems.