Abstract: Optimizing the layout of communication base stations and positioning substations for mobile communication, Internet of Things (IoT) communication, personnel and vehicle positioning in mines requires calculating electromagnetic wave propagation losses in mines. Therefore, it is necessary to study the calculation method of electromagnetic wave propagation loss in mines. However, Existing statistical calculation methods for electromagnetic wave propagation loss in general, outdoor open-space, indoor, tunnel, and mine environments primarily consider the effects of distance and frequency but do not account for underground-specific factors such as tunnel cross-sectional geometry. As a result, these methods yield significant errors when calculating electromagnetic wave propagation loss in the complex underground environment of coal mines. A relative wavelength calculation method for electromagnetic wave propagation loss in mines has been proposed, which depends not only on frequency and distance but also on factors such as the cross-sectional dimensions of the tunnel.The electromagnetic wave propagation characteristics of mines are revealed as follows: in the mine line-of-sight (LOS) scenarios: ① In auxiliary transportation tunnels, electromagnetic wave propagation exhibits a turning point. When the relative wavelength is less than 10, the relative spatial scale of propagation is small, making the electromagnetic wave propagation significantly affected by the tunnel, which hinders their propagation. Under these conditions, the propagation loss exponent in the mine exceeds that in free space, and the propagation loss increases with increasing wavelength. When the relative wavelength is greater than 10, the relative spatial dimensions for electromagnetic wave propagation are large, causing the wave to propagate within the tunnel similarly to free space; consequently, propagation loss increases as the wavelength decreases. ② Belt conveyors hinder the propagation of electromagnetic waves in tunnels, and the presence of belt conveyors in tunnels increases the loss of electromagnetic wave propagation. In the mine non-line-of-sight (NLOS) scenarios: ① The propagation of electromagnetic waves is influenced by both tunnel cross-section and frequency, with frequency having a greater impact on the propagation of electromagnetic waves than relative waves. ② In tunnels with the same cross-section, the average electromagnetic wave propagation loss exponent is 2.77 for the curved tunnel, 2.83 for the branch tunnel with the transmitter located in the side branch, and 3.02 for branch tunnels with the transmitter located in the main tunnel. The average loss exponent increases in this order. Therefore, electromagnetic wave propagation loss is lower in curved tunnels than in branch tunnels. Moreover, in branch tunnels, deploying the transmitter in the side tunnel yields better performance than deploying it in the main branch. Verification using measured mine data indicates that the proposed method achieves a mean absolute error (MAE) of 4.4 dB. Compared to commonly used statistical calculation methods based on channel standards, this represents a reduction in MAE of 2.1 to 6.4 dB, thereby improving the accuracy of electromagnetic wave propagation loss calculations in mines.