Abstract: ISS (in-seam seismic) exploration is a significant method for underground coal mine exploration. It has the advantages of strong energy and long transmission distance through the transmission of waves in coal seams to detect the fault or other geologic anomalies. Research on transmitted ISS exploration is relatively mature, and it has been widely used in underground coal mine production. In recent years, reflected ISS exploration is getting more and more attention. Detection of fault drop by reflected ISS exploration is the future research direction, and its theoretical basis is not sufficient. In order to clarify the relationship between in-seam wave reflected energy and fault drop, six fault models with different fault drop are established. In those models, fault drop varies from 1/8 times coal thickness to 2 times coal thickness. Observation system is designed for those models and numerical simulation is performed by high order staggered grid finite difference algorithm. By extracting the reflection wave amplitude information of six different models, the reflection wave energy with offset and fault drop change characteristics were analyzed and the effect of band-pass filtering on the energy distribution of the reflected in-seam wave were analyzed too. The analysis results indicate that: while fault drop is less than one time of the coal seam thickness, the reflection amplitude gradually increases with the increase of fault drop. When fault drop is larger than coal seam thickness, reflection amplitude decreases with increase of fault drop. In the situation of various fault drops, the trend of amplitude versus offset is consistent, but when fault drop is larger than 3/4 times of the coal seam thickness, the reflected energy of the near and far offset is significantly different. The 100～150 Hz and 150～200 Hz band-pass filters have significant effects on the energy of reflection in-seam waves with a fault drop less than 1/2 coal thickness and greater than 3/4 coal thickness, respectively. Therefore, the detection of fault drop based on the energy distribution characteristics of the reflected in-seam wave has sufficient theoretical basis and practical application value.