Abstract: The existence of subsidence column causes difficulties in coal mining, therefore, according to the structure, physical parameters and acoustic characteristics of coal seam collapse pillar, the coal powder, gypsum and cement are selected as raw materials, and the “rock-coal (containing collapse pillar)-rock” three-layer model is designed to study the coal seam with collapse pillar. The acoustic instrument controls the amplitude and frequency of the source, and the channel wave data is collected by transmission method. The amplitude energy attenuation coefficients of each channel are calculated by using the transmitted slot wave data, the wave velocity is analyzed by using the slot wave data waveform, and the existence position of collapse column is determined by using the time-frequency analysis method of Smoothed Pseudo Wigner-Ville distribution (SPWVD). The results show that the collapse column in the coal face can be effectively identified by analyzing the velocity of the transmitted slot wave and using SPWVD to analyze the energy characteristics of the transmitted slot wave, the collapse column in coal face can be identified effectively based on the maximum amplitude attenuation coefficient of slot wave.