Abstract: The seismic method is often used to analyze the correlation between porosity and permeability of caved coal and rock in goaf. Most of the existing studies adopt the effective medium theory, which convert the anisotropy and inhomogeneity of caved coal and rock in goaf into equivalent medium parameters. The influence of particle size, porosity, spatial characteristics and other factors on the seismic wave is less considered. Taking the caved coal and rock in goaf as the research object, the broken coal and rock samples are prepared according to the actual particle gradation and porosity of the caved coal and rock in goaf. The porosity and wavelength-particle size ratio of the caved coal and rock samples are verified to be consistent with those of the caved coal and rock in the actual goaf. On this basis, the influence of coal and rock particle size, porosity and seismic wave frequency on wave velocity, amplitude attenuation coefficient and permeability is analyzed through experiments.The P-wave velocity in broken coal and rock decreases with the increase of porosity, increases with the increase of particle size, and is less affected by seismic wave frequency. The amplitude attenuation coefficient increases with porosity and decreases with the increase of particle size, and is more easily affected by seismic wave frequency when porosity is larger. The permeability of broken coal and rock increases with the increase of porosity and particle size. According to the Kozney-Carman equation, the permeability prediction formula of coal and rock in goaf is established based on the experimental results. The calculated permeability of caved coal and rock in goaf is 1 225× 10⁻¹²-178 930×10⁻¹² m². In order to eliminate the influence of particle size on the amplitude attenuation coefficient of seismic wave and the permeability of coal and rock samples, the amplitude attenuation coefficient of unit wavelength (amplitude attenuation factor) is proposed. The empirical relationship between the amplitude attenuation factor and the permeability is obtained by least square fitting. The study provides a thought for preliminarily judging the permeability of caved coal and rock in goaf.