To what extent can particulate random media be characterised using direct wave backscattering from a single receiver/source? Here, in a two dimensional setting, we show using a machine learning approach that both the particle radius and concentration can be accurately measured when the boundary condition on the particles is of Dirichlet type. Although the methods we introduce could be applied to any particle type. In general backscattering is challenging to interpret for a wide range of particle concentrations, because multiple scattering cannot be ignored, except in the very dilute range. Across the concentration range from 1% to 20% we find that the mean backscattered wave field is sufficient to accurately determine the concentration of particles. However, to accurately determine the particle radius, the second moment, or average intensity, of the backscattering is necessary. We are also able to determine what is the ideal frequency range to measure a broad range of particles sizes. To get rigorous results with supervised machine learning requires a large, highly precise, dataset of backscattered waves from an infinite half-space filled with particles. We are able to create this dataset by introducing a numerical approach which accurately approximates the backscattering from an infinite half-space.PACS numbers: 42.25.Dd,43.20.Fn,05.10.Ln Under close inspection, many materials are composed of small randomly distributed particles or inclusions. So it is no surprise that the need to measure particle properties, such as their average size and concentration, spans many physical disciplines. For quick non-invasive measurements, waves, either mechanical, electromagnetic or quantum, are the preferred choice. However, measuring a broad range of particle concentrations and sizes is still an open challenge. For high concentrations the wave undergoes multiple scattering, which requires specialised methods to compute and interpret. And further, measuring a wide range of particle sizes means a wide range of frequencies needs to considered.The type of wave used depends on the type of particle: acoustic waves are used to measure liquid emulsions 1 , sediment on the ocean floor 2 and polycrystalline materials 3 . Microwaves are vital in remote sensing of ice 4 ; optics for aerosols 5 and cellular components, both micrometer 6 and nanoscale 7 structures, among many other applications. In all these applications, there are cases when transmission experiments are impractical, because either the material is too opaque or, for example, has an unknown depth. The next natural choice is to use reflected, or backscattered, waves.Here we ask can one source/receiver measure the properties of a random particulate medium? And is it possible a) arturgower@gmail.com; https://arturgower.github.io/ b) gowerrobert@gmail.com; https://perso.telecomparistech.fr/rgower/ to do so without measuring the backscattering for a range of scattering angles, and without knowing the depth of the medium? Figure 1 illustrates a backscattered wave in time measured...