We describe the scientific motivation, basic principles, and feasibility of a relatively new measurement technique, radio tomography, and show how it can be used to investigate the Earth's magnetosphere. We demonstrate that a multisatellite radio tomography experiment can produce two-dimensional images of plasma density in the Earth's magnetosphere at sufficient spatial (1/2 RE) and temporal (~10 s) resolution to address key problems of magnetospheric physics. The imaging technique incorporates well-established radio science methods and computed tomography. Several coplanar satellites are required in orbits that encompass the imaged area. We suggest that the large-scale images are more valuable when combined with in situ observations, supporting an unambiguous interpretation of the in situ data and an investigation of the interdependence of small- and large-scale plasma processes. ¿ 2000 American Geophysical Union