There are many different rainfall processes with different attributes in terms of variations in space, time, and intensity. The monitoring of fast varying rainfall in space and time is a challenge for existing systems including radars, rain gauges and the opportunistic sensing systems with microwave signals from commercial wireless communication networks.

Recently, we have proposed to use the microwave signals from the communication links of Low Earth Orbit satellites to retrieve rainfall, using a standard least-squares based signal processing method. While the proposed method has a potential to achieve 3D rainfall retrieval with high spatial resolutions by using a large number of cheap ground receivers, it retrieves the rainfall field only once during one overpass of a satellite by assuming that the rainfall field does not change. To deal with a fast time-varying rainfall, in this research, we propose to use the algebraic reconstruction technique (ART) to retrieve the rain field multiple times during one overpass of a satellite by dynamically updating the retrieved rainfall with the use of the estimated signal-to-noise (SNR) of the received microwave signals from the satellite in real-time. To test the feasibility of the proposed method, we utilize a synthetic rainfall field generated by the Weather Research and Forecasting model. For a circular LEO satellite trajectory with a satellite-to-ground signal path loss model applied to the synthetic rainfall field, our preliminary simulation results have demonstrated that the proposed ART based method has a great potential to achieve 3D retrieval of rainfall with high spatial and temporal variations.