Beamforming and signal processing with sparse, random arrays for space based radar

We investigate moving target detection with a space-based cluster of radar satellites, modeled as a random sparse array. For a linear array which moves along its axis, the optimum space-time processing leads to a DPCA-like solution. We analyze the performance degradation with array pitch (satellite cluster rotation) and yaw (Earth rotation) and known random element position errors. Results show that the signal/clutter improvement factor is an extremely sensitive function of the pulse repetition frequency and that it varies over a large dynamic range. Achievable improvement factor maxima are moderately degraded by array pitch and strongly degraded with increasing position errors. Yaw effects appear to be negligible. In general, with increased randomness, fewer elements contribute to form DPCA pairs so that satellite usage efficiency (and SNR) is degraded.