Remote sensing via \(\ell_1\)-minimization (Q404256)

The paper considers the detection of locations and reflectivities of remote targets by sending probing signals from an antenna array and recording the reflected signals. A general sparse recovery result and condition number estimates for certain random matrices are derived. The recovery of sparse vectors for sensing matrices with dependent rows and columns associated with a class of bounded orthonormal systems is proved. The type of matrix includes the sensing matrix arising in the inverse scattering problem as a special case. The result assumes that the nonzero coefficients of the signal to be recovered have random phases. The assumption of random phases is then removed and the sparse recovery for the inverse scattering setup for signals with fixed deterministic phases is shown. Some numerical simulation results are presented to illustrate the findings.