DOA estimation based on smoothed sparse reconstruction with time-modulated linear arrays

A novel direction of arrival (DOA) estimation method based on smoothed sparse reconstruction with time-modulated linear arrays (TMLA) is proposed in this paper. With the analysis of the noise variation after time modulation, a closed-form expression for the noise covariance under the optimized unidirectional phase center motion time sequence scheme is derived. Then a reweighted continuous function is designed for the smooth approximation to ℓ0-norm based on the orthogonality between the signal subspace and the noise subspace. Simulation results demonstrate that, compared with conventional DOA estimation method with time-modulated arrays, the proposed method possesses higher estimation accuracy and lower computational complexity, especially for the scenario with small snapshot and low signal-to-noise ratio. •The nonuniform noise is derived in a closed-form expression and whitened accordingly.•A reweighted continuous function is designed for the smooth approximation to ℓ0-norm.•A steepest ascent approach is applied to solve the objective function in an iterative fashion.