Single Snapshot-Spatial Compressive Beamforming for Azimuth Estimation and Backscatter Reconstruction

Development of radar signal processing is still emerging until this age including its capability to detect targets. In this paper, spatial compressive beamforming (SCB) method based on compressive sensing (CS) is applied in spatial domain studied to improve the azimuth angle estimation (AAE) of received backscatter FMCW signals as well as beamforming algorithm. Since only few azimuth angles occupied by the signals, in single snapshot form, they present sparse signals that can be reconstructed using a sparse recovery method such as LASSO. The use of $M$ elements rather than $N$ element where $M < N$ is accomplished by applying compression matrix C from Gaussian matrix and yields a compressive array. The rffect of noise power to acuuracy of the reconstruction is investigated. Performance of SCB compared to classical beamforming is evaluated as well in case of close and far targets. Results show the azimuth resolution of SCB with 181 angular grid points can reach up to 2 degree accurately while classical beamforming gives lower resolution about 15 degree. By choosing the regularization parameter $\lambda$ carefully in SCB, the replicated single snapshot backscatters are accurate enough since relative true error (RTE) achieves 0.85% for two closely adjacent targets less than 15 degree where classical beamforming presents 253.34%.