Sparse recovery of radar echo signals using Adaptive Backtracking Matching Pursuit

Compressive Sensing (CS) combines signal sampling and signal compression. CS directly acquires a signal provided it is either sparse by itself or sparse in some transform domain. In radar applications, it is not always possible to sample the radar signal ideally. Further, consecutive radar echo signals show some correlation which may be exploited. In this work, we start by modelling the radar echo signal and adopting a sensing mechanism to acquire it. For CS reconstruction, we propose Adaptive Backtracking Matching Pursuit which makes use of the `partially known support' to reconstruct the sparse version of radar echo signal.