Blind deconvolution on graphs: Exact and stable recovery

Chang Ye, Gonzalo Mateos

Abstract: Highlights•We study a blind deconvolution problem on graphs.•A requirement on invertibility of the graph filter enables an efficient convex relaxation.•We derive sufficient exact recovery conditions in the noise-free setting.•Stable recovery when the observations are corrupted by a small amount of noise.•Results broaden the scope of blind deconvolution of temporal signals to graphs.