Structural Sparsity in Multiple Measurements

We propose a novel sparsity model for distributed compressed sensing in the multiple measurement vectors (MMV) setting. Our extends concept of row-sparsity to allow more general types structured arising variety applications like, e.g., seismic exploration and non-destructive testing. To reconstruct data from observed measurements, we derive non-convex but well-conditioned LASSO-type functional. By exploiting convex-concave geometry functional, design projected gradient descent algorithm show its effectiveness extensive numerical simulations, both on toy real data.