Restricted p-isometry properties of nonconvex block-sparse compressed sensing

• We establish a p-RIP condition for robust recovery of nearly block-sparse signals via mixed l2/lp-minimization.
• We obtain a stronger version p-RIP condition for exact recovery of block-sparse signals via mixed l2/lp-minimization.
• We determine how many random, Gaussian measurements are sufficient for the p-RIP condition to be satisfied with high probability.
• A series of numerical experiments have been carried out to prove the validation of our presented theoretical results.

In this paper, by generalizing the notion of restricted p  -isometry constant (0