Tight Performance Bounds for Compressed Sensing With Conventional and Group Sparsity
This work addresses compressed sensing for group sparsity, offering improved theoretical guarantees that are incremental but with broader applicability.
The paper tackles the problem of recovering group sparse vectors from linear measurements by showing that ℓ₁-norm minimization can be used for group sparse recovery, introducing the group robust null space property (GRNSP) and linking it to a group version of the restricted isometry property (GRIP). It provides less conservative bounds for equal-sized groups, extends to varying group sizes, and offers new error bounds for ℓ_p-norms and measurement requirements.
In this paper, we study the problem of recovering a group sparse vector from a small number of linear measurements. In the past the common approach has been to use various "group sparsity-inducing" norms such as the Group LASSO norm for this purpose. By using the theory of convex relaxations, we show that it is also possible to use $\ell_1$-norm minimization for group sparse recovery. We introduce a new concept called group robust null space property (GRNSP), and show that, under suitable conditions, a group version of the restricted isometry property (GRIP) implies the GRNSP, and thus leads to group sparse recovery. When all groups are of equal size, our bounds are less conservative than known bounds. Moreover, our results apply even to situations where where the groups have different sizes. When specialized to conventional sparsity, our bounds reduce to one of the well-known "best possible" conditions for sparse recovery. This relationship between GRNSP and GRIP is new even for conventional sparsity, and substantially streamlines the proofs of some known results. Using this relationship, we derive bounds on the $\ell_p$-norm of the residual error vector for all $p \in [1,2]$, and not just when $p = 2$. When the measurement matrix consists of random samples of a sub-Gaussian random variable, we present bounds on the number of measurements, which are less conservative than currently known bounds.