This chapter focuses on a particular class of multiple measurement vector problems, where the solutions are row-sparse or joint-sparse. It presents the algorithms to solve the unconstrained problems. The constrained and unconstrained forms are equivalent for proper choice of ε and α. However, there is no analytical relationship between them in general. The cooling technique solves the constrained problem in two loops. The outer loop decreases the value of α. The inner loop solves the unconstrained problem for a specific value α. Just as there are greedy methods to solve the sparse and group-sparse recovery problems, there are techniques to solve the row-sparse recovery problem as well. It is a simple extension of the sparse recovery for single measurement vector (SMV). The chapter also presents the algorithm for orthogonal matching pursuit, used for solving the SMV problem.