There have been many results for the sparse signal recovery in Compressed sensing, such as a series of greedy methods and l1, minimization. Under proper conditions, we can solve the problem of l1, minimization or l1,-regularized least squares to recovery sparse signals. This paper provides several methods based on convex optimization, Interior point, gradient projection and fixed point methods are used to solve these problems even when the observations are contaminated with noise(y=Φχ+e). We also present numerical results of these methods. |