A New Conjugate Gradient Projection Method for Convex Constrained Nonlinear Equations

Figure 4

From top to bottom: the original signal, the measurement, and the reconstructed signal by two algorithms. (a) Original (n = 4096, number of nonzeros = 16), (b) measurement, (c) LJJ CGPM (MSE = 3.81e − 07, Iter = 42, time = 0.66 s), and (d) PDY (MSE = 3.21e − 07, Iter = 61, time = 1.20 s).

(a)

(b)

(c)

(d)