Abstract: This paper considers a class of mixed-constrained least squares problems arising from matrix equations and proposes an inertial Peaceman-Rachford splitting method for their solution. The original problem is first reformulated as an equivalent separable convex optimization problem. Exploiting the separability of the Peaceman-Rachford framework, the resulting subproblems are efficiently solved with the aid of the LSQR method. Under suitable assumptions, the global convergence of the proposed algorithm is established. Numerical experiments further validate the effectiveness and computational efficiency of the proposed method.