A Relaying Scheme Based on Diagonalization for Multi-Relay Symmetric MIMO Communication Networks

This letter investigates an effective relaying scheme for the existing multi-relay symmetric multiple-input-multiple-output (MIMO) communication networks, with a source-destination pair that communicates through amplify-and-forward (AF) multi-relay nodes. The maximum mutual information (MI) is taken as the design criterion with per relay power constraint. Considering the objective function can be maximized by diagonalizing the matrix of the determinant, a low-complexity and decentralized design is developed to solve this problem. Based on mathematics theory and variable substitution, the objective function is constituted with the sum of multiple diagonal matrices by eigenvalue decomposition (ED) and singular value decomposition (SVD), the original optimization problem is transformed into a standard scalar convex optimization problem. Finally, the Karush-Kuhn-Tucker (K.K.T) conditions are employed to obtain the optimal values of the diagonal elements for all power allocation matrices. Numerical results show that the proposed scheme can significantly improve the system performance.