Low-complexity precoding for multi-beam satellite systems

Conventional lattice reduction aided Tomlinson-Harashima precoding (LRA-THP) is acknowledged complexity. A low-complexity precoding algorithm was proposed for multi-beam satellite systems. Users are grouped based on the distance between them in the proposed algorithm. Then, the matrix inversion and QR decomposition are used to calculate the preprocessing matrix by which the inter-group interference is suppressed. During calculation, the characteristic of multi-beam satellite systems is exploited to reduce the dimension of the preprocessing matrix and the computational complexity can be decreased further by this step. The LRA-THP is employed in each group to eliminate the residual interference. Simulations show that the proposed algorithm can reduce the complexity over the conventional LRA-THP by 92.5% while the performance loss is about 0.2 dB. In addition, compared with the existing group precoding algorithm, the proposed algorithm can achieve the similar performance and has a 69.7% reduction in the computational complexity. The low complexity and flexibility of the proposed algorithm is of higher practical value. ©, 2015, Beijing University of Posts and Telecommunications. All right reserved.