Research and Simulation of Beamforming Algorithms in the TD-LTE System

In order to reduce multi-access interference and increase system throughput, beamforming algorithms are introduced into the LTE system. These three beamforming algorithms based on LMS, eigenvalue decomposition and conjugate normalization are studied in this paper, making simulation and comparison of the performances of these three algorithms on the simulation platform of LTE system. The results show that in the static environment, the performances of beamforming algorithms based on the eigenvalue decomposition and conjugate normalization are equivalent, the performances of them are better than LMS algorithm´s. And due to small computation, the conjugate normalization algorithm is consequently more practical. Meanwhile, the influences of the parameters of the granularity of the weights, different antenna configuration, ETU channel model and UE movement speed on the performances of beamforming algorithms are also simulated and studied. Results show that under the same conditions, the smaller the granularity of the weights is, the better the performance is; the better the correlation between antennas is, the better the performance is; the smaller the UE movement speed is, the better the performance is.