Sum‐rate optimization problem is a key issue in a time‐varying multi‐user multi‐input multi‐output (MU‐MIMO) distributed antenna system. Channel precoding matrix is the key technology in the achievable sum‐rate optimization problem. In this paper, two algorithms are proposed to solve the achievable sum‐rate optimization problem for a time‐varying MU‐MIMO distributed antenna system, namely one‐dimensional search algorithm (OSA) and cyclic MMSE (Minimum Mean Square Error) search algorithm (CMSA). In order to design the channel precoding matrix for the achievable sum‐rate optimization, a one‐dimensional search algorithm is proposed in the time‐varying MU‐MIMO distributed antenna system. For the problem with a large amount of computation in OSA, a cyclic MMSE search algorithm is proposed in the time‐varying MU‐MIMO distributed antenna system. Simulation results show that the proposed algorithms can effectively improve the sum‐rate compared to other algorithms. Simulation results also show that OSA is superior to CMSA in the sum‐rate, and the channel state information (CSI) coefficient has little effect on OSA and CMSA.