Massive MIMO (Multiple-Input Multiple-Output) has the advantage of providing excellent multiplexing/diversity gain. However, the large antenna array will cause high hardware cost. Hybrid precoding is an effective method to reduce the RF (Radio Frequency) chains. In this paper, we formulate a MMSE-based (Minimum Mean Square Error-based) optimization model, propose a HP-MMSE (Hybrid Precoding-Minimum Mean Square Error) algorithm and obtain the closed-form solutions of hybrid precoders. Simulation results show that the proposed algorithm can approach the optimal unconstrained BER (Bit Error Rate) performance with higher energy efficiency.