Joint Channel Estimation Based on the Structured Sparsity for FDD Massive MIMO Systems

The variability of wireless environment makes channel estimation increasingly important for 5G communications. However, the pilot overhead is high in massive multiple-input multiple-output systems. Thus, a joint channel estimation scheme based on the structured sparsity of the channel is proposed to reduce the pilot overhead. Specifically, the structured sparsity of the channel is revealed firstly, i.e., adjacent slots for a user enjoy the partially common non-zero rows in a channel matrix. Then, multiple time slots can be estimated jointly by exploiting this property. Simulation results show that the proposed algorithm can achieve accurate channel estimation with a low pilot overhead.