Channel Estimation for Sparse mm-Wave MIMO System

The fifth-generation (5G) cellular networks will provide gigabit-per-second data rates from massive antenna array combined with the emerging use of large and unexploited millimeter wave (mm-Wave) bands (30–300 GHz) in small cells. Channel estimation for sparse mm-Wave MIMO systems is a difficult task. This is because of a large number of coefficients to be estimated, lower scattering nature, and blockage of mm-Wave by many materials in the environment. This paper will be the opportunity to implement the sparse channel estimation in the 5G cellular networks. In this work, we propose compressed-sensing (CS) based solutions and implements hybrid MIMO architecture for the proposed algorithm, OMP algorithm, and oracle estimator with different mm-Wave MIMO setups. Simulation results show that as compared to the OMP algorithm, proposed algorithm requires 16.9 times less computation time, and significant improvement is seen in normalized mean squared error (NMSE). Also, in the analysis, we found that the performance of the hybrid MIMO approaches near-optimal to conventional fully digital precoder.