This thesis considers channel estimation and multiuser (MU) data transmission for massive MIMO systems with fully digital/hybrid structures in mmWave channels. It contains three main contributions. In this thesis, we first propose a tone-based linear search algorithm to facilitate the estimation of angle-of-arrivals of the strongest components as well as scattering components of the users at the base station (BS) with fully digital structure. Our results show that the proposed maximum-ratio transmission (MRT) based on the strongest components can achieve a higher data rate than that of the conventional MRT, under the same mean squared errors (MSE). Second, we develop a low-complexity channel estimation and beamformer/precoder design scheme for hybrid mmWave systems. In addition, the proposed scheme applies to both non-sparse and sparse mmWave channel environments. We then leverage the proposed scheme to investigate the downlink achievable rate performance. The results show that the proposed scheme obtains a considerable achievable rate of fully digital systems. Taking into account the effect of various types of errors, we investigate the achievable rate performance degradation of the considered scheme. Third, we extend our proposed scheme to a multi-cell MU mmWave MIMO network. We derive the closed-form approximation of the normalized MSE of channel estimation under pilot contamination over Rician fading channels. Furthermore, we derive a tight closed-form approximation and the scaling law of the average achievable rate. Our results unveil that channel estimation errors, the intra-cell interference, and the inter-cell interference caused by pilot contamination over Rician fading channels can be efficiently mitigated by simply increasing the number of antennas equipped at the desired BS.