Massive MIMO Transmission Scheme Based On Per Beam Synchronization

Nowadays,mass access from wireless terminals makes data flow increase exponentially,which requires breakthroughs in the future wireless communication technologies.Massive multiple-input multiple-output(MIMO),which can improve system throughput,is considered as a key technology for the fifth generation wireless systems.Millimeter wave(mmWave)also attracts much attention thanks to its large spectral band-widths and the feasibility to combine with massive MIMO.However,one concern in applying mmWave to wireless communication systems is the severity of doppler effect due to high frequencies.Due to the high spectral efficiency of orthogonal frequency division multiplexing(OFDM),OFDM combined with massive MIMO is a promising technology for wideband transmission with high data rate.Nevertheless,if massive MIMO systems adopt orthogonal pilot approaches as conventional MIMO systems,the pilot overhead issue would become the system bottleneck.To address above problems,we investigate massive MIMO transmis-sion scheme based on per beam synchronization.First of all,we introduce a beam domain channel model for massive MIMO and further study its char-acteristics.We introduce a physically spatial channel model and investigate the relationship between space domain and beam domain when the number of antennas is large,which paves the way for the following beam domain transmission scheme.Also,we demonstrate the channel sparsity in the beam domain,which is the basis of the channel estimation method proposed later.Furthermore,we obtain the approximate representa-tion of beam domain channel model and demonstrate the banding characteristics of doppler spread and delay spread.To be more specific,when compared with doppler spread and delay spread in the space domain,doppler spread and delay spread in the beam domain will be largely reduced,and doppler spread will be re-duced by a factor of the number of antennas at the mobile station.This banding characteristic is the theoretical basis of the proposed massive MIMO transmission scheme based on per beam synchronization.Secondly,based on above characteristics,we propose a transmission scheme based on per beam syn-chronization for massive MIMO systems under mobile scenarios.Considering the multi-beam diversity of user terminals,we develop a greedy user and beam scheduling algorithm based on the beam domain statistical channel state information.Based on the elementary synchronization techniques,we propose to synchronize in the beam domain.Simulations results show that,with this method,the estimation accuracy for synchro-nization parameters can be largely improved.Given the fact that doppler spread and delay spread are largely reduced in the beam domain,we propose a uplink and downlink joint transmission scheme based on per beam synchronization,which can reduce channel's dispersion both in the frequency domain and time domain.The-oretical deduction and simulation results both show that the effective channels after synchronization exhibit the slowing down effects both in the frequency domain and time domain.As a result,the coherence time and coherence bandwidth will increase a lot,which can bring many performance gains.Through many ex-periments,we obtain a conclusion that the proposed transmission scheme based on per beam synchronization outperforms the conventional antenna-based transmission scheme,especially when the velocity of mobile stations is high or the number of antennas at the mobile station is large.Based on per beam synchronization,we investigate the pilot design and channel estimation issue for massive MIMO-OFDM uplink transmission,which aims to reduce the heavy pilot overhead.By exploiting the channel sparsity in the beam-delay domain,the channel estimation issue for multiuser uplink transmis-sion is formed into a sparse compressed sensing(CS)model.We propose a cyclic shift pilot scheme,which satisfies Restricted Isometry Property under the CS frameworks.As the sparsity is efficiently utilized,the proposed scheme significantly reduces the pilot overhead.Furthermore,based on approximate message pass-ing(AMP)algorithm,an approximate Bayesian inference can be conducted for channel estimation with low complexity.Simulation results show that the proposed AMP-based algorithm with the pilot scheme outper-forms several existing baselines,especially in the low signal-to-noise ratio(SNR)regime.Moreover,when compared with conventional microwave massive MIMO systems,mm Wave massive MIMO systems exhibit better performance.