Downlink Channel Estimation For Massive MIMO And Large Intelligent Metasurfaces Assisted Communication System

Efficient use of the limited quantity of available spectrum to cater to the exponentially increasing demand for throughput has been the focus of communication and signal process-ing engineers for the last decades.With the advent of technologies such as the Internet of things(IoT)or machine-type communications,devices and appliances around us which have predominantly been offline are being equipped with sensors that generate data and are now driving the demand for throughput.Many new technologies are under investigation to cater to these use cases and to also increase throughput of the existing wireless environment.One of the enabling candidate technology is massive MIMO.Massive MIMO have a huge potential to provide a significant improvement in spectral and Energy efficiency.Deploying Large Intelligent Metasurface(LIM)on building walls,LIM-assisted Massive MIMO,to avoid link blockage and perform reflect beamforming in Massive MIMO network is another hopeful candidate technology that is gaining too much attention in recent times.LIM provide the possibility of directed forwarding of the incoming signal without employ-ing any power amplifier and signal processing units,but rather by suitably designing the phase shifts applied by each reflecting element,in order to constructively combine each re-flected signal.However,fully harvesting the potentials of massive MIMO,with or without LIM,requires an accurate channel estimation at the transmitter.In the conventional channel estimation scheme the overhead associated with channel estimation is directly proportional to the number of transmit antennas which result in too much overhead for DL channel esti-mation of massive MIMO networks.In TDD Massive MIMO networks this can be avoided by exploiting channel reciprocity.However,for FDD Massive MIMO,a duplex mode which is employed in most of existing cellular systems,DL channel estimation is deemed to be im-possible due to the high overhead even for static channel.Furthermore,the available channel estimation schemes for LIM-assisted MIMO need the LIM to be equipped with additional signal processing capability which result in higher cost and increased power consumption.The first contribution of this thesis is a low-overhead characteristic learning,tracking,and monitoring mechanism for the time-varying massive MIMO channel of FDD system.Specifically,this work exploited the common spatial sparsity and temporal correlation of the channels to develop the low-overhead estimation scheme.Firstly,using the virtual chan-nel representation and modeling the temporal correlation as an autoregressive process,the time-varying massive MIMO channel is formulated as a sparse signal model.Then,a sparse Bayesian learning(SBL)scheme based on the expectation maximization(EM)is proposed to determine the model parameters of the channel.To achieve the posteriors of different model parameters,the approximate message passing(AMP)is used in the expectation step.Furthermore,the Kalman filtering(KF)with a reduced dimension is used to track the down-link(DL)channel.To observe the change of model parameters and start the relearning process,a monitoring scheme based on the Bayesian Cramer-Rao bound(BCRB)is also developed.Finally,numerical results are provided to demonstrate the performance of the proposed scheme.The second contribution of this thesis is a scheme to estimate the cascaded channel in LIM-assisted Massive MIMO,the transmitter-LIM channel and LIM-receiver channel,un-der a scenario where the transmitter is equipped with massive MIMO and the LIM have no signal processing capability.By exploiting the sparsity of the channel and programmability of LIM,the transmitter-LIM and LIM-receiver cascaded channel estimation is formulated as a combined compressive sensing and self-calibration problems.We proposed two stage algorithm that utilize vector approximate message(VAMP)passing for the compressive sens-ing and the adaptive VAMP for the self-calibration.Numerical results are also presented to demonstrate the performance of the proposed scheme.