Information Theoretical Studies on MIMO Channel with Limited Channel State Informatio

Tremendous increase in throughput, reliability and security requirements in present and future wireless communication networks necessitates the migration towards the underutilized higher frequency bands. The premise of large scale multiple input multiple output (MIMO) technology deployment in these bands has the potential of fulfilling future network requirements. At the same time, large scale network deployment, or the so-called dense coverage (large number of small scale base stations), is another link level strategy that also has the potential of enhancing the overall network quality of service (QoS). Performance of MIMO communication systems is governed by the amount of channel state information (CSI) available at both transmitter and receiver especially when deployed in a dense coverage network which has the potential of high line of sight (LoS) opportunity.;This thesis aims to address throughput, reliability and physical layer security aspects of MIMO communication systems deployed in a fading environment with a stable path between transmitter and receiver with limited CSI feedback. The research involves four major research directions: (1) Transmitter optimization for public messages with minimal form of CSI feedback, (2) Secrecy capacity and optimal transmission strategy for confidential messages under the same limited CSI feedback model with eavesdropper uncertainty, (3) Establishing fundamental limits of covert communication of MIMO AWGN channel and highlight the potential of having a dominant channel mode in establishing high covert rates, (4) Message source authentication over MIMO channel with dominant mode.;We start by considering the MIMO channel with dominant LoS component where the only CSI available at the transmitter are the Rician factor and the physical direction of the receiver with respect to the transmitter antennas array. For this particular scenario, although the exact capacity still unknown in a closed form, we establish an upper bound using Jensen's inequality for which we derive the maximizing transmission strategy in a closed form. Despite of being suboptimal, the upper bound maximizing strategy, when compared to all previously proposed strategies, is shown to provide a substantial gain in terms of gap to capacity over a wide range of system parameters.;We extend our research with the previous setup to the scenario in which the exchanged message are subject to secrecy constraint. We establish the delay limited secrecy capacity of the channel with eavesdropper uncertainty. We introduce a novel class of eavesdropper CSI uncertainty that makes use of the location diversity of communicating nodes. Further, we obtain upper and lower bounds on the ergodic secrecy capacity when the exchanged message are not subject to delay constraint.;Different from low probability of intercept constraint, low probability of detection (LPD) is a more restrictive communication scenario in which the communication is required to be undetectable by a passive adversary. We extend our research to establish the fundamental limits of covert (LPD) communication over MIMO channel. We derive the exact scaling laws of the number of covert bits that can transmitted over MIMO channel in the two asymptotic limits of MIMO channel, large block length and large array limits.;A key advantage of MIMO channel with dominant mode is the geolocation awareness, i.e., transmitter and receiver have the ability to identify the physical direction of each others. In this correspondence, we study message authentication problem by leveraging the cooperation between existing cryptographic authentication schemes and physical layer based wireless authentication. In particular, we introduce the angle of arrival (AoA) based wireless authentication in which the actual AoA of a given message is checked for consistency with the information available at the receiver about the communication channel. Hardware implantation details and field experimental results are provided to offset the gap between theoretical studies of physical layer security and their application to practical communication systems.