서강대학교

초록/요약

Multiple-input multiple-output (MIMO) technology has attracted a great deal of attention due to its capabilities to offer a significant increase in data throughput and link reliability. MIMO systems equipped with distributed transmit antennas can provide enhanced data throughput as well as improved signal quality when appropriate cooperative transmission schemes among antennas are employed. In this dissertation, we first investigate the performance of single-user distributed MIMO (D-MIMO) systems under the influence of Rayleigh fading and log-normal shadowing for different cooperation strategies, and corresponding statistical distributions for both the signal-to-interference-plus-noise ratio (SINR) and the capacity are presented. Specifically, the ergodic capacity per remote antenna (RA) is determined when n RAs cooperatively transmit the desired signal to the target receiver, which is compared with the non-cooperative transmission (NCT) case. Using the results, a preferable mode of operation for given geographic locations of the receiver is determined. We propose to perform an adaptive operation between the cooperative and non-cooperative modes under the capacity maximization criteria, and quantify exact amounts of capacity gain over individual transmission modes. When RAs for signal transmission are selected based on the instantaneous channel condition, we can obtain additional capacity gains over aforementioned transmission schemes. By deriving a general expression for the SINR of such antenna selection based transmission, we determine an exact amount of the maximum achievable gain in capacity. For important special cases of selecting a single RA for non-cooperative transmission and selecting two RAs for cooperative transmission among three RAs surrounding the receiver, closed-form formulas are presented for the SINR and capacity distributions. For multi-user D-MIMO systems, we present a collaborative MIMO precoder and power control (PC) strategy, which is optimal in terms of obtaining the maximum sum-rate under a given power constraint for each antenna. Closed-form expressions for both the MIMO precoder and the corresponding sum capacity are derived for the two-user case. Performance gains of the optimal precoders are quantified over existing schemes including zero-forcing transmission with and without dirty-paper coding (DPC). Numerical evaluation is performed using sectorized cellular structures to demonstrate the effectiveness of the optimal scheme under practical settings. Optimal linear MIMO precoding schemes without using DPC are also investigated in terms of the sum-rate maximization and the log-sum-rate maximization for fairness improvement, applicable to any number of RAs and mobile users. We further propose a simple strategy to determine transmission power for each user, referred to as channel norm-based power control to obtain near-optimal performance of the presented optimal precoders. In particular, a closed-form solution to maximize the sum-rate using the PC is suggested. It is also shown that the proposed strategy can be efficiently utilized in conjunction with existing multi-user transmission schemes.