Takashi Seyama

Efficient selection of user sets for downlink non-orthogonal multiple access

Non-orthogonal multiple access (NOMA), which uses superposition coding at the transmitter side and successive interference cancellation at the receiver side, is a promising new downlink multiple access scheme. The product of user average throughputs among all users within a cell can be maximized by optimal proportional fair (PF) scheduling, but this is computationally complex because it requires an exhaustive search for all possible user sets and optimal power allocation among users. In this work, we propose a novel method of selecting user sets for PF scheduling in downlink NOMA. Unlike conventional suboptimal methods, the proposed method is based on the mathematical characteristics of the PF metric. The proposed method can reduce the complexity when the maximum number of multiplexed users is three and four as well as two by judging whether a user set is worth multiplexing or not on the basis of a simple condition between the weakest two users within the set. Computer simulation showed that the proposed method can significantly reduce the average number of searched user sets while achieving almost the same total and cell-edge throughput as the optimal method.

A Low Complexity PMI/RI Selection Scheme in LTE-A Systems

Recent wireless radio technologies have utilized precoding matrix and rank adaptation as a technique to increase frequency efficiency. In 3rd generation partnership project (3GPP) long term evolution (LTE) and LTE-advanced (LTE-A) systems, codebook based precoding is utilized. In the codebook based system, a mobile station (MS) needs to estimate optimum channel state information (CSI) which includes a channel quality indicator (CQI), a precoding matrix indicator (PMI) and rank indicator (RI) and to feed back the CSI to the base station (BS) in order to get higher frequency efficiency. Generally, the MS estimates the optimum precoding matrix from all possible precoding matrices which are predefined as the codebook. When such an exhaustive search algorithm is utilized in LTE-A MS, the MS needs huge computation power because the number of rank and precoding matrix patterns is large in a LTE-A system. In this paper, we propose a novel PMI/RI selection algorithm which is less complex than the exhaustive search algorithm in a LTE-A system. After evaluating the throughput performances by simulation and analyzing the computations, it is described that the proposed algorithm has the same throughput performances as the conventional algorithm and 13% of the computations compared with the conventional algorithm.

MLD-Based MU-MIMO Detection Scheme for LTE Downlink

Currently Multi-User MIMO (MU-MIMO) attracts attention to improve frequency usage efficiency. When using Maximum Likelihood Detection (MLD) for MU-MIMO detection, we need to know the modulation type of the transmitted signals. However in 3rd Generation Partnership Project (3GPP) Long Term Evolution (LTE) system, the user to whom MU-MIMO is applied cannot be informed about the modulation type of the signals for a multiplexed interference user. Therefore, linear filters such as Zero Forcing (ZF) and Minimum Mean Square Error (MMSE) are typically applied. In this paper, we propose an MLD-based MU-MIMO detection method with modulation-type estimation. In this method, we perform MLD by assuming possible modulation types to calculate metrics, and by comparing them, we determine the modulation type of the signals for an interference user. From the computer simulation results, we find that the proposed method can obtain better performance than MMSE with a slight increase in complexity from SU-MIMO detection.