Abstract
In this paper, a joint multi-user constellation is proposed for energy detection-based non-coherent massive multiple-input multiple-output system. This is motivated by the simple design and high energy efficiency it entails for both the transmitter and receiver. First, the orthogonal codes is employed to suppress the multi-user interference. However, this comes at the price of consuming more communications resources. In this study, the key to reduce code redundancy is the design of a joint constellation since it makes energy detection applicable when multiple users employ the same orthogonal codes. Although it is unsolvable initially, our analysis indicates that through minimizing the symbol-error rate (SER), the joint constellation design becomes feasible. Concretely, two analytical expressions of SER based on Gamma and Gaussian distributions
are derived. Via minimizing the error probability, an important result that the joint constellation should satisfy is obtained. Accordingly, an isometric constellation design is proposed to find constellations that enable non-coherent reception with multiple users, and achieve the minimum SER simultaneously. In addition, decoding regions of symbol decision are optimized to further improve the error performance. In the end, numerical simulations are carried out to highlight the effectiveness of our proposed scheme.
are derived. Via minimizing the error probability, an important result that the joint constellation should satisfy is obtained. Accordingly, an isometric constellation design is proposed to find constellations that enable non-coherent reception with multiple users, and achieve the minimum SER simultaneously. In addition, decoding regions of symbol decision are optimized to further improve the error performance. In the end, numerical simulations are carried out to highlight the effectiveness of our proposed scheme.
Original language | English |
---|---|
Pages (from-to) | 3812 - 3825 |
Number of pages | 14 |
Journal | IEEE Transactions on Wireless Communications |
Volume | 19 |
Issue number | 6 |
DOIs | |
Publication status | Published - Jun 2020 |