This paper studies advanced multi-access techniques to support a high volume of concurrent access demanded in future wireless networks. Sparse code multiple access (SCMA), as a code-domain Non-Orthogonal Multiple Access (NOMA), serves multiple users simultaneously by frequency-domain coding. Blind Interference Alignment on the other hand applies time-domain coding to accommodate multiple users. Unlike beamforming, both of them need no CSIT, which saves the control cost of channel information feedback and highly simplifies the implementation of multiple access. To achieve even higher multiplexing gain and diversity order, we propose a new multiple access framework, which makes use of both time and frequency coding by combining SCMA and BIA, and is named as SparseCode-and-BIA-based multiple access (CIAMA). Then we analyze two decoding schemes, the two-stage decoding scheme consisting of zero-forcing and Message Passing Algorithm (MPA), and the Joint Message Passing Algorithm (JMPA) enhanced by constructing a virtual factor graph. Simulation results indicate that although the two-stage decoding performance is inferior to both BIA and SCMA, it has low decoding complexity. On the other hand, the JMPA decoding scheme achieves the same diversity gain as STBC-based SCMA while gaining even higher multiplexing gain, which makes the CIAMA with JMPA decoding scheme an attractive MA scheme in future wireless communication networks.

翻译：本文研究先进的多种接入技术，以支持未来无线网络中所需的大规模并发接入。稀疏码分多址作为一种码域非正交多址，通过频域编码同时为多个用户提供服务；而盲干扰对齐则采用时域编码以容纳多个用户。与波束赋形不同，这两种技术均无需信道状态信息，从而节省了信道信息反馈的控制开销，并极大简化了多址接入的实现。为进一步提升复用增益和分集阶数，本文提出一种新的多址接入框架，通过结合SCMA与BIA同时利用时域和频域编码，命名为基于稀疏码和BIA的多址接入。随后分析了两种解码方案：由迫零和消息传递算法组成的两阶段解码方案，以及通过构建虚拟因子图增强的联合消息传递算法。仿真结果表明，尽管两阶段解码性能劣于BIA和SCMA，但其解码复杂度低；而JMPA解码方案在获得与基于空时分组码的SCMA相同的分集增益的同时，实现了更高的复用增益，这使得采用JMPA解码的CIAMA成为未来无线通信网络中极具吸引力的多址接入方案。