| ID | 原文 | 译文 |
| 3903 | 理论分析和蒙特卡罗仿真验证了 ASDST 系统的性能。 | The performance of ASDST system was verified by theoretical analysis and Monte Carlo simulation. |
| 3904 | 对人工智能在移动通信领域学术界和产业界的研究与应用现状进行了总结，指出了人工智能在提升移动通信系统性能方面的挑战和瓶颈。 | The research and application progress in mobile communication, and points out its obstacle for improving theperformance of mobile communication system were summarized. |
| 3905 | 创造性地提出性能外环与性能内环协同减小实际网络性能与理想网络性能间距离的新思路和新方法：对性能外环部分进行人工智能重构，对性能内环部分进行传统自适应或最优化，形成与性能外环部分的最佳协同。 | A new approach for reducing the gap between the prac-tical and the ideal network performance was put forward creatively, which included the artificial intelligence reconstruc-tion for performance outer loop part, the traditional adaptation or optimization for performance inner loop part, and theoptimal cooperation between the two parts. |
| 3906 | 若干成功应用的实例证明了该思路和方法的有效性。 | The effectiveness of this approach was proved by the several successful ap-plications. |
| 3907 | 最后指出，为了满足移动通信系统对人工智能解决方案“稳”“准”“快”的严苛需求，使能移动网络的自动化、智能化、智慧化，除了使人工智能重构的方案本身具有优异的性能外，还必须有基于大数据分析和模拟系统的反馈闭环系统架构，而架构中模拟器的构建——模拟系统是实现“稳”“准”“快”严苛需求的关键路径。 | Finally, it pointed out that in order to meet the severe demands of mobile communication system for the “sta-ble” “accurate” and “fast” artificial intelligence solution, and make the mobile network automatic, intelligent and wise, inaddition to the excellent performance of the artificial intelligence reconstruction scheme itself, there must also be a feed-back closed-loop system architecture based on the big data analysis and analog system, which is the key path to achievethis target. |
| 3908 | 研究了分组马尔可夫叠加传输的神经网络（NN）译码方案。利用 NN，实现了不同网络结构、数据表征形式的基本码译码器。在此基础上，将所实现的基本码译码器嵌入迭代译码机制中，提出了基于 NN 的分组马尔可夫叠加传输的滑窗译码算法， | A neural network (NN)-based decoding algorithm of block Markov superposition transmission (BMST) was researched. The decoders of the basic code with different network structures and representations of training data were im-plemented using NN. Integrating the NN-based decoder of the basic code in an iterative manner, a sliding window de-coding algorithm was presented. |
| 3909 | 并分析了其对应的性能下界。所提出的译码算法提供了一种将 NN 运用到长码译码的解决思路，即用 NN 替代译码中的部分模块。 | To analyze the bit error rate (BER) performance, the genie-aided (GA) lower boundswere presented. The NN-based decoding algorithm of the BMST provides a possible way to apply NN to decode longcodes. That means the part of the conventional decoder could be replaced by the NN. |
| 3910 | 仿真结果表明，利用 NN 实现的基本码译码器可以达到最大似然译码性能。基于 NN 的分组马尔可夫叠加传输的滑窗译码算法性能在中高信噪比区域与对应精灵辅助下界贴合，获得了额外的编码增益。 | Numerical results show that theNN-based decoder of basic code can achieve the BER performance of the maximum likelihood (ML) decoder. For theBMST codes, BER performance of the NN-based decoding algorithm matches well with the GA lower bound and exhib-its an extra coding gain. |
| 3911 | 考虑一个基于 k 阶串行干扰消除的功率域非正交接入的单跳网络，研究在定义功率域非正交多路接入的可靠性模型的基础上，在传输时延有界的前提下，研究如何通过用户组配和功率分配来最大化上行传输的平均可靠性，进而为其提出了一个复杂度为 O(nlogn)的启发式算法，其中 n 是用户数。 | A single-hop PD-NOMA network supporting k-successive interference cancellation (k-SIC) was considered.Given the real-time performance requirements, how to maximize the average reliability of uplink transmissions by the joint users grouping and power allocation was studied. A heuristic algorithm with complexity of O(nlogn) was presented,where n was the number of users. |
| 3912 | 此外，还证明了该算法在 k=2 的情况下是最优的。 | The algorithm was also proved to be the optimal if k=2. |