ID 原文 译文
47026 迭代块判决反馈均衡器（IBDFE, iterative block de-cision feedback equalizer）能以较低的复杂度达到与最佳接收相近的性能，有效降低了 FTN 接收机的复杂度，但其复杂度仍能进一步降低。 An effective low-complexity FTN receiver is iterative block decision feedback equalizer (IBDFE), whichcan achieve performance close to the optimal receiver with much lower complexity, but its complexity can be further reduced.
47027 因此，在 IBDFE 的基础上，简化了前馈和反馈滤波器的设计，为 FTN 通信系统提出了一种降低复杂度的 IBDFE（RC-IBDFE, reduced-complexity IBDFE）。 Therefore, a reduced-complexity IBDFE (RC-IBDFE) was proposed for FTN communication systems, which was based onIBDFE, but introduced a simplified method to design the feedforward and feedback filters.
47028 当加速因子为 τ={0.8, 0.7}时，数值结果和分析表明，与 IBDFE 相比，RC-IBDFE 的顽健性更好，复杂度降低了{45.7%, 25%}，当误比特率为 10 −5 时的性能增益为{0.605 4, 0.193 6} dB； When packing factor is τ={0.8, 0.7},numerical results and analysis indicate that, compared with IBDFE, RC-IBDFE has stronger robustness, its complexity is re-duced by {45.7%, 25%}, and its performance gain is {0.605 4, 0.193 6} dB when the bit error rate (BER) is 10 −5.
47029 与已有的低复杂度 IBDFE（LC-IBDFE, low-complexity IBDFE）相比，RC-IBDFE 的顽健性更好，复杂度降低了{13.6%, −25.8%}，当误比特率为 10 −5 时的性能增益为{0.289 7, 0.531 2} dB。 Comparedwith the existing low-complexity IBDFE (LC-IBDFE), RC-IBDFE has stronger robustness, its complexity is reduced by {13.6%,−25.8%}, and its performance gain is {0.289 7, 0.531 2} dB when the BER is 10 −5.
47030 针对非对称 PCMA 信号盲分离问题，在一种低复杂度的盲分离结构基础上，提出基于软信息联合修正提高强弱信号解调性能的盲分离算法。 On the basis of a blind separation structure with low complexity, an efficient blind separation algorithm basedon soft information joint correction was proposed for asymmetric PCMA to improve the demodulation performance ofstrong and weak signals.
47031 该算法利用强弱信号解调相互的影响，通过尝试修正高出错概率的强信号接收符号， By utilizing the demodulation mutual influence between the strong and weak signals, this algo-rithm tried to correct the receiving symbols of strong signal with high error probability.
47032 比较修正前后强弱信号符号星座质量（软信息）的优劣来决定是否修正当前强弱信号的硬判决值，从而降低强弱信号解调误码率。 Comparing the symbol constella-tion quality (soft information) of signals before and after correction, it can be decided whether the hard decision values ofstrong and weak signals need to be modified, which efficiently reduce the demodulation error rate of strong and weaksignals.
47033 仿真结果表明，当强信号信噪比高于 17 dB 时，强弱信号解调误码率分别降低近 2 个数量级，且算法计算复杂度低于传统基于重构抵消算法的计算复杂度。 The simulation results show that, the demodulation error rate of strong and weak signals can be reduced by nearlytwo orders of magnitude after joint correction especially when the signal to noise ratio of strong signal is higher than17 dB, and the computational complexity of this algorithm is lower than that of the conventional reconstruction cancella-tion algorithm.
47034 针对强化学习中已有连续动作空间算法未能充分考虑最优动作的选取方法和利用动作空间的知识，提出一种对自然梯度进行改进的行动者评论家算法。 The existed algorithms for continuous action space failed to consider the way of selecting optimal action andutilizing the knowledge of the action space, so an efficient actor-critic algorithm was proposed by improving the naturalgradient.
47035 该算法采用最大化期望回报作为目标函数， The objective of the proposed algorithm was to maximize the expected return.