ID 原文 译文
47176 针对现有不确定数据聚类算法中构建不确定数据模型以及距离度量时存在的影响结果准确性与聚类性能等问题，提出一种基于快速高斯变换的不确定数据聚类算法。 An uncertaindata clustering algorithm based on fast Gaussian transform was proposed, to solve the problems about the impact on theaccuracy of clustering results and the clustering efficiency caused by the uncertainties, during the construction of uncer-tain data models and the distance measurement, which existed in the current researches.
47177 首先在不假设数据分布的前提下，构建符合不确定性分布特征的数据模型； First, the data model according tothe characteristic of the uncertainty distribution was constructed, without the premise of assuming the data distribution.
47178 然后结合不确定对象的 2 个重要特征：属性特征与表示不确定数据分布特征的概率密度函数，度量不确定数据对象间的相似性； And the similarity between uncertain data objects was measured by combining the two important features of uncertainobjects, attribute features and the probability density function representing the characteristic of uncertainty distribution.
47179 并以此为基础提出不确定数据聚类算法； And then the uncertain data clustering algorithm was proposed.
47180 最后在 UCI 以及真实数据集上的实验结果表明，所提算法在运行效率和聚类准确性方面均能取得较好效果。 Finally, the experiment results on UCI and real datasetsindicate the better efficiency and accuracy of proposed algorithm.
47181 针对非合作接收不同符号速率 PCMA 混合信号，提出一种基于双网格逐幸存路径处理（DG-PSP, doublegrid per-survivor processing）的分离算法。 A blind separation algorithm was proposed for PCMA signals with different symbol rates based on double gridper-survivor processing (DG-PSP).
47182 将 2 路信号分量的信道状态与输入分别视为 2 组动态网格，通过分别迭代更新 2 组网格状态实现混合信号重构，进而实现 PCMA 信号盲分离。 The channel states and two input signal components were treated as two dynamic gr-ids, receiving mixed-signal reconstructed by respectively iterative updating the two groups of grid status, thus achievingblind signal separation.
47183 重点研究了联合译码的迭代分离结构，并对不同参数估计误差下算法性能做了详细分析比较。 The joint iterative decoding separation structure was focused, and a detailed analysis and comparisonunder different error estimation of parameters was shown.
47184 所提算法复杂度与传统 PSP 算法相当， The complexity of the algorithm is similar to the traditional PSPalgorithm.
47185 同时仿真结果表明，算法性能逼近联合界，当采用迭代处理时，在误比特率 10 −2 数量级上经过一次迭代就能获得 2 dB 的信噪比增益，经过 2 次迭代则有近 3 dB 的信噪比增益。 Simulation results show that, a gain of about 2 dB in signal-noise ratio can be obtained after the first iteration at abit error rate of 10 −2 , and a gain of nearly 3 dB in signal-noise ratio can be obtained after the second iteration.