ID 原文 译文
44726 该算法扩展了 MSK 解调抗多普勒频偏的范围，且在复杂度和判决时间上优于同类算法，略高于 1 符号差分算法对应值，为在多普勒频偏条件下 MSK 的解调提供了新的手段。 The algorithm extends the range of MSK demodulation anti-Doppler frequency offset, and is superior to similar algorithm in complexity and decision delay time, slightly higher than 1 symbol differential algorithm cor-responding value, which provides a new means for demodulation of MSK under Doppler frequency offset.
44727 射频层析成像（RTI，radio tomographic imaging）技术作为免携带设备定位（DFL，device-free localization）的主要方式之一，在被定位目标不携带任何定位装置的情况下仍能实现定位，具有广泛的应用前景。 As one of the main methods of device free localization (DFL), the radio tomographic imaging (RTI) method that can locate a target without attaching any devices has wide application prospects.
44728 针对现有 RTI技术中椭圆权重模型的不足，基于菲涅耳衍射理论提出一种改进的椭圆阴影权重模型来提高 RTI 成像质量，并论证了这种方法的可行性； To overcome the shortcoming of the existing ellipse weight model, based on the Fresnel diffraction theory an improved ellipse weight model was proposed to enhance the imaging quality of RTI and demonstrates the feasibility of this model.
44729 同时为了克服背景噪点以及伪目标图像的影响，在此基础上提出基于十字模型的前景提取算法，进一步提高 RTI 定位性能。 Meanwhile, a foreground detection algorithm based on the cross model was proposed to reduce the impact of background noises and pseudo-targets, thereby further improving the imaging quality.
44730 室内外实验结果表明，该方法的成像质量和定位精度都要优于现有 RTI 方法。 The indoor and outdoor experimental results verify that the imaging quality and the positioning accuracy of the proposed method are better than the existing RTI methods.
44731 目前大部分中文垃圾邮件过滤系统受文本稀疏及模型特征局限的影响较大，其特征高维和特征局限的缺陷成为制约过滤效果的重要因素。 In view of the shortcoming that high dimension of features in the Chinese spam filtering system,
44732 针对特征高维问题，提出一种基于中心词扩展的 TF-IDF（term frequency-inverse document frequency）特征提取算法，增加了特征节点的表达能力，实现了特征降维。 a TF-IDFfeatures extraction algorithm was proposed based on the central word extension, the algorithm improves the expression capacity of the node in the network and reduces the dimension of feature.
44733 针对分类模型特征局限和属性间条件独立性假设不成立问题， Further, a three-layer structure model based on GWO_GA structure learning algorithm was proposed to expand the limit of text features and improve the diversity of text features.
44734 提出一种基于 GWO_GA（grey wolf optimizer-genetic algorithm）结构学习算法的3层贝叶斯网络模型， The new structure learning algorithm relaxes the conditional independence assumption of feature properties.
44735 放松了条件独立性假设，增加了特征多样性， A fine classification layer was added between class layer and feature layer to increase feature coverage.