成果

2024

1. Paper

   基于线性卷积混叠过程的自监督式光纤传感信号分离

   陈照 , and  刘泽超^*

   北京航空航天大学学报, 2024

   Abs Bib

   针对光纤传感信号分离问题，提出了一种基于线性卷积混叠过程的自监督式信号分离方法，其主要包含三部分：线性卷积混叠模型、语义特征提取模型和基于查询的信号分离模型。在模型训练过程中，相比于线性叠加模型，混叠模型可依更贴合实际传感过程的线性卷积混叠方式对输入的多个子信号进行混叠，动态生成混叠信号，服务于后续分离模型的自监督式学习；然后利用语义特征提取模型将某一子信号映射至特征空间；最后将其特征作为查询因子，并与混叠信号一起输入到分离模型中，最终输出目标子信号，在可成倍扩充训练样本的同时也可实现零样本条件下的目标信号分离。为验证方法的有效性，在室内环境下开展实验并采集了跑步时及单频正弦振动下的光纤传感信号及两者的混叠信号，且在该实测数据上的实验结果表明了该方法的有效性。

   @article{convSigSep,
     title = {基于线性卷积混叠过程的自监督式光纤传感信号分离},
     journal = {北京航空航天大学学报},
     volume = {},
     pages = {},
     year = {2024},
     issn = {1001-5965},
     doi = {http://dx.doi.org/10.13700/j.bh.1001-5965.2024.0409},
     url = {https://bhxb.buaa.edu.cn/bhzk/cn/article/doi/10.13700/j.bh.1001-5965.2024.0409},
     author = {陈照 and 刘泽超},
     keywords = {信号分离, 自监督, 卷积混叠, 光纤传感, 零样本},
   }

2. Paper

   POFNet: Enlarging the Dynamic Range of Phase Demodulation by Signal Fusion

   Zechao Liu, Hongkun Zheng, Chen Zhu, and Lingmei Ma^*

   In 2024 Conference on Lasers and Electro-Optics (CLEO) , 2024

   Abs Bib

   To break the π-phase criterion in phase demodulation, POFNet is proposed to fuse the OPD and the phase signal to estimate the unwrapping errors and recover larger phase. The results showcase its superiority performance.

   @inproceedings{pofnet_cleo,
     author = {Liu, Zechao and Zheng, Hongkun and Zhu, Chen and Ma, Lingmei},
     booktitle = {2024 Conference on Lasers and Electro-Optics (CLEO)},
     title = {POFNet: Enlarging the Dynamic Range of Phase Demodulation by Signal Fusion},
     year = {2024},
     volume = {},
     number = {},
     pages = {AF2D.7},
     doi = {https://doi.org/doi:10.1364/cleo_at.2024.af2d.7},
     url = {https://opg.optica.org/abstract.cfm?URI=CLEO_AT-2024-AF2D.7},
     publisher = {Optica Publishing Group},
   }

3. Paper

   Large Dynamic Range and High Frequency Acoustic Signal Detection with SNR Enhancement

   Hongkun Zheng, Lingmei Ma^* , Zechao Liu, Caiyun Li, and 3 more authors

   In 2024 Conference on Lasers and Electro-Optics (CLEO) , 2024

   Abs Bib

   An OPD-based demodulation method for acoustic signal detection is proposed, breaking the limit of phase demodulation. Experiment results demonstrated superior performance at high frequency, and the SNR increases proportionally with the frequency of the signal.

   @inproceedings{zhkcleo2024,
     author = {Zheng, Hongkun and Ma, Lingmei and Liu, Zechao and Li, Caiyun and Zhuang, Yiyang and Zhu, Chen and Rao, Yunjiang},
     booktitle = {2024 Conference on Lasers and Electro-Optics (CLEO)},
     journal = {CLEO 2024},
     keywords = {Fiber optic sensors; Light sources; Michelson interferometers; Optical fibers; Phase; Phase unwrapping},
     pages = {AM2A.3},
     publisher = {Optica Publishing Group},
     title = {Large Dynamic Range and High Frequency Acoustic Signal Detection with SNR Enhancement},
     year = {2024},
     url = {https://opg.optica.org/abstract.cfm?URI=CLEO_AT-2024-AM2A.3},
     doi = {https://doi.org/10.1364/CLEO_AT.2024.AM2A.3},
   }

4. Paper

   Detection of Large Dynamic Range Acoustic Signal Using OPD-based Fiber-Optic Interferometric Demodulation with SNR Enhancement

   Hongkun Zheng, Lingmei Ma^* , Zechao Liu, Caiyun Li, and 4 more authors

   Journal of Lightwave Technology, 2024

   Abs Bib

   The measurement of acoustic signals at high frequencies with a large dynamic range remains a challenge for phase-demodulation-based fiber acoustic sensing systems due to the phase wrapping phenomenon. Here, an optical path difference (OPD)-assisted demodulation method aiming at breaking the limitations of phase demodulation is reported. By interrogating an imbalanced Michelson interferometer (IMI) with a frequency-modulated pulse light, a time-resolved interference spectrum is obtained and used for OPD demodulation. Theoretical analysis reveals that the variation amplitude of demodulated OPD signal not only is linearly dependent on the amplitude of the acoustic signal applied to the IMI, but also experiences a gain that is directly proportional to the signal frequency. Consequently, compensation of this gain introduces a noise suppression that favors high-frequency signals, resulting in a higher signal-to-noise ratio (SNR) at high frequencies. The method’s capability for detecting both single-frequency and multi-frequency acoustic signals is demonstrated by simulations and experiments, and the results show that the upper-frequency limit of the proposed scheme is at least 40 times the limit of the phase demodulation method for an acoustic signal with an amplitude of 1.135 μϵ. The proposed method can be easily extended to distributed acoustic sensing systems and is of great potential to break the limitation of a restricted dynamic range encountered in traditional phase demodulation methods.

   @article{zhkjlt,
     author = {Zheng, Hongkun and Ma, Lingmei and Liu, Zechao and Li, Caiyun and Jie, Ruimin and Zhuang, Yiyang and Zhu, Chen and Rao, Yunjiang},
     journal = {Journal of Lightwave Technology},
     title = {Detection of Large Dynamic Range Acoustic Signal Using OPD-based Fiber-Optic Interferometric Demodulation with SNR Enhancement},
     year = {2024},
     volume = {},
     number = {},
     pages = {1-9},
     keywords = {Acoustics;Demodulation;Optical fiber sensors;Interference;Dynamic range;Interferometers;Frequency modulation;acoustic measurement;interferometer;demodulation;optical fiber sensor;optical path difference},
     doi = {10.1109/JLT.2024.3371702},
   }

2023

1. Paper

   Deep Hyperspherical Clustering for Skin Lesion Medical Image Segmentation

   Zuowei Zhang, Songtao Ye , Zechao Liu, Hao Wang^*, and 1 more author

   IEEE Journal of Biomedical and Health Informatics, 2023

   Abs Bib

   Diagnosis of skin lesions based on imaging techniques remains a challenging task because data (knowledge) uncertainty may reduce accuracy and lead to imprecise results. This paper investigates a new deep hyperspherical clustering (DHC) method for skin lesion medical image segmentation by combining deep convolutional neural networks and the theory of belief functions (TBF). The proposed DHC aims to eliminate the dependence on labeled data, improve segmentation performance, and characterize the imprecision caused by data (knowledge) uncertainty. First, the SLIC superpixel algorithm is employed to group the image into multiple meaningful superpixels, aiming to maximize the use of context without destroying the boundary information. Second, an autoencoder network is designed to transform the superpixels’ information into potential features. Third, a hypersphere loss is developed to train the autoencoder network. The loss is defined to map the input to a pair of hyperspheres so that the network can perceive tiny differences. Finally, the result is redistributed to characterize the imprecision caused by data (knowledge) uncertainty based on the TBF. The proposed DHC method can well characterize the imprecision between skin lesions and non-lesions, which is particularly important for the medical procedures. A series of experiments on four dermoscopic benchmark datasets demonstrate that the proposed DHC yields better segmentation performance, increasing the accuracy of the predictions while can perceive imprecise regions compared to other typical methods.

   @article{hypersc,
     author = {Zhang, Zuowei and Ye, Songtao and Liu, Zechao and Wang, Hao and Ding, Weiping},
     journal = {IEEE Journal of Biomedical and Health Informatics},
     title = {Deep Hyperspherical Clustering for Skin Lesion Medical Image Segmentation},
     year = {2023},
     volume = {27},
     number = {8},
     pages = {3770-3781},
     keywords = {Skin;Lesions;Image segmentation;Feature extraction;Biomedical imaging;Medical diagnostic imaging;Optimization;Melanoma;clustering;imprecision;skin lesion;deep learning;belief functions;image segmentation},
     doi = {10.1109/JBHI.2023.3240297},
   }

2. Paper

   Phase correction based SNR enhancement for distributed acoustic sensing with strong environmental background interference

   Caiyun Li , Zechao Liu, Yiyang Zhuang, Hongkun Zheng, and 5 more authors

   Optics and Lasers in Engineering, 2023

   Abs Bib

   Fiber-optic distributed acoustic sensing (DAS) systems based on phase-sensitive optical time domain reflectometry (Φ-OTDR) measure acoustic waves by demodulating the phase variations of the Rayleigh backscattering (RBS) signals in a sensing optical fiber. However, in harsh environments, strong environmental background interference, coupled with the interference fading of the RBS, would introduce severe distortions in DAS signal that cannot be corrected or even can be worsened by phase unwrapping, thus resulting in low signal to noise ratio (SNR) and even undetectable signal. In this work, a novel method based on trend prediction is proposed to correct the distortions induced by phase unwrapping error around the fading points. The method is further validated in processing the data acquired from a field test performed in ocean environments using a home-built DAS system. By locating and erasing the distortion points, and then detrending, the acoustic signal buried in strong environmental background interference is retrieved with an SNR improvement greater than 10 dB. The results show that the proposed phase correction method can effectively enhance DAS’s SNR for those challenging applications with strong background interference.

   @article{aunwrap,
     title = {Phase correction based SNR enhancement for distributed acoustic sensing with strong environmental background interference},
     journal = {Optics and Lasers in Engineering},
     volume = {168},
     pages = {107678},
     year = {2023},
     issn = {0143-8166},
     doi = {https://doi.org/10.1016/j.optlaseng.2023.107678},
     url = {https://www.sciencedirect.com/science/article/pii/S0143816623002075},
     author = {Li, Caiyun and Liu, Zechao and Zhuang, Yiyang and Zheng, Hongkun and Zhu, Chen and Hu, Weiwang and Wang, Jianguo and Ma, Lingmei and Rao, Yun-Jiang},
     keywords = {Fiber-optic distributed acoustic sensing systems (DAS), Phase-sensitive optical time domain reflectometry (φ-OTDR), Unwrapped phase, Interference fading, Large dynamic range, Phase distrotion},
   }

3. Paper

   Mining and reasoning of data uncertainty-induced imprecision in deep image classification

   Zuowei Zhang, Liangbo Ning , Zechao Liu, Qingyu Yang, and 1 more author

   Information Fusion, 2023

   Abs Bib

   Existing deep image classification techniques strive to suppress data uncertainty for various reasons such as blur, occlusion, noise, and label error and advance to higher accuracy. However, they ignore data uncertainty-induced imprecision and thus do not work as intended. In this paper, we propose a deep open-source framework to mine and reason the imprecision of such training and test sets, which can present better performance and reduce the risk of misclassification. First, we design a label reassignment mechanism. It allows the network to reassign training labels and allow imperfect training samples with multiple labels. As a result, they are removed from the original classes and considered new imprecise samples to represent partial ignorance. Second, we propose a new imbalanced data enhancement architecture to learn a generalized representation of each (precise and imprecise) class. It helps the network fuse the auxiliary information from both precise and imprecise classes, which is beneficial to extract more distinctive class features from single-labeled samples and characterize uncertainty-induced imprecision in the test set by imprecise test samples. Afterward, methodological analyses and empirical evaluations are conducted. The proposed framework is demonstrated to present better performance on different typical networks (Resnet50, MobileNetV2, DenseNet121, EfficientNetB0, ShuffleNetV2, SENet, SqueezeNet, Xception) based on five publicly available datasets (Imagewoof-5, Flowers, Monkey, Butterfly, and Cifar-10). In addition, several targeted deep techniques for uncertain images or imprecise results are also employed as comparisons to prove the superiority of the proposed framework.

   @article{ZHANG2023202,
     title = {Mining and reasoning of data uncertainty-induced imprecision in deep image classification},
     journal = {Information Fusion},
     volume = {96},
     pages = {202-213},
     year = {2023},
     issn = {1566-2535},
     doi = {https://doi.org/10.1016/j.inffus.2023.03.014},
     url = {https://www.sciencedirect.com/science/article/pii/S1566253523001033},
     author = {Zhang, Zuowei and Ning, Liangbo and Liu, Zechao and Yang, Qingyu and Ding, Weiping},
     keywords = {Imprecision, Data uncertainty, Image classification, Deep techniques, Multiple labels},
   }

4. Paper

   SNR Improvement for Distributed Acoustic Sensing with Strong Environmental Background Interference

   Caiyun Li , Zechao Liu, Hongkun Zheng, Yiyang Zhuang, and 5 more authors

   In 28th International Conference on Optical Fiber Sensors , 2023

   Abs Bib

   A novel method is proposed to correct the distortions induced by phase unwrapping error. The method is further validated in processing the data acquired from a field test performed in ocean environments using a DAS system.

   @inproceedings{unwrapofs,
     author = {Li, Caiyun and Liu, Zechao and Zheng, Hongkun and Zhuang, Yiyang and Zhu, Chen and Hu, Weiwang and Wang, Jianguo and Ma, Lingmei and Rao, Yun-Jiang},
     booktitle = {28th International Conference on Optical Fiber Sensors},
     journal = {28th International Conference on Optical Fiber Sensors},
     keywords = {Distortion; Erbium fibers; Fiber Bragg gratings; Optical time domain reflectometry; Phase unwrapping; Spatial resolution},
     pages = {W4.71},
     publisher = {Optica Publishing Group},
     title = {SNR Improvement for Distributed Acoustic Sensing with Strong Environmental Background Interference},
     year = {2023},
     url = {https://opg.optica.org/abstract.cfm?URI=OFS-2023-W4.71},
     doi = {10.1364/OFS.2023.W4.71},
   }

5. Patent

   基于自监督式信号融合的DAS动态范围提升方法和设备

   刘泽超,  马玲梅,  李彩云,  朱琛, and 3 more authors

   2023

   Abs Bib

   本发明涉及一种基于自监督式信号融合的DAS动态范围提升方法和设备,方法包括：并行获取双路光纤传感数据,即动态范围未受限制但信噪比较低的DAS信号和因动态范围受限而存在相位解卷绕错误的DAS相位信号,输入自设计的自监督式信号融合模型中,求取相位卷绕系数,并取整,根据取整后的相位卷绕系数和第二光纤传感信号,构建解卷绕后的相位估计值并输出。与现有技术相比,本发明可实时有效修正DAS相位信号因动态范围受限而导致的相位解卷绕错误,恢复真实相位,提升动态范围,且不以牺牲信噪比为代价。

   @patent{CN116956226B,
     author = {刘泽超 and 马玲梅 and 李彩云 and 朱琛 and 郑洪坤 and 彭威 and 田帅飞},
     title = {基于自监督式信号融合的DAS动态范围提升方法和设备},
     edition = {CN116956226B},
     year = {2023},
     pages = {16},
     address = {311121 浙江省杭州市余杭区中泰街道科创大道之江实验室},
   }

6. Patent

   一种光纤传感行人识别方法、装置和存储介质

   彭威 ,  刘泽超,  马玲梅,  王皓, and 1 more author

   2023

   Abs Bib

   本发明涉及一种光纤传感行人识别方法、装置和存储介质,方法包括以下步骤：S1、获取行人经过时的传感数据,对传感数据进行解调,得到相位数据；S2、对相位数据进行切面降采样；S3、对切面降采样后的数据通过矩形滑窗进行切割,得到时间序列信号；S4、将时间序列信号进行尺度划分,得到第一高尺度数据、第一中尺度数据和第一低尺度数据；S5、将划分得到的数据输入训练好的阶梯形的稠密卷积网络结构中,得到分类识别结果。与现有技术相比,本发明基于多尺度和注意力机制,实现高精度的行人动作识别。

   @patent{CN116541744A,
     author = {彭威 and 刘泽超 and 马玲梅 and 王皓 and 程徐},
     title = {一种光纤传感行人识别方法、装置和存储介质},
     edition = {CN116541744A},
     year = {2023},
     pages = {20},
     address = {311121 浙江省杭州市余杭区中泰街道之江实验室南湖总部},
   }

7. Patent

   基于畸变点检测的光纤传感信号相位错误修正方法和装置

   刘泽超,  李彩云,  马玲梅,  胡威旺, and 2 more authors

   2023

   Abs Bib

   本发明涉及一种基于畸变点检测的光纤传感信号相位错误修正方法和装置,方法包括以下步骤：获取光纤传感数据,进行解调,得到相位数据,该相位数据包括光纤上多个位置处的相位变化信号；对相位数据中的各个位置信号分别进行相位解卷绕,然后进行重采样,获取各个位置的相位解卷绕数据；根据相位解卷绕数据的信号变化量,确定畸变点；根据畸变点的分布确定畸变点的类型,可分为尖峰畸变点、连续台阶畸变点和单次畸变点,从而进行对应的修正处理,然后再经去趋势操作,输出修正后的相位信号。与现有技术相比,本发明可实时有效修正复杂情况下的相位错误,抑制衰落噪声,提升信噪比。

   @patent{CN115855123B,
     author = {刘泽超 and 李彩云 and 马玲梅 and 胡威旺 and 彭威 and 王皓},
     title = {基于畸变点检测的光纤传感信号相位错误修正方法和装置},
     edition = {CN115855123B},
     year = {2023},
     pages = {27},
     address = {311121 浙江省杭州市余杭区中泰街道之江实验室南湖总部},
   }

8. Patent

   基于分层变分自编码的跨源舰船特征融合学习与识别方法

   文载道 ,  刘泽超,  刘准钆, and  潘泉

   2023

   Abs Bib

   本发明公开了一种基于分层变分自编码的跨源舰船特征融合学习与识别方法,获取目标舰船的光学或合成孔径雷达待识别图像；使用训练好的分层式变分自编码网络中的第一编码器提取待识别图像中的舰船类别间差异性特征及数据源间差异性特征；使用训练好的分层式变分自编码网络中的第二编码器对舰船类别间差异性特征及数据源间差异性特征进行分析,确定待识别图像中的目标舰船的类别及待识别图像的数据源类别；本发明利用分层式变分自编码网络,从大量的无法配准的异源舰船目标图像中自动提取表示性/解释性与判别性兼具的结构化特征,实现跨源舰船特征融合学习,以及舰船目标的精准识别。

   @patent{CN111291639B,
     author = {文载道 and 刘泽超 and 刘准钆 and 潘泉},
     title = {基于分层变分自编码的跨源舰船特征融合学习与识别方法},
     edition = {CN111291639B},
     year = {2023},
     pages = {15},
     address = {710072 陕西省西安市友谊西路127号},
   }

2022

1. Patent

   一种基于深度学习的光纤传感水声信号识别方法及装置

   高嘉豪,  彭威 ,  刘泽超,  王皓, and 3 more authors

   2022

   Abs Bib

   本发明提供一种基于深度学习的光纤传感水声信号识别方法,该方法降低了光纤传感水声信号识别的难度,通过最优聚类模型,将无监督学习方式转化为有监督学习的方式,使识别未知的目标事件信号成为可能；以光纤传感系统自身固有噪声信号分解分量作为训练数据,构建开集识别网络,可用于识别任意不属于系统噪声的目标事件信号,有效提高了模型的泛化能力。

   @patent{CN114818839B,
     author = {高嘉豪 and 彭威 and 刘泽超 and 王皓 and 马玲梅 and 饶云江 and 叶松涛},
     title = {一种基于深度学习的光纤传感水声信号识别方法及装置},
     edition = {CN114818839B},
     year = {2022},
     pages = {17},
     address = {311121 浙江省杭州市余杭区之江实验室南湖总部},
   }

2. Patent

   一种两段式光纤传感水声信号补偿方法和装置

   彭威 ,  刘泽超,  王皓,  马玲梅, and 1 more author

   2022

   Abs Bib

   本发明公开了一种两段式光纤传感水声信号补偿方法和装置。该水声信号补偿方法在传感信号频域分解处理的过程中,首先通过遗传算法寻找传感信号变分模态分解的最优惩罚因子和迭代阈值参数,使得信号处理过程中尽可能降低信号损失；其次,将损失信号通过基于多尺度排列熵的补偿算法,回补至各模态分量,使得水声传感信号在频域分解处理的过程中尽可能少的损失有用信息,提升传感信号的信噪比。

   @patent{CN114754857B,
     author = {彭威 and 刘泽超 and 王皓 and 马玲梅 and 饶云江},
     title = {一种两段式光纤传感水声信号补偿方法和装置},
     edition = {CN114754857B},
     year = {2022},
     pages = {19},
     address = {310023 浙江省杭州市余杭区之江实验室南湖总部},
   }

3. Patent

   一种基于自适应VMD的φ-OTDR水声信号处理方法和装置

   彭威 ,  刘泽超,  王皓,  马玲梅, and 2 more authors

   2022

   Abs Bib

   本发明公开了一种基于自适应VMD的φ-OTDR水声信号处理方法和装置。该水声信号处理方法将信号从时域转换至频域进行分析,对光纤传感器中不同位置的传感信号进行变分模态分解处理,基于信号噪声的特性,提取全变分、分形维数、排列熵、能量特征用于噪声信号的特征离散化。同时,我们根据最大化簇间间距和最小化簇内间距原则设计了一种信号可分离性指标,来观测和优化变分模态分解过程,使得传感信号分解得到的模态分量可以更清晰的划分噪声和目标信号。

   @patent{CN114077854B,
     author = {彭威 and 刘泽超 and 王皓 and 马玲梅 and 饶云江 and 叶松涛},
     title = {一种基于自适应VMD的φ-OTDR水声信号处理方法和装置},
     edition = {CN114077854B},
     year = {2022},
     pages = {15},
     address = {310023 浙江省杭州市余杭区之江实验室南湖总部},
   }

4. Patent

   一种基于参考传感器的分布式声波传感降噪系统及方法

   马玲梅,  应马可,  胡威旺 ,  刘泽超, and 2 more authors

   2022

   Abs Bib

   本发明公开了一种基于参考传感器的分布式声波传感降噪系统及方法,该系统包括相位型光时域反射模块,参考传感器,噪声补偿算法模块；该方法包括：S1:建立相位型光时域反射模块；S2:设立参考传感器用以获得所述相位型光时域反射模块的噪声特征；S3:通过噪声补偿算法模块对所述噪声特征计算补偿。本发明通过参考传感器收集本地信号,通过参考端收集的信号训练深度神经网络,对噪声进行预测,由此实现噪声补偿的功能,本发明提出多种参考传感器的架构,其实现方法简便,手段灵活,可在不同使用环境中实现高可靠,低延迟的实时降噪补偿功能。

   @patent{CN113654642B,
     author = {马玲梅 and 应马可 and 胡威旺 and 刘泽超 and 王皓 and 饶云江},
     title = {一种基于参考传感器的分布式声波传感降噪系统及方法},
     edition = {CN113654642B},
     year = {2022},
     pages = {17},
     address = {310023 浙江省杭州市余杭区文一西路1818号},
   }

2020

1. Paper

   A new incomplete pattern belief classification method with multiple estimations based on KNN

   Zong-fang Ma, Hong-peng Tian , Ze-chao Liu, and Zuo-wei Zhang^*

   Applied Soft Computing, 2020

   Abs Bib

   The classification of missing data is a challenging task, because the lack of pattern attributes may bring uncertainty to the classification results and most classification methods produce only one estimation, which may have a risk of misclassification. A new incomplete pattern belief classification (PBC) method with multiple estimations based on K-nearest neighbors (KNNs) is proposed to deal with missing data. PBC preliminarily classifies the incomplete pattern using its KNNs obtained by the known attributes. The pattern whose KNNs contain only one class information can be directly divided into this class. If not, the p (p≤c) estimations will be computed according to the different KNNs in different classes when p classes are included in the KNNs of the pattern and it will yield p pieces of classification results by the chosen classifier. Then, a weighted possibility distance method is used to further divide the p classification results with their KNNs’ classification information. The pattern with similar possibility distances in different classes will be reasonably classified into a proper meta-class under the framework of belief functions theory, which truly reflects the uncertainty of the pattern caused by missing values and effectively reduces the error rate. Experiments on both artificial and real data sets show that PBC is effective for dealing with missing data.

   @article{bf-knn,
     title = {A new incomplete pattern belief classification method with multiple estimations based on KNN},
     journal = {Applied Soft Computing},
     volume = {90},
     pages = {106175},
     year = {2020},
     issn = {1568-4946},
     doi = {https://doi.org/10.1016/j.asoc.2020.106175},
     url = {https://www.sciencedirect.com/science/article/pii/S1568494620301150},
     author = {Ma, Zong-fang and Tian, Hong-peng and Liu, Ze-chao and Zhang, Zuo-wei},
     keywords = {Missing data, -nearest neighbors, Possibility distance, Belief functions theory, Uncertainty},
   }