检索结果-内蒙古大学图书馆

Acoustic backdoor attacks on speech recognition via frequency offset perturbation

Applied Soft Computing 2025年 177卷

作者： Tang, Yu Xu, Xiaolong Sun, Lijuan Jiangsu Key Laboratory of Big Data Security & Intelligent Processing Nanjing University of Posts and Telecommunications No. 9 Wenyuan Road Jiangsu Nanjing210023 China School of Computer Science Nanjing University of Posts and Telecommunications No. 9 Wenyuan Road Jiangsu Nanjing210023 China School of computer Science and Technology Huaiyin Normal University Jiangsu Huai'an223300 China

With the increasing deployment of deep learning-based speech recognition systems, backdoor attacks have become a serious security threat, enabling adversaries to implant hidden triggers that activate malicious behaviors while preserving model performance on benign inputs. However, existing acoustic backdoor attacks, whether in the time or frequency domain, often struggle to achieve sufficient stealthiness, as poisoned samples either disrupt semantic integrity or introduce perceptible artifacts. Moreover, these methods typically fail to strike an effective balance among attack efficacy, stealthiness, and robustness. To address these limitations, we propose Shadow Frequency (SF), a novel backdoor attack that leverages psychoacoustic-guided frequency offset perturbations to inject imperceptible yet model-sensitive signals near dominant spectral components. This design ensures auditory imperceptibility while maintaining high attack effectiveness and robustness. Experimental results show that SF achieves over 96% ASR with minimal impact on clean data accuracy, and remains effective under common defenses, validating its practicality for real-world deployment. © 2025 Elsevier B.V.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Personalized Privacy-Preserving Routing Mechanism Design in Payment Channel Network

引用

Journal of Computer Science & Technology 2024年第6期39卷 1380-1400页

作者： Peng-Cheng Zhao Li-Jie Xu Jia Xu Jiangsu Key Laboratory of Big Data Security and Intelligent Processing Nanjing University of Posts and TelecommunicationsNanjing 210023China

Payment Channel Network(PCN)provides the off-chain settlement of *** is one of the most promising solutions to solve the scalability issue of the *** routing techniques in PCN have been ***,both incentive attack and privacy protection have not been considered in existing *** this paper,we present an auction-based system model for PCN routing using the Laplace differential privacy *** formulate the cost optimization problem to minimize the path cost under the constraints of the Hashed Time-Lock Contract(HTLC)tolerance and the channel *** propose an approximation algorithm to find the top K shortest paths constrained by the HTLC tolerance and the channel capacity,i.e.,top K-restricted shortest ***,we design the probability comparison function to find the path with the largest probability of having the lowest path cost among the top K-restricted shortest paths as the final ***,we apply the binary search to calculate the transaction fee of each *** both theoretical analysis and extensive simulations,we demonstrate that the proposed routing mechanism can guarantee the truthfulness and individual rationality with the probabilities of 1/2 and 1/4,*** can also ensure the differential privacy of the *** experiments on the real-world datasets demonstrate that the privacy leakage of the proposed mechanism is 73.21%lower than that of the unified privacy protection mechanism with only 13.2%more path cost compared with the algorithm without privacy protection on average.

关键词： blockchain payment channel network routing mechanism differential privacy personalized privacy-preserving

来源：评论

学校读者我要写书评

暂无评论

FDSR-INT: A Flexible On-Demand In-Band Telemetry Approach for Aerial Computing Networks

引用

IEEE Internet of Things Journal 2025年

作者： Xu, Jinghan Xu, Xiaolong Zhao, Juan Gao, Honghao Nanjing University of Posts and Telecommunications School of Computer Science Nanjing210023 China Nanjing University of Posts and Telecommunications Jiangsu Key Laboratory of Big Data Security and Intelligent Processing The School of Computer Science Nanjing210023 China Jinling Institute of Technology School of Computer Science Nanjing211169 China Shanghai University School of Computer Engineering and Science Shanghai200444 China Gachon University College of Future Industry Gyeonggi-Do 461-701 Korea Republic of

In-band network telemetry (INT) is a new network measurement technique that provides real-time, fine-grained packet-level network measurements. However, standard INT lacks the flexibility to perform configurable on-demand network measurements. In this work, we propose a flexible on-demand network measurement mechanism (FDSR-INT) based on a dual bitmap by combining the programmability of segment routing based on IPv6 (SRv6) and the telemetry efficiency of the INT to achieve customizable network measurements. By designing the dual bitmap, the telemetry information of personalized probe nodes is supported, and the telemetry efficiency is improved. SRv6 is employed to direct measurement probe packets, ensuring coverage of a specified set of network targets. Its inherent programmability enables INT to conduct customized per-hop network measurements. The designed flexible INT field structure, which appends traceability information along with telemetry information, reduces the information traceability overhead in the control plane and improves the network compatibility of the mechanism. We solve the optimal probing path by solving a traveler problem in an auxiliary graph and design a greedy pathcutting algorithm to maximize the number of nodes for single packet probing while satisfying the packet length constraints to improve the success rate of the probing task. Finally, we implemented FDSR-INT using P4 and verified its performance experimentally in the constructed *** integrated simulation dynamic environment. FDSR-INT saves 30% of the data plane bandwidth and 54% of the data plane bandwidth on *** interfaces compared with SONM-SR-INT, etc. Furthermore, it has low control plane processing overhead and probe path transmission overhead. © 2014 IEEE.

关键词： Bandwidth

来源：评论

学校读者我要写书评

暂无评论

Identifying eclipsing binary stars with TESS data based on a new hybrid deep learning model

arXiv

引用

arXiv 2025年

作者： Shan, Ying Chen, Jing Zhang, Zichong Wang, Liang Zou, Zhiqiang Li, Min College of Computer Nanjing University of Posts and Telecommunications Nanjing210023 China Jiangsu Key Laboratory of Big Data Security and Intelligent Processing Nanjing210023 China Nanjing Institute of Astronomical and Optics & Technology Chinese Academy of Sciences Nanjing210042 China CAS Key Laboratory of Astronomical Optics & Technology Nanjing Institute of Astronomical Optics & Technology Nanjing210042 China University of Chinese Academy of Sciences Beijing100049 China University of Chinese Academy of Sciences Jiangsu Nanjing211135 China

Eclipsing binary systems (EBs), as foundational objects in stellar astrophysics, have garnered significant attention in recent years. These systems exhibit periodic decreases in light intensity when one star obscures the other from the observer’s perspective, producing characteristic light curves (LCs). With the advent of the Transiting Exoplanet Survey Satellite (TESS), a vast repository of stellar LCs has become available, offering unprecedented opportunities for discovering new EBs. To efficiently identify such systems, we propose a novel method that combines LC data and generalized Lomb-Scargle periodograms (GLS) data to classify EBs. At the core of this method is CNN Attention LSTM Net (CALNet), a hybrid deep learning model integrating Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM) networks, and an Attention Mechanism based on the Convolutional Block Attention Module (CBAM). We collected 4,225 EB samples, utilizing their 2-minute cadence LCs for model training and validation. CALNet achieved a recall rate of 99.1%, demonstrating its robustness and effectiveness. Applying it to TESS 2-minute LCs from Sectors 1 to 74, we identified 9,351 new EBs after manual visual inspection, significantly expanding the known sample size. This work highlights the potential of advanced deep-learning techniques in large-scale astronomical surveys and provides a valuable resource for further studies on EBs. © 2025, CC BY-SA.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Resource allocation in cognitive wireless powered communication networks with wirelessly powered secondary users and primary users

引用

Science China(Information Sciences) 2019年第2期62卷 224-229页

作者： Ding XU Qun LI Wireless Communication Key Lab of Jiangsu Province Nanjing University of Posts and Telecommunications Jiangsu Key Lab of Big Data Security and Intelligent Processing Nanjing University of Posts and Telecommunications

Dear editor,Wireless powered communication networks(WPCNs) are popular especially for wireless sensor networks where sensors can be wirelessly powered. For coordinating wireless power and information transfer, Ju and Zhang in their pioneer work [1] proposed a "harvest-then-transmit" protocol for WPCN, where the time for wireless power transfer (WPT) and wireless information transfer(WIT) is divided into two phases:the WPT phase and WIT phase. Meanwhile, cognitive radio (CR)

关键词： WIT Resource allocation in cognitive wireless powered communication networks with wirelessly powered secondary users and primary users

来源：评论

学校读者我要写书评

暂无评论

Learning Comprehensive Representation via Selective Activation and Dual-Level Orthogonality for Pedestrian Attribute Recognition

引用

IEEE Transactions on Circuits and Systems for Video Technology 2025年

作者： Wu, Junyi Huang, Yan Gao, Min Niu, Yuzhen Chen, Yuzhong Wu, Qiang Zhao, Jianqiang Ministry of Education Engineering Research Center of Big Data Intelligence China Ai Research Center Sdic Intelligence Xiamen Information Co. Ltd Xiamen China Xiamen Meiya Pico Information Security Research Institute Co. Ltd Xiamen China Institute of Automation Beijing China Fuzhou University Fujian Key Lab for Intelligent Processing and Wireless Transmission of Media Information College of Physics and Information Engineering Fuzhou China University of Technology Sydney School of Electrical and Data Engineering Ultimo Australia

Multi-label Pedestrian Attribute Recognition (PAR) involves identifying a series of semantic attributes in person images. Existing PAR solutions typically rely on CNN as the backbone network to extract pedestrian features. Unfortunately, CNNs process only one adjacent region at a time, resulting in the disappearance of long-range relations between different attribute-specific regions. To address this limitation, we adopt the Vision Transformer (ViT) instead of CNN as the backbone for PAR, aiming to build long-range relations and extract more robust features. However, PAR suffers from an inherent attribute imbalance issue, causing ViT to naturally focus more on attributes that appear frequently in the training set and ignore some pedestrian attributes that appear less. The native features extracted by ViT are not able to tolerate the imbalance attribute distribution issue. To tackle this issue, we propose a novel component and a dual-level loss: the Selective Feature Activation Method (SFAM), the Orthogonal Feature Activation Loss (OFALoss), and Orthogonal Weight Regularization Loss (OWRLoss). SFAM smartly suppresses the more informative attribute-specific features, thus compelling the PAR model to pay greater attention to attribute-specific regions that are often overlooked. The proposed OFALoss enforces an orthogonal constraint on the original feature extracted by ViT and the suppressed features from SFAM, promoting the comprehensiveness of feature representation in each attribute-specific region. Furthermore, OWRLoss is employed for decreasing correlations among entries of the last shared classification layer, which can alleviate the highly correlated of weight vectors caused by non-uniform distribution. This can prevent excessive mutual interference among different attributes during attribute recognition. Our model-agnostic approach is plug-and-play, requiring no additional training parameters in the training process. We conduct experiments on several benchmark P

关键词： Dual-Level Orthogonal Regularization Pedestrian Attribute Recognition Selective Feature Activation

来源：评论

学校读者我要写书评

暂无评论

FIMI： A Constant Frugal Incentive Mechanism for Time WindowCoverage in Mobile Crowdsensing

引用

Journal of Computer Science & Technology 2017年第5期32卷 919-935页

作者： Jia Xu Jian-Ren Fu De-Jun Yang Li-Jie Xu Lei Wang Tao Li School of Computer Science Nanjing University of Posts and Telecommunications Nanjing 210023 China Jiangsu Key Laboratory of Big Data Security and Intelligent Processing Nanjing University of Posts and Telecommunications Nanjing 210023 China Jiangsu Key Laboratory of Big Data Security and Intelligent Processing Nanjing University of Posts and Telecommunications Nanjing 210023 China Department of Computer Science Colorado School of Mines Golden 80401 U.S.A.

Mobile crowdsensing has become an efficient paradigm for performing large-scale sensing tasks. An incentive mechanism is important for a mobile crowdsensing system to stimulate participants and to achieve good service quality. In this paper, we explore truthful incentive mechanisms that focus on minimizing the total payment for a novel scenario, where the platform needs the complete sensing data in a requested time window (RTW). We model this scenario as a reverse auction and design FIMI, a constant frugal incentive mechanism for time window coverage. FIMI consists of two phases, the candidate selection phase and the winner selection phase. In the candidate selection phase, it selects two most competitive disjoint feasible user sets. Afterwards, in the winner selection phase, it finds all the interchangeable user sets through a graph-theoretic approach. For every pair of such user sets, FIMI chooses one of them by the weighted cost. Further, we extend FIMI to the scenario where the RTW needs to be covered more than once. Through both rigorous theoretical analysis and extensive simulations, we demonstrate that the proposed mechanisms achieve the properties of RTW feasibility (or RTW multi-coverage), computation efficiency, individual rationality, truthfulness, and constant frugality.

关键词： crowdsensing incentive mechanism constant frugality

来源：评论

学校读者我要写书评

暂无评论

Server load prediction algorithm based on CM-MC for cloud systems

引用

Journal of Systems Engineering and Electronics 2018年第5期29卷 1069-1078页

作者： XU Xiaolong ZHANG Qitong MOU Yiqi LU Xinyuan Jiangsu Key Laboratory of Big Data Security & Intelligent Processing Nanjing University of Posts and Telecommunications State Key Laboratory of Information Security Chinese Academy of Sciences Shanghai Stock Exchange Technology Co. LtdShanghai Stock Exchange Institute of Big Data Research at Yancheng Nanjing University of Posts and Telecommunications

Accurate prediction of server load is important to cloud systems for improving the resource utilization, reducing the energy consumption and guaranteeing the quality of service(QoS).This paper analyzes the features of cloud server load and the advantages and disadvantages of typical server load prediction algorithms, integrates the cloud model(CM) and the Markov chain(MC) together to realize a new CM-MC algorithm, and then proposes a new server load prediction algorithm based on CM-MC for cloud systems. The algorithm utilizes the historical data sample training method of the cloud model, and utilizes the Markov prediction theory to obtain the membership degree vector, based on which the weighted sum of the predicted values is used for the cloud model. The experiments show that the proposed prediction algorithm has higher prediction accuracy than other typical server load prediction algorithms, especially if the data has significant volatility. The proposed server load prediction algorithm based on CM-MC is suitable for cloud systems, and can help to reduce the energy consumption of cloud data centers.

关键词： cloud computing load prediction cloud model Markov chain energy saving

来源：评论

学校读者我要写书评

暂无评论

MedBLIP: A Multimodal Method of Medical Question-Answering Based on Fine-Tuning Large Language Model

引用

Computerized Medical Imaging and Graphics 2025年

作者： Lejun Gong Jiaming Yang Shengyuan Han Yimu Ji Jiangsu Key Lab of Big Data Security & Intelligent Processing School of computer science Nanjing University of Posts and Telecommunications Nanjing210023 China Institute of High Performance Computing and Bigdata Nanjing University of Posts and Telecommunications Nanjing 210023 China

Medical visual question answering is crucial for effectively interpreting medical images containing clinically relevant information. This study proposes a method called MedBLIP (Medical Treatment Bootstrapping Language-Image Pretraining) to tackle visual language generation tasks related to chest X-rays in the medical field. The method combine an image encoder with a large-scale language model, and effectively generates medical question-answering text through a strategy of freezing the image encoder based on the BLIP-2 model. Firstly, chest X-ray images are preprocessed, and an image sample generation algorithm is used to enhance the text data of doctor-patient question-answering, thereby increasing data diversity. Then, a multi-layer convolutional image feature extractor is introduced to better capture the feature representation of medical images. During the fine-tuning process of the large language generation model, a new unfreezing strategy is proposed, which is to unfreeze different proportions of the weights of the fully connected layer to adapt to the data in the medical field. The image feature extractor is responsible for extracting key features from images, providing the model with rich visual information, while the text feature extractor accurately captures the essential requirements of the user's question. Through their synergistic interaction, the model can more effectively integrate medical images and user inquiries, thereby generating more accurate and relevant output content. The experimental results show that unfreezing 31.25% of the weights of the fully connected layer can significantly improve the performance of the model, with ROUGE-L reaching 66.12%, and providing a more accurate and efficient answer generation solution for the medical field. The method of this study has potential applications in the field of medical language generation tasks. Although the proposed model cannot yet fully replace human radiologists, it plays an indispensable role

关键词：

来源：评论

学校读者我要写书评

暂无评论

Multi-task regression learning for survival analysis via prior information guided transductive matrix completion

引用

Frontiers of Computer Science 2020年第5期14卷 99-112页

作者： Lei Chen Kai Shao Xianzhong Long Lingsheng Wang Jiangsu Key Laboratory of Big Data Security and Intelligent Processing Nanjing University of Posts and TelecommunicationsNanjing210023China MIIT Key Laboratory of Pattern Analysis and Machine Intelligence Nanjing University of Aeronautics and AstronauticsNanjing211106China

Survival analysis aims to predict the occurrence time of a particular event of interest,which is crucial for the prognosis analysis of ***,due to the limited study period and potential losing tracks,the observed data inevitably involve some censored instances,and thus brings a unique challenge that distinguishes from the general regression *** addition,survival analysis also suffers from other inherent challenges such as the high-dimension and small-sample-size *** address these challenges,we propose a novel multi-task regression learning model,i.e.,prior information guided transductive matrix completion(PigTMC)model,to predict the survival status of the new ***,we use the multi-label transductive matrix completion framework to leverage the censored instances together with the uncensored instances as the training samples,and simultaneously employ the multi-task transductive feature selection scheme to alleviate the overfitting issue caused by high-dimension and small-sample-size *** addition,we employ the prior temporal stability of the survival statuses at adjacent time intervals to guide survival ***,we design an optimization algorithm with guaranteed convergence to solve the proposed PigTMC ***,the extensive experiments performed on the real microarray gene expression datasets demonstrate that our proposed model outperforms the previously widely used competing methods.

关键词： survival analysis matrix completion multi-task regression transductive learning multi-task feature selection

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：