Please wait a minute...

����Ŀ¼

    15 August 2019, Volume 5 Issue 4
    Comprehensive Review
    Survey of FPGA based recurrent neural network accelerator
    Chen GAO,Fan ZHANG
    2019, 5(4):  1-13.  doi:10.11959/j.issn.2096-109x.2019034
    Asbtract ( 847 )   HTML ( 39)   PDF (1423KB) ( 293 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    Recurrent neural network(RNN) has been used wildly used in machine learning field in recent years,especially in dealing with sequential learning tasks compared with other neural network like CNN.However,RNN and its variants,such as LSTM,GRU and other fully connected networks,have high computational and storage complexity,which makes its inference calculation slow and difficult to be applied in products.On the one hand,traditional computing platforms such as CPU are not suitable for large-scale matrix operation of RNN.On the other hand,the shared memory and global memory of hardware acceleration platform GPU make the power consumption of GPU-based RNN accelerator higher.More and more research has been done on the RNN accelerator of the FPGA in recent years because of its parallel computing and low power consumption performance.An overview of the researches on RNN accelerator based on FPGA in recent years is given.The optimization algorithm of software level and the architecture design of hardware level used in these accelerator are summarized and some future research directions are proposed.

    Special Column:Researches on Key Technologies of Privacy Protection and Its Innovative Applications
    Survey on the privacy-preserving content based image retrieval
    Ying WU,Xuan LI,Biao JIN,Rongrong JIN
    2019, 5(4):  14-28.  doi:10.11959/j.issn.2096-109x.2019035
    Asbtract ( 614 )   HTML ( 33)   PDF (1868KB) ( 318 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    With the widespread popularity of smart devices and social media,the number of image data is exponentially increasing.The data owners tend to outsource the local data to the cloud servers,where data is stored,shared and retrieved.However,the content of users’ image data contains a lot of sensitive information,which may be exposed to the attackers and the incomplete trusted cloud servers,resulting in a serious risk of privacy leakage to users.The research progress of the content-based image retrieval technology under the privacy-preserving in recent years were reviewed,and the key image cryptographic technology were summarized,including homomorphic encryption,randomized encryption and comparative encryption.Around these techniques,the typical solutions are analyzed and compared in detail,and the improvement strategies of index construction are introduced.Finally,the future research directions are discussed.

    Research on differential privacy protection parameter configuration method based on confidence level
    Senyou LI,Xinsheng JI,Wei YOU
    2019, 5(4):  29-39.  doi:10.11959/j.issn.2096-109x.2019036
    Asbtract ( 432 )   HTML ( 9)   PDF (1505KB) ( 141 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    In order to solve the problem that the user's real data information is disclosed during the data release and analysis process,and reduce the probability of an attacker gaining real results through differential attacks and probabilistic inference attacks,a differential privacy protection parameter configuration method based on confidence level is proposed.Analysis of attacker confidence under attacker probabilistic inference attack model and make it no higher than the privacy probability threshold set according to the data privacy attribute.The proposed method can configure more reasonable privacy protection parameters for different query privilege of query users,and avoids the risk of privacy disclosure.The experimental analysis shows that the proposed method analyzes the correspondence between attacker confidence level and privacy protection parameters based on query privilege,noise distribution characteristics and data privacy attributes,and derives the configuration formula of privacy protection parameters,which configure the appropriate parameters without violating the privacy protection probability threshold.

    Regulatory Bitcoin privacy-preserving mixing service
    Zijian BAO,Qinghao WANG,Yongxin ZHANG,Bin WANG,Ning LU,Wenbo SHI
    2019, 5(4):  40-51.  doi:10.11959/j.issn.2096-109x.2019037
    Asbtract ( 596 )   HTML ( 17)   PDF (1159KB) ( 285 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    With the deepening of research on Bitcoin privacy protection,Bitcoin,a new type of digital currency,has become more difficult to regulate.A Bitcoin privacy protection confusion scheme with regulatory RBmix is proposed for this problem.The model of RBmix uses a fair blind signature algorithm and introduces trusted third parties with regulatory,anonymity,scalability,Bitcoin compatibility,and anti-DoS aggression.Experimental results show that the RBmix protocol has good scalability and execution efficiency.

    Trajectory privacy protection mechanism of obfuscating paths based on vehicles cooperation
    Ziwen ZHAO,Ayong YE,Junlin JIN,Lingyu MENG
    2019, 5(4):  52-62.  doi:10.11959/j.issn.2096-109x.2019038
    Asbtract ( 549 )   HTML ( 8)   PDF (2433KB) ( 196 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    Vehicles sent location information to third parties to obtain location-based services in the Internet of vehicles,which may lead to vehicle trajectory privacy leakage.To address the privacy problem,trajectory privacy protection mechanism of obfuscating paths based on vehicles cooperation was proposed.Firstly,after the vehicle trajectory entropy reaches the custom trajectory protection threshold,the overhead of paths confusion saves by reducing the number of vehicle paths confusion.Then,paths confusion algorithm was designed to increase chances of vehicle paths confusion at intersections,which can improve the degree of vehicle trajectory protection.Finally,the simulation experiment verified the validity and efficiency of the method from the trajectory entropy and trajectory tracking success rate.

    Research on QR code logistics privacy based on segmented encryption and time-limited control
    Liang LIU,Wenbo GUO,Yuwei YANG,Huaiyu GUO
    2019, 5(4):  63-70.  doi:10.11959/j.issn.2096-109x.2019039
    Asbtract ( 577 )   HTML ( 17)   PDF (1819KB) ( 165 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    In view of the disadvantages of the existing logistics system in the market in user privacy protection,a privacy protection scheme based on QR code segmentation encryption,hierarchical authorization and time-limited control was proposed.This scheme segments all the recipient information into RSA and Base64 encryption,after integration,it is embedded into the QR code.In the process of logistics transportation and delivery,different levels of QR code decryption permissions are granted to different branches or transfer centers to view the designated content.At the same time,the QR code automatically becomes invalid after the user signs for it,so as to protect the user's privacy.The core idea of the scheme is to minimize the contact group of recipient information and reduce the possibility of user information disclosure.

    Papers
    Data security challenges and countermeasures in financial technology
    Jianming ZHU,Hongrui YANG
    2019, 5(4):  71-79.  doi:10.11959/j.issn.2096-109x.2019044
    Asbtract ( 1053 )   HTML ( 31)   PDF (1015KB) ( 324 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    While financial technology brings opportunities to the financial industry,it also brings new challenges.With the rapid development of financial technology,the big data explosion caused by the digitalization,networking and intelligence of financial services has put forward new and greater requirements for data security governance.Based on the analysis of the status quo of financial technology development and financial data security,a financial data security “shell” model for the characteristics of financial data was proposed:the internal is the security goals of ensuring data confidentiality,integrity,and availability; the external is a variety of threats including data leaking,tampering and destroying etc; the middle is various coping strategies,including access strategies,prevention and control strategies,detection strategies,perceived strategies and etc.According to this model,relevant suggestions and countermeasures for data security governance in financial technology are proposed.

    Multidimensional detection and dynamic defense method for link flooding attack
    Yang WANG,Guangming TANG,Cheng LEI,Dong HAN
    2019, 5(4):  80-90.  doi:10.11959/j.issn.2096-109x.2019040
    Asbtract ( 627 )   HTML ( 15)   PDF (1206KB) ( 270 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    Aiming at the shortcomings of the existing link flooding attack defense methods,a multi-dimensional index detection algorithm is proposed,which performs multi-dimensional detection on the abnormal forwarding links through the five-dimensional elements of connection length,low-speed ratio of data packets,uniformity of data packet distance,average low-speed ratio of data packets,and change rate of low-speed ratio of data packets,thus effectively solving the problem of high false alarm rate of the existing detection methods.Furthermore,a controller switch dynamic deployment method based on coloring theory is proposed,which solves the problem of difficult to be actually deployed in the actual environment with limited switch variant types existing in the existing defense mitigation mechanisms.Experimental analysis show the feasibility of the proposed method.

    Improved method of Tor network flow watermarks based on IPD interval
    Jie DU,Yongzhong HE,Ye DU
    2019, 5(4):  91-98.  doi:10.11959/j.issn.2096-109x.2019041
    Asbtract ( 400 )   HTML ( 8)   PDF (3965KB) ( 159 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    Tor is an anonymous network mechanism that provides services for hiding traffic sources,but it has the problem that the entry traffic flows of Tor are clearly identifiable.Bridge protocols such as obfs4 come into being to solve this problem,which brings new challenges that have not yet been overcome.An IPD interval scheme is proposed,which uses the clustering characteristics of k-means to improve the original scheme,so that the added flow watermark can be detected efficiently in the three modes of obfs4 bridges.The results of experiments show that the improved algorithm has higher detection rate and recognition rate,and has good adaptability to variable netflow traffic,which is conducive to the construction of a nice secure network environment.

    Analyzing the complex properties of bank account transaction network
    Fang LYU,Xinlin ZHANG,Wei WANG,Junheng HUANG,Bailing WANG
    2019, 5(4):  99-107.  doi:10.11959/j.issn.2096-109x.2019043
    Asbtract ( 535 )   HTML ( 12)   PDF (2194KB) ( 191 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    Firstly,a general directed and weighted transaction network model is built on the basis of the transaction characteristics of bank accounts.Secondly,referring to the definition of complex network,the complex network properties of the transaction network are verified from two levels of network structure and node.The network structure properties includes static properties,named scale-free property and small-world property,and dynamic evolution property,that is self-organization property.In addition,the attractor property is utilized for verifying the dynamical characteristic of complex network nodes.Finally,on the real bank account transaction data contains illegal pyramid marketing activities,the following conclusions were drawn:the network has scale-free property,small-word property,partially self-organization property and strange attractors.

    Trusted access scheme for intranet mobile terminal based on encrypted SD card
    Jiyang LI,Pengyuan ZHAO,Zhe LIU
    2019, 5(4):  108-118.  doi:10.11959/j.issn.2096-109x.2019042
    Asbtract ( 508 )   HTML ( 14)   PDF (2970KB) ( 163 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    In order to solve the problem of information security caused by illegal access of untrusted mobile terminals to intranet,a trusted access scheme for intranet mobile terminals based on encrypted SD card is designed.By using encrypted SD card as trusted hardware equipment,trusted start-up,integrity verification and Intranet trusted access of mobile terminal devices are realized.A secure storage and encrypted communication mechanism for data interaction between mobile terminal and Intranet is also created.The experimental results show that the scheme can effectively authenticate the security of mobile terminals without changing the basic architecture of mobile terminals,and to some extent protect the security of the Intranet environment.

Copyright Information
Bimonthly, started in 2015
Authorized by:Ministry of Industry and Information Technology of the People's Republic of China
Sponsored by:Posts and Telecommunications Press
Co-sponsored by:Xidian University, Beihang University, Huazhong University of Science and Technology, Zhejiang University
Edited by:Editorial Board of Chinese Journal of Network and Information Security
Editor-in-Chief:FANG Bin-xing
Executive Editor-in-Chief:LI Feng-hua
Director:YI Dong-shan
Address:F8,You Dian Publisher Building,No.11,Chengshousi Road,Fengtai District,Beijing 100078,PR China
Tel:+8610-81055479, 81055456
Fax:+86-81055464
ISSN 2096-109X
CN 10-1366/TP
visited
Total visitors:
Visitors of today:
Now online: