Table of Content

  

20 September 2016
Volume 51 Issue 9

Information retrieval: the final direction of human language technology research in DARPA

CAO Rong, HUANG Jin-zhu, YI Mian-zhu

JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2016, 51(9):  11-17.  doi:10.6040/j.issn.1671-9352.1.2015.C05

Abstract ( 1502 )   PDF (3020KB) ( 1389 )   Save

References | Related Articles | Metrics

In order to procure the relevant research and application of human language technology in American defense field, the design ideas, progress and goal of DARPA since its coming into being is combed comprehensively and systematically. The focus of DARPA covered speech recognition, machine translation, information retrieval, language resource construction, language technology evaluation and system integration, etc. Judged from the distribution of research fields, expected goals and application ideas, the ultimate goal of DARPA human language technology research is highly-efficient, highly-adapted, highly-intelligent IR technology, which is applied to expand the scope of military operations, downsize operators and lift operation speed. Meanwhile, DARPAs innovation and military requirement-driven practice are worth learning.

Cooperative group signature scheme with controllable linkability

LIU Xin, XU Qiu-liang, ZHANG Bo

JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2016, 51(9):  18-35.  doi:10.6040/j.issn.1671-9352.0.2016.003

Abstract ( 1296 )   PDF (2837KB) ( 1249 )   Save

References | Related Articles | Metrics

The existing group signature schemes with controllable linkability(abbreviated as GS-OL)suffer from the following two deficiencies, i.e., group members’ private keys are likely to leak and the Open Authority may harm users’ privacy. Moreover, there is still room for improvement in the computational efficiency of users. To correct this situation, an extension of the existing GS-OL security model was made, and an improved scheme with higher efficiency was provided. By using the Boneh-Boyen signature scheme without pairings, the users were liberated from performing expensive pairing computations. Borrowing the ideas of cooperative group signatures, the problem of private key leakage of group members was also solved. Moreover, in order to decentralize the Open Authoritys power, the technique of distributed key generation, committed zero-knowledge proof of knowledge, and distributed proof of knowledge of preimages of group homomorphism were incorporated in the new scheme. In addition, a practical anonymous subscription system based on the new GS-OL scheme was proposed. Compared with previous systems, the new system meets several practical properties such as allowing customers’ tokens to expire and supporting privacy-preserving data mining, and it has obvious advantage in users’ operational efficiency.

A new quantum secret sharing scheme

XU Sheng-wei

JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2016, 51(9):  36-40.  doi:10.6040/j.issn.1671-9352.3.2015.027

Abstract ( 1363 )   PDF (1440KB) ( 890 )   Save

References | Related Articles | Metrics

In the practical applications, member expansion is a usual demand during the development of a secret sharing network. However, there are few consideration on network expansibility in the existing quantum secret sharing schemes.We proposed an expansible quantum secret sharing scheme with relatively simple and economical quantum resources. Its trait, no requirement of any agents assistant during the process of member expansion, can help to prevent potential menaces of insider cheating. We also gave a discussion on the security of this scheme from three aspects.

Research on the localization of firmware vulnerability based on stain tracking

DAI Zhong-hua, FEI Yong-kang, ZHAO Bo, WANG Ting

JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2016, 51(9):  41-46.  doi:10.6040/j.issn.1671-9352.2.2015.245

Abstract ( 1895 )   PDF (2130KB) ( 1993 )   Save

References | Related Articles | Metrics

In the process of vulnerability detection, because of the limited physical device and closed operating system, we cant confirm and utilize the bug discovered by Fuzzing test in time. So this paper concentrated on embedded firmware and proposed a firmware vulnerability analysis and utilization method based on stain tracking. The method uses dynamic analysis technique in the simulation environment. With the help of this method, we could rapidly locate the exception position and trace back to the origin, then its possible to assess the corresponding solution rapidly. Furthermore, we did experiments on many devices such as router and IP camera and successfully utilized many 0day bugs on ARM and MIPS architecture. According to the results, the firmware vulnerability analysis and utilization method based on device simulation debugging is referable to vulnerability positioning and utilization of embedded firmware.

A zero-knowledge proof based remote desktop authentication protocol

ZHU Zhi-qiang, MA Ke-xin, SUN Lei

JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2016, 51(9):  47-52.  doi:10.6040/j.issn.1671-9352.2.2015.139

Abstract ( 1526 )   PDF (2062KB) ( 825 )   Save

References | Related Articles | Metrics

Aiming at the possibility of user attributes revelation during the user identity authentication period under the desktop cloud environment, a remote desktop authentication model was presented. On the basis of the model, a zero-knowledge proof based remote desktop authentication protocol ZPBRDAP was presented. The protocol consisted of attributes registration protocol and identity authentication protocol, which could accomplish user identity authentication without exposing user identity attributes. At last,the validity, the security and the efficiency of the protocol was analyzed. It is demonstrated that the protocol has a strong usability.

Multi-component property based remote attestation

LI Xiao-ce, PAN Xiao-zhong, MAI Tao-tao

JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2016, 51(9):  53-58.  doi:10.6040/j.issn.1671-9352.2.2015.180

Abstract ( 1063 )   PDF (1935KB) ( 881 )   Save

References | Related Articles | Metrics

On the basis of component property-based attestation, multi-component property based attestation(MCPBA)is proposed to solve the attestation problem with the security property consists of some components in a special structure. This scheme can satisfy the attestation requirements of different kind of properties. A several-for-one mapping between components and property is established based on blocks of messages signature in Camenisch and Lysyanskaya signature scheme. MCPBA has an efficient component property attestation method, which is suitable for both multi and single component-property. The component property based attestation that proved secure in RO model under strong RSA assumption is the secure base of MCPBA, to make certain of privacy protection. MCPBA is suitable for the property attestation that has a special components structure.

A system of monitoring and protecting Android privacy leakage

JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2016, 51(9):  59-67.  doi:10.6040/j.issn.1671-9352.1.2015.293

Abstract ( 2771 )   PDF (3349KB) ( 6254 )   Save

References | Related Articles | Metrics

In order to protect the data of Android operating system, we present a system called FlowMonitor to monitor and protect privacy flow based on Android. FlowMonitor adds taint marks and identity tags onto data. Moreover, it is a real-time, dynamic monitoring system. In order to avoid leaking private data by collusion attack, it uses the mechanism of taint's identity authentication based on the idea of RBAC, and finds all probable malicious applications. Based on open and closed policies, labels were designed for each data to control the use of private data. The user can control applications using private data based on their choices. Evaluation results show that Flowmonitor out performs taint Droid. It can not only find out all probable malicious applications which steal private data by collusion attack, but also control applications using each private data. Extensive experimental results demonstrate that FlowMonitor effectively protects the user privacy.

Research on watermarking payload of the spatial images based on MPUI model

MAO Jia-fa, HUANG Yan-hong, NIU Xin-xin, XIAO Gang, ZHU Li-nan, SHENG Wei-guo

JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2016, 51(9):  68-75.  doi:10.6040/j.issn.1671-9352.3.2015.069

Abstract ( 1501 )   PDF (2821KB) ( 759 )   Save

References | Related Articles | Metrics

Based on the MPUI model, we study the maximum watermarking payload of spatial domain image, which is not only related to embedding intensity but also to the factors such as the size of image, image complexity and visual sensitively, and so forth. We derived the relationship among maximum watermarking payload, size of image and embedding intensity, and adopt multivariate regression analysis method to research image complexity, visual sensitivity, and embedding rate model. Synthesize these internal and external factors, with the aid of subjective and objective evaluation system, we estimate the estimation method of watermark embedded payload and verify its effectiveness through experiments.

Reversible data hiding in encrypted image based on an integer transform

JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2016, 51(9):  76-83.  doi:10.6040/j.issn.1671-9352.2.2015.221

Abstract ( 1281 )   PDF (3247KB) ( 1053 )   Save

References | Related Articles | Metrics

This paper proposed a novel reversible data hiding scheme in encrypted image which exploits an integer transform to achieve both high embedding capacity and security.In the proposed scheme, content owner implements an integer transform on the pixel groups which are selected by a transform key and composed of two adjacent pixels.Based on properties of the transform,the original pixels values can be recovered perfectly if one of LSBs(Least Significant Bit)in the transformed pixel group is altered.Data hider can seek the transformed pixel groups out from the encrypted image with the same transform key, select one pixel in each transformed group using an embedding key and embed additional data reversibly by replacing corresponding LSBs of the chosen pixels. Only if the receiver has both the transform key and embedding key, can the secret data be extracted without error.Only if the receiver has both the transform key and encryption key, can the original image be recovered perfectly. The experimental results demonstrate that additional data can be embedded and extracted directly in encrypted domain and the original image can be recovered losslessly. Besides, the embedding rate can reach 0.47 bpp(bit per pixel).

A P2P network privacy protection system based on anonymous broadcast encryption scheme

LI Yu-xi, WANG Kai-xuan, LIN Mu-qing, ZHOU Fu-cai

JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2016, 51(9):  84-91.  doi:10.6040/j.issn.1671-9352.3.2015.067

Abstract ( 1672 )   PDF (2292KB) ( 1285 )   Save

References | Related Articles | Metrics

Aiming at the problem that the existing broadcast encryption scheme cannot protect the users privacy well, we proposed a broadcast encryption scheme with the receiver anonymity based on composite order bilinear groups and Lagrange interpolation method. And aiming at the low privacy protection in the existing P2P network system, we constructed a P2P network privacy protection system based on the proposed anonymous broadcast encryption scheme. Then,we designed and implemented function modules of the system by Android, and verified every modules function. The results show that the proposed anonymous broadcast encryption algorithm can distribute the users data and effectively protect the privacy of the receiver, at the same time, solve the problems that exist in the existing in P2P networks which rely on expensive anonymous technology to protect user data privacy and the excessive open of user permission.

An approach of detecting LDoS attacks based on the euclidean distance of available bandwidth in cloud computing

JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2016, 51(9):  92-100.  doi:10.6040/j.issn.1671-9352.3.2015.090

Abstract ( 1451 )   PDF (3403KB) ( 885 )   Save

References | Related Articles | Metrics

According to the architecture of the cloud computing Data Center Networks(DCNs), the Low-rate Denial of Service(LDoS)attack is modeled from the view of network architecture. Furthermore, the euclidean approach is applied to the available bandwidth to detect LDoS attacks. As LDoS attacks force the links co-located in the same routing domain to increase their available bandwidths, the average euclidean distance is applied as the measurement for detecting LDoS attacks. And then, the traditional Probe Gap Model(PGM)is improved to test the available bandwidth specifically in cloud computing. Experiments in practical network are conducted to test the attack effect and the detection performance. Test results verify: 1)LDoS attacks present more damages than Flooding Denial of Service(FDoS)attacks in cloud computing DCNs, 2)The proposed detection approach can detect LDoS attack accurately, and achieves 98% detection probability.

Fast authentication mechanism based on Diffie-Hellman for wireless mesh networks

SU Bin-ting, XU Li, FANG He, WANG Feng

JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2016, 51(9):  101-105.  doi:10.6040/j.issn.1671-9352.2.2015.096

Abstract ( 1271 )   PDF (1940KB) ( 1092 )   Save

References | Related Articles | Metrics

In order to ensure the celerity and security in mobile client switching process of wireless Mesh networks, the fast login authentication and fast handover authentication scheme based on Diffie-Hellman algorithms were proposed. In login authentication process, the Mesh client is authenticated successfully with 4 rounds of message exchange by predistribution of tickets, followed by calculating their shared handover key and delivering it to any of its neighbor. With the shared key, the handover authentication process takes a 3-message handshake to accomplish mutual authentication when clients move to a new router. During the process, the authentication server does not need to be involved. The analysis of the scheme security show that the proposed schemes have less communication and computation overhead, lower authentication latency and higher authentication efficiency. And it is secure if the user is considered trustworthy.

K-Canopy:a fast data segmentation algorithm for the topic detection

CHEN Qiang, DU Pan, CHEN Hai-qiang, BAO Xiu-guo, LIU Yue, CHENG Xue-qi

JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2016, 51(9):  106-112.  doi:10.6040/j.issn.1671-9352.1.2015.057

Abstract ( 1297 )   PDF (1858KB) ( 767 )   Save

References | Related Articles | Metrics

This paper presented a pre-clustering algorithm for tasks of topic detection on big data. To support the parallelization of the successive topic detection task,the proposed algorithm was designed to segment the dataset according to the semantic association among data points as evenly and efficiently as possible. The experimental result shows that our proposed algorithm is effective at segmenting dataset while preserving semantic association inside data blocks, and is helpful for improving the efficiency and effectiveness of topic detection.

EIP: discovering influential bloggers by user similarity and topic timeliness

ZHU Sheng, ZHOU Bin, ZHU Xiang

JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2016, 51(9):  113-120.  doi:10.6040/j.issn.1671-9352.2.2015.228

Abstract ( 1441 )   PDF (3273KB) ( 993 )   Save

References | Related Articles | Metrics

Enormous information flowing through Online Social Media nowadays, spreading through hundreds of millions of users with different influence in the network. EIP(extended influence-passivity), an extension of IP(influence passivity)algorithm, is proposed to identify influencers in social network based on users forwarding activity. EIP measures the influence and passivity of users taking both pair wise topical similarity and timeliness feature of information into account. An evaluation performed with about 100 000 user dataset crawled from Sina micro-blog shows that EIP outperforms than other algorithms, including the original IP and TwitterRank.

News event extraction based on kernel dependency graph

LIN Li

JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2016, 51(9):  121-126.  doi:10.6040/j.issn.1671-9352.1.2015.C01

Abstract ( 1376 )   PDF (2448KB) ( 1007 )   Save

References | Related Articles | Metrics

The main process of event extraction based on kernel dependency graph(KDG)is to find and match the semantic structure. A Vietnamese-English-Chinese FrameNet on the South China Sea news has been built to explore the specific application of semantic structure extraction including KDG semantic analysis model, KGD automatic generation and event templates extraction based on KDG. The research focuses are the analysis models for typical KDGs, special KDGs of Null Instantiated Frame Element and frame element fusion; the representation method of KDG for event information extraction; the process of KGD automatic generation from annotated sentences. It is demonstrated that event extraction based on KDG is driven by linguistic motivation and feasible to find out the semantic clues of news texts by its intuitiveness. At the present, it is possible to generate KDG automatically from the frame-semantic annotated sentences and extract the corresponding event templates.

A forensic analysis algorithm of registry reverse reconstruction based on physical memory

GAO Yuan-zhao, LI Bing-long, WU Xi-xi

JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2016, 51(9):  127-136.  doi:10.6040/j.issn.1671-9352.2.2015.140

Abstract ( 1335 )   PDF (3362KB) ( 717 )   Save

References | Related Articles | Metrics

The reconstruction and analysis of the registry is one of the most important and difficult aspects of the Windows physical memory forensics. By analyzing the logical structure of the registry files in the hard disk and exploring the data structure features of the registry in the physical memory based on the Windows debugging tools, we proposed a clear and definite method to locate the registry physical addresses in the memory. Furthermore, by analyzing the tree-structured relationship between the entries of the registry, we designed a registry reconstruction algorithm and implemented a dendrogram visualization algorithm for the reconstructed registry based on Graphviz. The results of the experiment show that we can reconstruct of the names and values of the registry entries, retrieve the virus in the system based on the information we got, and finally display the process and results of the virus infection via the registry visualization.

A distributed training method for L1 regularized kernel machines based on filtering mechanism

JI Xin-rong, HOU Cui-qin, HOU Yi-bin, ZHAO Bin

JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2016, 51(9):  137-144.  doi:10.6040/j.issn.1671-9352.3.2015.064

Abstract ( 1468 )   PDF (2325KB) ( 770 )   Save

References | Related Articles | Metrics

To decrease the amount of data transferred and the computing cost during training a kernel machine in wireless sensor network, a distributed training method for L1-regularized Kernel Minimum Square Error machine based on filtering mechanism was proposed. First, filtering mechanism of samples was presented and used on each node. Second, with consistency constraint on the local model of each node and its local optimal one obtained by exchanging the local model with its all neighbours, the distributed optimization problem of L1-regularized Kernel Minimum Square Error machine was solved by Augmented Lagrange Method of Multipliers, and the local optimization problem of L1-regularized Kernel Minimum Square Error machine on each node was solved by Alternating Direction Method of Multipliers. Then, the spares model obtained on each node was transferred to its all neighbor nodes. This process iterates until the local model on each node converges. For carrying out this method,a novel distributed training algorithm for L1-regularized Kernel Minimum Square Error based on filtering of samples was proposed. Simulation results prove the validity of the proposed algorithm in terms of convergence, sparse rate of model, the amount of data transferred and the number of samples used in model training.

Linear complexity of balanced quaternary generalized cyclotomic sequences with Period pq

JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2016, 51(9):  145-150.  doi:10.6040/j.issn.1671-9352.2.2015.230

Abstract ( 1247 )   PDF (1593KB) ( 572 )   Save

References | Related Articles | Metrics

Combined the theory of Gray mapping and cyclostomes, a new class of sequences over Z4 with Period pq was established. we will determine the corresponding Fourier spectral sequence of the new sequence on the finite field of F_r(r≥5, prime). Then, we will obtain the linear complexity of the new sequence from the weights of its Fourier spectral sequence. Results show that the sequences have larger linear complexity and can resist the attack by B-M algorithm. Its a good sequence from the viewpoint of cryptography.