Author Login Editor-in-Chief Peer Review Editor Work Office Work

15 March 2017, Volume 43 Issue 3
    

  • Select all
    |
  • HUANG Tinghui,WANG Yuliang,WANG Zhen,CUI Gengshen
    Computer Engineering. 2017, 43(3): 1-6. https://doi.org/10.3969/j.issn.1000-3428.2017.03.001
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Based on the analysis of the key technologies of Spark,such as flexible distributed data set and Spark task scheduling,it is concluded that time of I/O in data processing has a great effect on the computing performance of Spark. Aiming at this problem,this paper studies the run mode of Spark consolidating files that can reduce the number of cache files and improve the I/O efficiency of Spark to some extent,but it still has the disadvantage of high memory cost. Further more,the paper proposes an improved process of Spark Shuffle which designs a mode that every Mapper only generates one cache file,and every Mapper’s bucket shares the same memory buffer,thus these improve I/O efficiency and reduce the memory overhead. Simulation results show that,compared with the default mode of Spark,the I/O time of a wide dependent process is shortened by 42.9%,which improves the memory utilization and the efficiency of the Spark platform.
  • FENG Chaozheng,JIANG Yi,HE Jun,MA Xiangjun
    Computer Engineering. 2017, 43(3): 7-10,17. https://doi.org/10.3969/j.issn.1000-3428.2017.03.002
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    The Auto-Sharding mechanism in MongoDB database finishes shard migration only through the data quantity,which causes unbalanced load imbalance.Aiming at this problem,this paper proposes an optimized Auto-Sharding mechanism based on the access characteristics of hot and cold data.It uses the naive Bayes algorithm to determine the data access characteristics of hot and cold data,and takes the proportion of the hot data in a data block as the heat load to determine the data migration time.It establishes new data migration strategy through the heat load differences between data blocks.Experimental results show that the data throughput of the improved mechanism is obviously better than that of the original Auto-Sharding mechanism under high concurrent condition.
  • KAN Wenxiao,HUANG Qiulan,CHEN Gang
    Computer Engineering. 2017, 43(3): 11-17. https://doi.org/10.3969/j.issn.1000-3428.2017.03.003
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Desktop grid is a heterogeneous computing resource environment,which challenges the cross platform application deployment and heterogeneous resource integration.With good encapsulation and isolation,virtualization technology is used for integrating heterogeneous resources and shielding the difference of different platforms.Hence,a desktop grid system is implemented based on VirtualBox Virtnal Machine(VM).After testing different configuration parameters of VM,self-adaptive memory allocation module is designed.And based on the different stages of job operation,analysis of the influence on job operation caused by different performance of hard disk I/O and bandwidth I/O is given.According to the analysis result,VM image diversity and the performance loss caused by virtual machine start-up and bandwidth,the remote image library and dyamic job file system are designed,which meet the diverse requirement of VM image and quick deployment of job running environment.Meanwhile,cache service is used to make up for the loss caused by the use of VM.Application results show that this virtual desktop grid system can meet the needs of high energy physics computing very well.
  • XIA Cuicui,LIU Mengchi,HU Jie
    Computer Engineering. 2017, 43(3): 18-23,31. https://doi.org/10.3969/j.issn.1000-3428.2017.03.004
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    The information of Web entity is scattered in different places and lack of semantics. Traditional search engines based on keyword matching cannot get accurate results for lacking of context information. To get accurate information on the Web,Information Networking Model(INM) is used to directly describe relationships between entities in the real world,and to represent and model the semantic information of Web data. The semantic information about one entity is organized into an object,which makes it more quickly to acquire complete entity information. Based on the semantic database generated with INM,a semantic information search system named TLDW is built,which can be used to search semantic information of entities in the education domain. Experimental results show that first task searching time is less than 100 ms,and search results contain rich entity semantic information like context relationship. On the basis of the cache optimization,searching task can be finished within 20 ms.
  • ZHAO Jinming,QIAN Lei,WU Dong,HAO Ziyu
    Computer Engineering. 2017, 43(3): 24-31. https://doi.org/10.3969/j.issn.1000-3428.2017.03.005
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    To solve the problems of inefficient queries,low concurrency,small database capacity and bad scalability of the existing NativeXML database,a large-scale distributed NativeXML database prototype is designed based on efficient and multi-purpose computing platform.It adopts hardware-software co-design and designs a query filtering structure with high concurrency,fully pipelining and configurability in the hardware to process queries efficiently,while using the software to flexibly configure the database and do other complex processing like sorting and filtering.This paper also gives the solutions for the problems of existing NativeXML database.Experimental results show that the proposed prototype is feasible and effective.It has about 30 times speedup over Tamino database in query efficiency,supports high concurrency and has good scalability.
  • XIAN Xuefeng,CUI Zhiming,FANG Ligang,GU Caidong,SUN Xun
    Computer Engineering. 2017, 43(3): 32-39. https://doi.org/10.3969/j.issn.1000-3428.2017.03.006
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at the problems that the data source based on the selection method of data source quality exists in selection process are heavy crawling price and high repetition rate,this paper proposes a two-layer selection model for source selection and integration. The selection model is built based on the quality and utility of the data source,and a recursive incremental data source selection and integration strategy is presented based on the model. The strategy adopts a data source quality classifier to filter majority low-grade Deep Web resources,only leaveing several high-quality ones as the input of the second layer utility classifier. The second layer classifier chooses the processed candidate resources recursively,which enables the integrated system to extract as much high qualified resources while escaping to get high coverage over k. Experimental results show that,combined the ascendency of two classifiers,the designed model can reduce the space of candidate data resources while assuring the quality,and it simultaneously avoids processing huge amounts of repeated data and reduces the integrated cost of Deep Web resources extraction effectively.
  • CHEN Xuan,YIN Baoqun,SHI Hao
    Computer Engineering. 2017, 43(3): 40-45. https://doi.org/10.3969/j.issn.1000-3428.2017.03.007
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    The manifold structure of the traffic matrix is studied by using classical manifold learning algorithms,and the intrinsic dimension of the flow matrix can be got. However,the Locally Linear Embedding(LLE) algorithms rely on the selection of neighbor points. The traditional method is often obtained by experiment. Even so,this does not fully reveal the manifold structure of traffic matrix. Aiming at the above shortcomings,this paper improves the LLE algorithm by improving the LLE algorithm. It applies the improved LLE algorithm to real OD traffic matrix taken from the backbone network (Abilene).Experimental result shows the improved LLE algorithm has a smaller reconstruction error,compared with standard LLE algorithm,it can reveal a low-dimensional manifold structure exactly.
  • CHEN Zhitao,AN Hong,QIU Xiaojie,LI Feng,CHEN Junshi
    Computer Engineering. 2017, 43(3): 46-50,56. https://doi.org/10.3969/j.issn.1000-3428.2017.03.008
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at the impact of Dynamic Voltage Frequency Scaling(DVFS) on program’s performance and power consumption,based on interval division,this paper uses existing performance monitoring unit provided by commercial processor,it proposes a performance prediction model which takes the relation between memory access latency and frequency into account. It presents an energy-optimized DVFS(eDVFS) control policy that aims to minimize energy consumption. Experimental results show that eDVFS can get 23% at most and 6.85% on average in energy saving compared with Linux ‘ondemand’ adjusting policy.
  • YANG Qi,CHE Ming
    Computer Engineering. 2017, 43(3): 51-56. https://doi.org/10.3969/j.issn.1000-3428.2017.03.009
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    To control the mechanical device of 3D digital sandbox,a 3D standard module digital sandbox based on embedded system is designed and implemented. Aiming at the problems that the errors of hardware system of mechanical device lead to abnormality of process interruption and the target data are lost when the power is off,a set of hardware interrupt error compensation and data protection schemes are proposed to increase accuracy and reliability of the mechanical device. The conceptual interrupt filter layer which can filter pulses and judge the validity of interrupt is designed to compensate the hardware process interrupt error to reserve the correct pulses and increase accuracy of interrupt module. The data protection scheme against power failure is based on Electrically Erasable Programmable Read-only Memory(EEPROM) storing target data,data status and standard module status. The recovery of device is determined by the standard module status,and the validity of target data is determined by the data status,which provide valid data for device recovery. Theor etical analysis and experimental results show that compared with the original methods,the error probability of the scheme decreases to 0.5%,which can effectively reduce the error rate of mechanical devices and recover the data validly to improve the stability and reliability of mechanical device.
  • ZHANG Jing,SUN Shaojie,FAN Hongbo,XIAO Zhibin
    Computer Engineering. 2017, 43(3): 57-61,68. https://doi.org/10.3969/j.issn.1000-3428.2017.03.010
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Signal task scheduling algorithm is the key to the effectiveness of the implementation of Cyber Physical System(CPS).When the system is fully loaded,the Least Slack First(LSF) algorithm,Earliest Deadline First(EDF) algorithm or Highest Value First(HVF) algorithm cannot finish the task scheduling perfectly and the system energy consumption is higher than normal condition.Therefore,this paper proposes an improved scheduling algorithm.The algorithm can schedule system task dynamically by introducing task scheduling priority and actual task scheduling priority with combining task energy consumption,task completion value and task urgency.Experimental results show that for the same task set and the same number of tasks,the system energy consumption of the improved algorithm is less than that of LSF algorithm and EDF algorithm.When the system is fully loaded,compared with HVF algorithm,the improved algorithm can use less energy consumption to achieve the same total tasks’ value.
  • YAN Xiaoxing,CHE Ming,GAO Xiaojuan
    Computer Engineering. 2017, 43(3): 62-68. https://doi.org/10.3969/j.issn.1000-3428.2017.03.011
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    To construct moderate remote data transmission systems,reliable remote transmission system based on BeiDou short-message protocol is designed and implemented. It adopts packet loss feedback and retransmission mechanism to increase reliability of data transmission and proposes packet assembly and fragmentation mechanism to improve data transmission efficiency and guarantee the correct data transmission.Meanwhile,the system communication protocol is customized by expanding message content of BeiDou short-message communication protocol. Moreover,an embedded Beidou Communication Device (BCD) is developed and a corresponding management mechanism is provided,realizing robust transmission of remote data. It uses BCDs as intermediary of data transmission for experiment. Result shows that the packet loss rate is decreased by 46.877%on average with packet loss feedback and retransmission mechanism than no feedback,which ensures the reliability of BeiDou remote data transmission.
  • ZENG Youcai,JING Xiaorong
    Computer Engineering. 2017, 43(3): 69-74. https://doi.org/10.3969/j.issn.1000-3428.2017.03.012
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    For the co-channel interference in large-scale Multiple Input Multiple Output(MIMO) systems,this paper proposes a multiuser precoding scheme based on 3D codebooks. On the basis of uniform planar array antennas,the vertical codebook is designed by the distribution characteristics of vertical declivitous angle,and the existing Discrete Fourier Transform(DFT) codebook is used as the horizontal codebook. 3D codebooks based on the Kronecker product operation are reconstructed. Codeword selection and user scheduling strategy are combined to minimize the multiuser interference in the large-scale MIMO system.Simulation results show that,compared with the 3D-DFT codebook precoding scheme,the proposed scheme can more accurately distinguish the differences of the vertical channel information,and hence effectively suppress multiuser interference and improve system rate significantly.
  • SUN Zhongyue,JIA Xinghua
    Computer Engineering. 2017, 43(3): 75-78. https://doi.org/10.3969/j.issn.1000-3428.2017.03.013
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Taking into account the impact of network static topology on network capacity,a method is proposed to enhance the network capacity by deleting some edges with high betweenness.With this method packets can make a detour round the hub nodes and network load can be balanced.According to edge betweenness sorting,it ensures to delete edges under the condition of network connection.Simulation experiment is done under different node forwarding capacity model.Results show that the method is simple to operate and is effective.It not only improves the network capacity but also reduces the cost of network maintenance.
  • LONG Ken,LIU Yuezhen
    Computer Engineering. 2017, 43(3): 79-83. https://doi.org/10.3969/j.issn.1000-3428.2017.03.014
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at the problem of maximizing traverse rate in massive Multiple Input Multiple Output(MIMO) system over the 3D relevant channel,the spatial degrees of freedom and accuracy factor of the available Channel State Information(CSI) are introduced in capacity analysis. System capacity expression is derived under the spatial degrees of freedom and different accuracy of CSI. When capacity is maximized,conditions that the joint optimization problem is simplified into for the user number optimization are given. It can be derived that the unique optimal solution is obtained by using a midpoint. Simulation results show that under the condition that accuracy factor of CSI is 0.1 and the spatial degrees of freedom is 200,the optimal algorithm based on midpoint achieves almost the same capacity and user number as the poor search algorithm based on the number of users and regularization parameter,with rapid convergence speed and lower complexity.
  • ZHANG Yunyan,WANG Hui
    Computer Engineering. 2017, 43(3): 84-88. https://doi.org/10.3969/j.issn.1000-3428.2017.03.015
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at the problem that the multi-user MIMO indoor Visible Light Communication(VLC) system based on Tomlinson Harashima Precoding(THP) has high Bit Error Rate(BER) and low power utilization,this paper proposes a Multi-user MIMO(MU-MIMO) communication system based on lattice reduction THP.It uses lattice reduction THP to optimize channel matrix and make it full rank.A multi-user received signal balanced distribution scheme is designed based on this system.Simulation results indicate that MU-MIMO communication system has high robustness and low BER even under the condition of high Signal to Noise Ratio(SNR) and unstable channel.The fair distribution problem of the information received by users effectively by the multi-user received signal balanced distribution scheme.
  • PENG Yi,CUI Ziru,CHEN Changkai
    Computer Engineering. 2017, 43(3): 89-93. https://doi.org/10.3969/j.issn.1000-3428.2017.03.016
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In Wireless Cognitive Network(WCN),conventional cooperative spectrum sensing algorithms are used to enhance the accuracy of spectrum sensing,but cooperative spectrum sensing algorithm increases communication costs and the energy consumption of secondary users.For this reason,a Selective Trigger Cooperative Spectrum Sensing(STCSS) algorithm is proposed,this algorithm determines whether to trigger the secondary cooperative spectrum sensing based on the threshold of current spectrum sensing energy.This makes cooperative spectrum sensing is triggered only when necessary,thereby reduces energy consumption and communication overhead.Simulation results show that,compared with Cooperative Spectrum Sensing(CSS) and Fine And Fast Sensing Algorithm(FAF),STCSS algorithm enhances the detection probability and reduces energy consumption and overhead of communication.
  • XU Zhexin,ZHANG Yilin,LIN Xiao,WU Yi
    Computer Engineering. 2017, 43(3): 94-98,104. https://doi.org/10.3969/j.issn.1000-3428.2017.03.017
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Smart home router is designed to realize the interconnection between the home equipment sensor network and the Internet.It uses Zstack protocol stack and open source OpenWrt system to processe ZigBee and WIFI network packets respectively,so as to achieve barrier-free transmission of packets in the application layer and provide corresponding data service.Service sofrware of smart home scene design is used to provide management services for devices in smart home system.These two services are combined with the WiFi router and ZigBee coordinator to form a router router suitable for smart home system.Using Packet Sniffer and Wireshark to grab the packet from ZigBee and Internet,the function of the proposed router is analyzed and verified.It shows that this router can meet the demand of surfing the Internet from consumer devices,such as mobile phone,tablet PC or other smart home devices.It reduces the installation costs and the hardware complexity of smart home devices.
  • XU Zhoubo,HU Kui,CHANG Liang,GU Tianlong
    Computer Engineering. 2017, 43(3): 99-104. https://doi.org/10.3969/j.issn.1000-3428.2017.03.018
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In order to solve the problem of high storage redundancy and low search efficiency caused by the increasing number of routing table entries in the proess of routing lookup,an improved routing lookup algorithm based on Algebraic Decision Diagram(ADD) is proposed. According to the characteristic of symbolic algorithm,routing table entries are expressed as a pseudo Boolean function. Considering the features of routing table structure and the advantages of symbolic algorithm,a prefix-based routing table is constructed by ADD structure,and the algorithm of routing table update,deleting and look-up is also given. With the open source routing table provided by International Project Management Association(IPMA),the experimental results show that the algorithm can effectively reduce the number of memory access in routing table operation and save the routing table storage space.
  • KONG Fanzeng,ZHENG Nae,CHEN Guojun,YU Haitao
    Computer Engineering. 2017, 43(3): 105-109,114. https://doi.org/10.3969/j.issn.1000-3428.2017.03.019
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Existing Time of Arrival/Angle of Arrival(TOA/AOA) positioning algorithms based on Wireless Sensor Network(WSN) have low accuracy under the condition that the position of Anchor Nodes(AN) has error. Fusing Geometric Dilution of Precision(GDOP) weighting,an improved hybrid TOA/AOA positioning algorithm is proposed in this paper. According to the measurement error of positioning parameters from the ANs and its position error,GDOP calculation formula is derived for TOA/AOA positioning algorithm which is based on single AN. According to the measurement error of all ANs and the relationship between BN and ANs in the WSN,the GDOP formula is derived based on weighting fusion algorithm. Simulation results show that positioning accuracy of the improved algorithm is higher than that of the positioning algorithm based on mean weighting.
  • ZHOU Shengyuan,PAN Huilan
    Computer Engineering. 2017, 43(3): 110-114. https://doi.org/10.3969/j.issn.1000-3428.2017.03.020
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In order to improve convergence speed and tracking stability for high-order Quadrature Amplitude Modulation(QAM) demodulation carrier recovery,a new kind of Costas loop is designed based on software radio thought and the basic principle of traditional Costas loop. It is designed with sign phase detector instead of the traditional multiplier,and multi-factor adjustment replaces a set of coefficients in loop filter. Carrier recovery performance is verified by modeling and simulation using System Generator toolbox under Matlab/Simulink software for 1024QAM demodulation.Simulation result shows that the improved Costas loop can recover the carrier from received signal accurately for high-order QAM,with stable tracking,faster convergence,and lower bit error rate compared with traditional one,which improves the communication quality.
  • HUANG Shipan,ZHENG Lin,YANG Chao,JIANG Xiang,HUANG Zhichang
    Computer Engineering. 2017, 43(3): 115-119,125. https://doi.org/10.3969/j.issn.1000-3428.2017.03.021
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In the case of low Time-Bandwidth(TB) product,there are two problems in the Binary Orthogonal Keying(BOK) matching detection,the mutual interference between the positive and negative frequency modulation slope and the large amplitude of the self-matching sidelobe.In order to solve these problems,a new modulation method of delay offset Chirp-BOK is proposed.It reduces the mutual interference between the positive and negative frequency modulation slope signals by adding delay offset between the positive and negative frequency modulation slope symbols,so as to suppress the sidelobe and improve the Peak-to-Side-Lobes Ratio(PSLR).Through the engineering solution and simulation analysis for the best delay offset of this method,results show that,when compared with the commonly used side-lobe suppression method by the weighted window function,this method can dramatically improve the system detection performance without changing the structure of the Chirp matched filter and bandwidth under the condition of low TB product.
  • WU Fan,TAN Xiaobin,SHEN Jingjing,YAN Pei
    Computer Engineering. 2017, 43(3): 120-125. https://doi.org/10.3969/j.issn.1000-3428.2017.03.022
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Current positioning algorithms based on Radio Frequency Identification(RFID) location systems with dense passive tags face the issue of low localization precision due to the variation of behavior of tags,the ignorance of antenna directivity,external environment noise interference.To solve these problems and improve the positioning accuracy,the paper proposes a 2D passive RFID indoor positioning algorithm(IPABACM) by taking the directionality of the antenna into account in dense passive RFID tag distribution applications.Considering the effects of noise and tag performance differences in the production,the readied tags are assigned weights and trained through neural network.The deviation caused by antenna pattern’s uncertainty is eliminated by superimposing the antenna coverage area as a more regular pattern,which significantly improves the positioning accuracy.Experimental results show that compared with the traditional passive RFID localization algorithm,the IPABACM algorithm can provide relatively better accuracy and less positioning time.
  • SHEN Chaofan,HUANG Leijun
    Computer Engineering. 2017, 43(3): 126-133,139. https://doi.org/10.3969/j.issn.1000-3428.2017.03.023
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    With Molecular Sequence Compression(MSR) algorithm in the Redundancy Elimination(RE) technique as an example,this paper analyzes the impact of high Bit Error Rate(BER) and long latency of satellite links on the effectiveness and efficiency of the MSR algorithm,and proposes an acknowledgement-based history synchronization mechanism called Post-acknowledgement Synchronization(PAS).It combines PAS mechanism with MSR algorithm.Both the sender and receiver cache the historical packets that have been acknowledged by the receiver so that they have constantly synchronized history.Thus the receiver can decompress every packet that has been compressed by the sender,avoiding decompression failures.Experimental results show that MSR-PAS effectively suppresses RE-induced packet losses,while ensuring data compression effect and end to end delay stability.Accordingly,MSR-PAS is suitable for satellite IP networks.
  • WU Fu,ZHENG Lin,LI Xiaoji
    Computer Engineering. 2017, 43(3): 134-139. https://doi.org/10.3969/j.issn.1000-3428.2017.03.024
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    The traditional wireless communication networks with limited performance caused by its single structure cannot satisfy the quality requirement of emergency communication.To solve this problem,a muti-path Bandit algorithm based on the converged network,which consists of cognitive self-organization network and mobile cellular network,is proposed.The routing process in communication is divided into multi-slot path selecting stages.The trade-off between delay and energy efficiency for multi-path selecting is made to reasonably distribute energy consumption.Simulation results show that compared with the non-emergency business application and greedy algorithm,the network lifetime of muti-path Bandit algorithm are improved by 3%~20% under the converged network emergency business application.
  • FANG Kun,RU Le,YU Yunlong,JIA Xufeng,LIU Shuguang
    Computer Engineering. 2017, 43(3): 140-146. https://doi.org/10.3969/j.issn.1000-3428.2017.03.025
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    As the information transmission is sheltered easily by landform and terrain for the low-altitude Unmanned Aerial Vehicles(UAV) Ad Hoc networks,communication is interrupted frequent. Furthermore,the traditional congestion handling system cannot meet the quality requirement of UAV communication. In order to solve the above questions,a cross-layer method for reliable data transmission is proposed. The interruption probability of three communication mode is calculated based on channel conditions in the physical layer. The mechanism of the congestion degree division,according to the MAC layer information shared,based on the time that the packets enter in the buffering queue. The size of sliding windows is controlled by congestion in the transport layer,which can adaptively control the packet transmission rate of UAV. The best routing for reliable information transmission is selected by optimizing interruption probability and congestion degree in the network layer. Simulation results show,compared with Congestion Aware Routing protocol for Mobile(CARM) Ad Hoc network and Dynamic Congestion Aware Routing(DCAR) protocol,this method reduces interruption probability validly,and with the increase of information transmission rate and the speed of UAV,the low-altitude UAVs Ad Hoc networks get higher packet delivery ratio and lower average delay.
  • LIU Xinliang,DU Ruiying,CHEN Jing,WANG Chiheng,YAO Shixiong,CHEN Jiong
    Computer Engineering. 2017, 43(3): 147-153. https://doi.org/10.3969/j.issn.1000-3428.2017.03.026
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Through analyzing the specific implementation of the Security Socket Layer/Transport Layer Security(SSL/TLS) protocol in the client,this paper takes advantage of the vulnerabilities and flaws of SSL/TLS protocol session master key which is handled by browser and the secure parameters which are delivered in the process of protocol handshake. Combined with the Netfilter mechanism to hijack the session,it proposes a new security threat scheme for SSL/TLS protocol(SKAS). What’s more,it gives three defense methods of random number one-way encryption,bidirectional encryption and protection session master key security based on the security research of SKAS scheme. After the actual experiments,it is verified the feasibility of SKAS threat scheme. The success rate of attacks can reach more than 90% and this scheme also can achieve a wide attack and a deep threat. The three kinds of defense method can resist the SKAS threat and guarantee communication data security of SSL/TLS protocol between the client and server.
  • SUN He,WU Lifa,HONG Zheng,YAN Huiying,ZHANG Yafeng
    Computer Engineering. 2017, 43(3): 154-162. https://doi.org/10.3969/j.issn.1000-3428.2017.03.027
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Extracting a complete and accurate function call graph is the foundation of malware similarity analysis based on function call graph.This paper proposes a malware function call graph extraction method which integrates both dynamic and static analysis methods.It extracts executable path of malicious programs on the basis of static disassembly,and an active discovery strategy of hidden information is used to find out the hidden instructions and function calls in the malware.A dynamic feedback mechanism is used to ensure the information synchronization during the process of static and dynamic analysis.The experimental results show that the proposed method can deal with all kinds of reverse analysis technologies and extract a complete and accurate function call graph from malwares.
  • YU Qingying,LUO Yonglong,CHEN Fulong,ZHENG Xiaoyao
    Computer Engineering. 2017, 43(3): 163-171. https://doi.org/10.3969/j.issn.1000-3428.2017.03.028
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Foucusing on the issue that the existing spatial outlier detection methods fail to effectively solve the problem of guaranteeing both the data security and the validity of detection results at the same time,a privacy-preserving spatial outlier detection method is proposed,which uses statistical results of behavior attributes in spatial neighborhood and Mahalanobis distance to detect spatial outliers,and protects the privacy information by using the secure multi-party computation protocol based on semi-honest model,including secure distance computation,secure median computation of the combined vector and secure standardization protocols.Experimental results show that the method guarantees both the ability of privacy preserving and the effect of spatial outlier detection.
  • ZHAO Chen,YU Huifang,LI Jianmin
    Computer Engineering. 2017, 43(3): 172-175. https://doi.org/10.3969/j.issn.1000-3428.2017.03.029
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    At present,most group signature is only secure in the standard alone.Its security becomes weak in multiple protocol environment.In view of this,the security of group signature protocol is studied in the multiple concurrent execution under the universally composable framework.Since the protocol can be analyzed modularly under the universally composable framework,an ideal function of group signature protocol is defined in this framework.Then a group signature protocol is proposed to realize the ideal function,and its universally composable security as well as unforgeability is also proved.Furthermore,a concrete instance of this universally composable group signature protocol is given.Analysis results show that this protocol security is suitable for applications in multiple concurrent environment.
  • NI Lin,SHI Lei,HAN Kun,LI Shaoqing
    Computer Engineering. 2017, 43(3): 176-180. https://doi.org/10.3969/j.issn.1000-3428.2017.03.030
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    The introduction of uncontrolled third-party IP soft core introduced in the process of integrated circuit design can easily be embedded with hardware trojans.Traditional,functional testing method is difficult to achieve full coverage detection.So,the paper analyzes the structure of hardware trojans and its implementation characteristics in the IP soft cores,proposes a detection method of hardware Trojans based on feature matching,and presents a complete identification process.In addition,on the basis of building a feature library of hardware Trojans by Trust-Hub,three different types of hardware trojans are designed to be analyzed in the RTL description of AES algorithm.Experimental results show that the method can effectively detect hardware trojans with the help of the characteristic analysis of target carrier.It can realize accurate detection for hardware trojans under the reasonable selection principle when no more than 10% redundancy code is acquired.
  • MENG Kui,LIU Mengchi,HU Jie
    Computer Engineering. 2017, 43(3): 181-186. https://doi.org/10.3969/j.issn.1000-3428.2017.03.031
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Intention recognition methods,which are mainly based on feature template,have complicated hand-crafted feature extraction process and are difficult to capture semantic information of texts.Aiming at this problem,this paper proposes a new query intention identification model based on character level recurrent network.In order to effectively extract deep semantic features of a sentence and decrease long distance information dependent constraints,this paper uses Long Short -Term Memory Neural Network(LSTM) as a linear transformation of neural network layer,and uses a reverse LSTM layer to extract future information character.To avoid error propagation problem caused by inaccurate word segmentation results,it uses Chinese characters as inputs of the model,and uses distributed representation of characters to improve extractions of semantic features of sentences.Experimental results show that the method has an accuracy of 90.7%,which is higher than the characteristics template method,and it can improve the classification performance of user query intention.
  • ZHOU Xiong,WANG Lili
    Computer Engineering. 2017, 43(3): 187-192,199. https://doi.org/10.3969/j.issn.1000-3428.2017.03.032
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In the Web service decision making process,there are problems of new and inaccurate information,so this paper proposes a Service-oriented Architecture(SOA) system for Web service recommendation.Based on the characteristics of the multi-criteria decision analysis,it constructs the hierarchical SOA structure based service evaluation system.It also proposes multi-criteria similarity evaluation algorithm of weight time variation to improve Web service recommendation accuracy.Combined with the weight-time-varying process,the criteria weights for each period are further considered,which can improve the scientific decision making of Web services.Experiments on the Web service recommendation data set of Yahoo domain name verifies that the proposed algorithm has a higher Web service recommendation accuracy and lower sorting loss.
  • HUANG Wenming,MO Yang
    Computer Engineering. 2017, 43(3): 193-199. https://doi.org/10.3969/j.issn.1000-3428.2017.03.033
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In view of the drawback that the decision rules of classification regard for K Nearest Neighbor(KNN),the importance of every sample as the same,the classification results are not good.This paper proposes a method based on the text weighted KNN text classification algorithm and applies it to the classification of spam messages.After feature selection,considering the influence of frequency of feature words appearing in the text on text importance,the paper puts forward the first weighting formula.It uses association rule algorithm to mine frequent term sets from the spam message text and puts forward the second formula.Finally,it uses the two weighting formulas for the composite weighting on every message text so as to distinguish the influence of every training sample on category determination,thus improving on the classification decision rules.Experimental results show that the method has a promotion in accuracy,recall rate and F1 value which are important indexes compared with the un-improved KNN classification of spam filtering.
  • SHEN Yi,HU Yunshan,ZENG Guang,HAN Wenbao
    Computer Engineering. 2017, 43(3): 200-203,212. https://doi.org/10.3969/j.issn.1000-3428.2017.03.034
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    A single-radix Fast Fourier Transform(FFT) in-place permutation algorithm is proposed according to the original lookup table algorithm.A new lookup table configuration method is designed,and the computing method of original lookup table is optimized.The general law of selecting the optimal lookup table scale under different permutation calculation scale is obtained.Simulation results show that the scale of the lookup table is variable.The table can be searched by the cycle, the scale of the lookup table can be selected,the number of cycles can be reduced and the computing speed can be improved.
  • TIAN Xuedong,ZHANG Kaige,ZHOU Nan,ZHANG Zhiming,TIAN Bingjie
    Computer Engineering. 2017, 43(3): 204-212. https://doi.org/10.3969/j.issn.1000-3428.2017.03.035
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Aiming at the characteristics of mathematical expressions,such as the diversity of mathematical symbols,the complexity of structures and the richness of semantics,a mathematical expression retrieval results ranking algorithm is proposed.It utilizes the advantages of the hesitant fuzzy sets in dealing with the problems with multiple features and membership values to calculate the similarity between mathematical expressions,based on which the mathematical expressions are ranked.It measures the similarity of mathematical expressions through constructing the hesitant fuzzy membership with the multidimensional characteristics of the symbols,structures,grammar and semantics of mathematical expressions.Experimental results show that the algorithm can not only realize ordered output of search result data,but also improve the performance of mathematical expression retrieval system.
  • XIE Yangxiaojie,ZHAO Ling
    Computer Engineering. 2017, 43(3): 213-219,224. https://doi.org/10.3969/j.issn.1000-3428.2017.03.036
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In order to improve the sensitivity of Electrocardiogram(ECG) identification to noise and changes of signal,and to prioritize heartbeat features,a hybrid ECG human identification method is proposed.Heartbeat and appearance properties are jointly analyzed in the proposed method,in which Fisher linear discriminant analysis method is used to obtain the main attributes from these properties and the heartbeat features are calculated in Fisher space.The feature in time domain is used to calculate ECG characteristics.After the heartbeat feature in time domain is deducted by a series of heartbeat,the morphological characteristics are obtained by standardizing and scaling ECG signals.By maximizing the rate between the between-class individual hash and in-class individual hash,the most discriminative features are selected.Experimental results show that the recognition accuracy of the proposed method in a large database of healthy individuals is 99.24%.When the body is in the case of heart disorders,the Equal Error Rate(EER) is only 0.76%.In a mixed state,EER is 1.31%.So the robustness is well verified.
  • XIANG Bingwei,JING Xinxing,YANG Haiyan
    Computer Engineering. 2017, 43(3): 220-224. https://doi.org/10.3969/j.issn.1000-3428.2017.03.037
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Focusing on the issue that the robustness of the existing vehicular speech recognition system degrades drastically under practical application environments,a noise classification and compensation method based on Support Vector Machine(SVM) is proposed.Firstly,the noise of each application scene is collected to construct the SVM noise classifier which is used to classify the noise in the mute segment of the speech signal,and the corresponding noise training template is selected according to the noise type.The Delta-Spectral Cepstral Coefficients(DSCC) is used as the characteristic parameter,further suppresses the noise in the speech segment for vehicle speech recognition system.Experimental results show that the proposed method can effectively improve the noise robustness of vehicle speech recognition system and has higher speech recognition rate than sparse coded speech enhancement and PNCC feature enhancement methods.
  • LI Wei,XIAO Yanghua,WANG Wei
    Computer Engineering. 2017, 43(3): 225-231,240. https://doi.org/10.3969/j.issn.1000-3428.2017.03.038
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Classification is an important issue in constructing knowledge graph.However,the existing categories are edited by human beings for Chinese Baike websites,leading to increasing manpower cost and lost or error of categories.Aiming at this problem,this paper proposes a method to recognize all entities of Baidu Baike.under people domain automatically and add categories for these entities.Entity Set Expansion(ESE) is used for existing raw data of like categories,attributes and subtitles,and Markov Logic Network(MLN) is used to jointly inference the category of entities.Experimental result shows that,compared with Support Vector Machine(SVM) and Logistic Regression(LR) algorithm,the proposed method performs much better both in precision and recall aspects.
  • LIU Lu,ZHENG Liming
    Computer Engineering. 2017, 43(3): 232-240. https://doi.org/10.3969/j.issn.1000-3428.2017.03.039
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Considering that the range of neighborhood sizes and the type of mutation operators have huge effect on the exploration mode in the algorithm evolution process in Multi-objective Evolutionary Algorithm based on Decomposition(MOEA/D),this paper proposes an optimized MOEA/D algorithm.Four different neighborhood sizes and four mutation strategies with different features are combined in pairs as a part of candidate pool.In the evolutionary process,the combination with better performance is selected from the candidate pool with higher possibility according to the principle of negative feedback.Experimental results indicate that the proposed algorithm has strong robustness,and good diversity while ensuring convergence.
  • QIN Feng,WANG Heng,ZHENG Xiao,WANG Xiujun
    Computer Engineering. 2017, 43(3): 241-246,252. https://doi.org/10.3969/j.issn.1000-3428.2017.03.040
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Traditional sentiment analysis method only considers a single text,and the recognition rate of sentiment polarity is low when the micro-blog text is short and colloquial.In order to solve this problem,a kind of sentiment analysis method combined with linguistic context is proposed.Regarding micro-blog sentiment analysis problem as a tag sequence learning task,this paper uses Hidden Markov Support Vector Machine(SVMhmm) to blend the micro-blog context into micro-blog sentiment analysis.Experimental results show that this method can analyse the micro-blog sentiment polarity better than the Naive Bayes or Support Vector Machine(SVM) micro-blog sentiment analysis model.
  • YANG Weifang,LI Xueming,QIAO Baoxue
    Computer Engineering. 2017, 43(3): 247-252. https://doi.org/10.3969/j.issn.1000-3428.2017.03.041
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    For the problem of the classical Heat Conduction(HC) recommendation algorithm with low accuracy and the Mass Diffusion(MD) recommendation algorithm with low diversity,an improved hybrid recommendation algorithm based on HC and MD is proposed.Considering the influence of user activity on the recommendation algorithm,it introduces a tunable parameter θ to adjust the influence of user activity and reallocate resources so as to obtain better recommendation results.Experimental results show that the precision and diversity of the algorithm are increased by about 5.81% and 4.15% on the Netflix dataset,5.08% and 3.60% on the MovieLens dataset respectively,compared with the previous HC and MD hybrid algorithm,when parameter θ obtains optimal value.
  • LI Chuanpeng,QIN Pinle,ZHANG Jinjing
    Computer Engineering. 2017, 43(3): 253-260. https://doi.org/10.3969/j.issn.1000-3428.2017.03.042
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In order to remove the noise in the image more effectively,a method based on deep learning is proposed.The method uses the integrated image as the input and output of the network,and uses hidden layer to compose a nonlinear mapping from the noisy image to denoised image.The network has a symmetric network structure consisting of convolution subnet and deconvolution subnet.Convolution subnet learns about image features,and the deconvolution subnet recovers the original image on the basis of characteristic graph and obtains more texture details combining rectified linear unit.This method uses the VOC2012 data set as the training set and Tensorflow to train the network model in the GPU environment.Experimental result demonstrates that this method can remove the noise in the image more effectively and obtain a higher Peak Signal to Noise Ratio(PSNR).In addition,it takes shorter time,and has better visual effect and perfect practicability.
  • CHEN Pengxu,LIN Maosong,LIANG Yanyang,LIU Hongwei
    Computer Engineering. 2017, 43(3): 261-265,270. https://doi.org/10.3969/j.issn.1000-3428.2017.03.043
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    The actual wire suspended robot monitoring image is usually corrupted by the mixed noise in practical applications.This paper proposes a mixed noise image denoising algorithm for the mixed noise removal.The image consists of Additive White Gaussian Noise(AWGN) and Impluse Noise(IN).Aiming at pulse noise,two thresholds are proposed to detect noise.The image sparse representation and nonlocal similarity prior are integrated into the denoising model based on the existing weighted coding algorithm.Finally the denoising image is obtained.Experimental results show that the proposed algorithm under different noise ratio has good recovery performance,retains the texture details of the image,and has strong practicability.
  • SUN Juntao,ZHANG Shunli,ZHANG Li
    Computer Engineering. 2017, 43(3): 266-270. https://doi.org/10.3969/j.issn.1000-3428.2017.03.044
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In order to obtain the appearance model that contains both the distinction between object and background and the object characteritics,this paper presents a joint Support Vector Machine(SVM).Combining the characteristics of the one-class SVM and the binary SVM,it proposes an optimized object function and gives the solution steps by its dual form with Lagrange multiplier.Based on it,an object tracking algorithm is proposed.The robustness of the object appearance model is enhanced,and the ability to identify the object and background is improved.Experimental results on the public test videos show that the proposed algorithm can realize precise object tracking,and has good stability.
  • BAI Yunjiao,ZHANG Quan,SHANG Yu,ZHANG Pengcheng,LIU Yi,GUI Zhiguo
    Computer Engineering. 2017, 43(3): 271-276,281. https://doi.org/10.3969/j.issn.1000-3428.2017.03.045
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In order to preserve edge and remove noise during the image denosing,a model combining the patch similarity based Anisotropic Diffusion(AD) and the shock filter is proposed for image denoising and enhancement.The proposed model makes use of the patch similarity based AD model to remove noise and introduces the shock filter to enhance the important structure features of the image.The function with respect to the modulus of the image gradient is constructed to adaptively adjust the enhancement coefficients in the homogenous region,detail and edge region of the image.This function suppresses the noise amplification and the overshoot phenomenon when the proposed model enhances the details of the image.Experimental results show that the proposed model,which is superior to the traditional AD model,patch similarity based AD model and shock filter combined AD model in the aspects of visual effects and objective evaluation indexes,can not only effectively remove noise,but also better preserve the detail and edge features of the image.
  • NIE Jinyan,WEI Yantao,QU Shaocheng
    Computer Engineering. 2017, 43(3): 277-281. https://doi.org/10.3969/j.issn.1000-3428.2017.03.046
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In order to improve the accuracy and speed of object recognition,a template selection algorithm is presented.By using the labeling information of training sample images,this algorithm makes the different kinds of local neural response separate and the same ones come close.It not only reduces the redundancy of the template sets,but also obtains a small number of templates with strong discrimination ability.Experimental results show that,compared with the random selection algorithm,the proposed algorithm can significantly improve the efficiency of object recognition under the premise of ensuring accuracy.
  • ZHAO Yifeng
    Computer Engineering. 2017, 43(3): 282-288. https://doi.org/10.3969/j.issn.1000-3428.2017.03.047
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Concerning that the Gaussian Mixture Probability Hypothesis Density(GM-PHD) filter does not check one-to-one assumption and it is difficult to track crossing targets,an improved multi-target tracking method with weight constraint is proposed based on GM-PHD filter.Firstly,the weight matrix is constructed and the target with the maximum weight is searched from the generated targets.Then,according to weight relationship,targets in all rows are re-normalized except that with the maximum weight.Normalization and weight restriction are operated iteratively.Finally,the target weight is generated in the update steps of GM-PHD filter for complete filtering operation.The method is evaluated through Monte Carlo simulation,and the performance of filters are tested on the conditions of clutter,different target speeds and different frame rates,with simulations of crossing and dense targets.Experimental results shows that,compared with the methods using GM-PHD filter and Sequential Monte Carlo PHD (SMC-PHD) filter,the proposed method has better overall performance.
  • WU Zetai,CAI Renqin,XU Shuyan,WU Xiaosi,FU Yuli
    Computer Engineering. 2017, 43(3): 289-293. https://doi.org/10.3969/j.issn.1000-3428.2017.03.048
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In this paper,based on fingerprinting,the application of K Nearest Neighbor(KNN) method in indoor positioning is researched.In order to improve the positioning accuracy,this paper puts forward a new formula for calculating the similarity.Aiming at the problem of large amounts of computation for KNN method,it combines the clustering algorithm and KNN method,and proposes a new positioning algotithm.Experimental results show that,compared with the KNN,the proposed algorithm has comparable accuracy,and it significantly reduces the positioning time,which can satisfy the requirements of indoor and outdoor positioning.
  • LI Qiang,WANG Mei,LIU Zhenghong
    Computer Engineering. 2017, 43(3): 294-298. https://doi.org/10.3969/j.issn.1000-3428.2017.03.049
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In order to improve the precision of Received Signal Strength Indication(RSSI) indoor positioning,this paper puts forward an indoor RSSI positioning method based on reference tag and rotating Radio Frequency Identification(RFID) reader.In this method,tag is detected repeatedly for reconstructing reader’s reading area by manually rotating the reader.The coarse positioning points can be obtained according to two target reading area nodes.In order to get the accurate positioning value,these coarse positioning points are processed by optimization analysis or clustering analysis algorithm.Experimental results show that the positioning error of this method is as low as 4 cm,9 cm and 18 cm in indoor environment of 2.8 m×2.8.m,5.6 m×5.6 m and 8.4 m×8.4 m respectively,which meets the positioning requirements of the indoor scene like warehouse,supermarket and so on.
  • YAN Haizhou,XU Bugong,SHI Dongjiang,ZHENG Weide
    Computer Engineering. 2017, 43(3): 299-303. https://doi.org/10.3969/j.issn.1000-3428.2017.03.050
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    LZW algorithm can not make full use of the length of dictionary for prefix encoding when compressing data and has redundancy.When considering the length of the dictionary is l,the prefix can be located within [0,l-1] only,and the LZW algorithm uses 「lbl bit to encode the prefix.Therefore,the data of the interval [l,2「lbl-1] is not fully used.Aiming at this problem,this paper proposes an improved compression algorithm of prefix mapping encoding.It does not encode and output the prefix which meets the condition,but maps it to the interval [l,2「lbl-1].In this case,the number of prefix encoding is not decreased but implies the information that the next prefix will be encoded with less bits.Compared with LZW agorithm,the improved algorithm has advantages that it does not increase the amount of calculation and memory space,and can be used more generally in data compression.The effectiveness of the improved algorithm is verified when tested with English text data in simulation experiments and applied into residual data compression.
  • ZHOU Wenyi,WANG Jiyuan
    Computer Engineering. 2017, 43(3): 304-308,315. https://doi.org/10.3969/j.issn.1000-3428.2017.03.051
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    To solve the over-fitting problem of Random Forest(RF) learning method for training data,a Fuzzy Forest(FF) learning method is proposed by improving the decision function of every decision node.Firstly,it builds decision function of every decision node on decision tree by using Gauss membership functions to convert the clear decision paths to fuzzy ones.Then,it produces fuzzy path with the product of the fuzzy decision values of all decision nodes which the sample goes through from root node to leaf node.Finally,it computes the prediction result according to each fuzzy path and prediction parameter of corresponding leaf node.The new FF learning method is applied in the field of pedestrian detection for learning and classifying both Haar and Histogram of Oriented Gradient(HOG) features.Experimental results show that FF is better than classical classifiers such as Adaboost,Support Vector Machine(SVM) and RF for improving the recognition rate of pedestrian detection.
  • ZHANG Tiantian,ZHANG Ning,MAO Guoyong
    Computer Engineering. 2017, 43(3): 309-315. https://doi.org/10.3969/j.issn.1000-3428.2017.03.052
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    In order to research the influence of individual characteristic on information spreading in online social network,this paper presents a simple information spreading model that combines individual characteristic,social reinforcement and information attribute and considers the memory effects,social reinforcement,information attribute,and individual characteristic.It also considers the ability to spread information of the individual that obeys uniform distribution and power-law distribution in the regular network,the WS small network and the BA scale-free network,and carries out simulation experiment.At the same time,the empirical research is performed in the real network Facebook.No matter what kind of situation,transmission speed of information is fastest in the BA scale-free network,and slowest in the regular network.The stable time of the regular network is the longest.The spread figure of Facebook is similar to the results of the BA network.
  • ZHANG Wen,YANG Zhijiang,HU Zhenghua,MENG Yizhuo
    Computer Engineering. 2017, 43(3): 316-321. https://doi.org/10.3969/j.issn.1000-3428.2017.03.053
    Abstract ( ) Download PDF ( )   Knowledge map   Save
    Downloading the Digital Elevation Model(DEM) data of big band object through the traditional bounding box method has problems of low efficiency and large amounts of data download.In order to solve these problems,this paper proposes two algorithms to obtain DEM data along long route with spatial analysis and rectangular stitching.One is smooth buffer DEM clipping and extraction algorithm based on spatial analysis,and the other is DEM clipping and stitching algorithm based on rectangular stitching.The data of electric power transmission lines are used to test the algorithms,and the results show that the proposed algorithms have great improvement in both efficiency and effectiveness,compared with traditional bounding box method.