Current Issue

25 November 2017, Volume 38 Issue Z2

Previous Issue    Next Issue

Papers

Multi relay node selection strategy for energy capture collaborative relay network

Xian-zhong TIAN,Min GUO,Jia-cun HE,Gao LIU,Yi-nan ZHU

2017, 38(Z2):  1-7.  doi:10.11959/j.issn.1000-436x.2017276

Asbtract ( 358 )   HTML ( 5)   PDF (750KB) ( 578 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

An energy capture cooperative relay network was studied.Unlike traditional cooperative relay networks,the relay nodes generally did not have a fixed energy supply but could carry out energy capture.For this network,a natural question was how to determine whether the relay nodes perform energy capture or data forwarding,and how to select the relay node to assist the information transmission between the source node and the destination node.Based on this problem,a multi-relay node selection scheme was proposed,which select the operation according to the energy of the current time slot of the relay node.In each time slot,the relay node that meet the battery threshold requirements was selected to assist in transmission.The Markov chain was used to model the charging and discharging process of the relay node battery,and the probability of interruption and throughput of the wireless network was obtained.And the value of the battery threshold when the throughput was maximized,that is,when the probability of network breakage was minimum was solved.Finally,the advantages of this scheme were verified by numerical experiments.

Design and implementation of a layer-II fast handoff protocol in emergency communication system

Fang-min LI,Le ZENG,Yi SHEN,Tao ZHANG

2017, 38(Z2):  8-16.  doi:10.11959/j.issn.1000-436x.2017260

Asbtract ( 235 )   HTML ( 7)   PDF (1193KB) ( 625 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Immobility,limited bandwidth and poor support to real-time service are common shortcomings in traditional emergency communication system.Moreover,the data transmission speed is highly limited in high-speed services by the long handoff delay and fixed handoff trigger threshold when the mobile terminals moving between network nodes.In order to overcome these shortcomings,a new emergency communication system with multifrequency hierarchical clustering structure was designed.A new layer-II fast handoff protocol was designed for such network structure to deal with handoff issue in real-time services.In this handoff protocol,the handoff trigger threshold is self-adaption based on the moving velocity of mobile terminals.

Deep belief network-based link quality prediction for wireless sensor network

Lin-lan LIU,Jiang-bo XU,Yue LI,Zhi-yong YANG

2017, 38(Z2):  17-25.  doi:10.11959/j.issn.1000-436x.2017257

Asbtract ( 337 )   HTML ( 12)   PDF (1041KB) ( 708 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

After analyzing the existing link quality prediction models,a link quality prediction model for wireless sensor network was proposed,which was based on deep belief network.Support vector classification was employed to estimate link quality,so as to get link quality levels.Deep belief network was applied in extracting the features of link quality,and softmax was taken to predict the next time link quality.In different scenarios,compared with the model of link quality prediction based on logistic regression,BP neural network and Bayesian network methods,the experimental results show that the proposed prediction model achieves better precision.

Keystroke features recognition based on stable linear discriminant analysis

Wei-guo SHEN,Wei WANG

2017, 38(Z2):  26-29.  doi:10.11959/j.issn.1000-436x.2017272

Asbtract ( 268 )   HTML ( 13)   PDF (714KB) ( 379 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

A novel keystroke features recognition method based on stable linear discriminant Analysis (SLDA) was put forward.First of all,it maximum the dispersion between different sequences,while minimizing the dispersion between the same sequence set,maintain the best discriminant characteristics of the keystroke sequences.Secondly,the local similarity graph between keystroke sequences is constructed,minimizing the dispersion of the local similarity sequences,to keep the local similarity of keystroke sequences.Finally,based on the principles above,the feature of keystroke sequences are extracted,and the nearest neighbor classification criterion is used to judge the outputs.The effectiveness of the proposed method is certified by experiment results.

Android malware detection method based on SimHash

Bo CHEN,Yong-tao PAN,Tie-ming CHEN

2017, 38(Z2):  30-36.  doi:10.11959/j.issn.1000-436x.2017271

Asbtract ( 973 )   HTML ( 20)   PDF (832KB) ( 716 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

A new similarity detection scheme based on hierarchical SimHash algorithm was proposed.The scheme extractd contents from different aspects to represent the APK file,then used the improved SimHash to respectively represent the file.The scheme analyzed the APK file by extracting the AndroidManifest.xml file in it,the sum of the Smali code from the decompilation of dex file,instructions extracted in Smali files,Java code set,and instructions extracted in Java code files.Through the study of Voted Perceptron voting algorithm,the scheme used trust weight method,by valuating a trust weight in every layer,then combined all the result with weight in every layer as a resule of scheme,the result can be more reasonable and more convincing.

Experiment based on orbit satellite and application for sparse microwave imaging

Jing TIAN,Li-hong KANG,Xi-long SUN,Zhou LI,Bing-chen ZHANG

2017, 38(Z2):  37-42.  doi:10.11959/j.issn.1000-436x.2017263

Asbtract ( 358 )   HTML ( 9)   PDF (1598KB) ( 637 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

The experiment aiming to verify the sparse microwave image methodology based on orbit SAR satellite was presented.During the experiment,the imaging parameters were changed in order to achieve the uniform de-sampling echo data.The image quality evaluation method for sparse microwave imaging was proposed and quantified evaluation results for the experiment were also presented.Finally,the remote applications of sparse microwave imaging were prospected.

Leakage-resilient certificateless signcryption scheme

Yan-lin QIN,Xiao-ping WU,Wei HU

2017, 38(Z2):  43-50.  doi:10.11959/j.issn.1000-436x.2017267

Asbtract ( 294 )   HTML ( 13)   PDF (770KB) ( 375 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

In practical applications,the potential adversary may exploit partial information about the secret keys by side-channel attacks ,traditional certificateless signcryption schemes can’t resist these key-leakage attacks.A leakage-resilient certificateless signcryption scheme based on Elliptic Curve Cryptography and bilinear pairing was presented.In the random oracle,proved that the security of the scheme is based on the decisional Diffie-Hellman assumption.The scheme is also proved semantically secure against adaptive posterior chosen-ciphertext key-leakage attacks (KL-CCA2),and existentially unforgeable against chosen-message key-leakage attacks(KL-CMA).The proposed scheme was free from non-interactive zero knowledge proof system and needs no bilinear paring operation in signcryption phase.Compared with other schemes of the same kind,the proposed scheme can resist key-leakage attacks and maintains high efficiency.

Simulation and analysis of software defined cognitive frequency hopping multi beam satellite system

Heng-zhi LI,Chun-feng WANG

2017, 38(Z2):  51-55.  doi:10.11959/j.issn.1000-436x.2017273

Asbtract ( 373 )   HTML ( 9)   PDF (1407KB) ( 580 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Traditional multi-beam satellites cannot adapt to the changing traffic because of the fixed bandwidth and cannot adjust dynamically for the traffic load.In the future,the multi-beam broadband satellite system must have great flexibility and can be dynamically adjusted according to the change of the traffic flow.Beam hopping for multi-beam broadband satellite network system has been proposed to improve the traditional multi-beam broadband satellite network system performance,whose principle is to assign different beams to different time slots,rather than the allocation of bandwidth,so that in each time slot,the entire available bandwidth is allocated to each beam,and the time window is periodically applied to the beam selection system ,in the full band hopping mode,the band can be selected optimally in the duration of each beam to meet user transmission bandwidth and delay requirements.With the development of cognitive radio technology,cognitive beam hopping for multi-beam satellite system can further improve the efficiency of spectrum utilization,to meet the growing shortage of spectrum resources,to achieve large-capacity high-bandwidth broadband satellite network system,and to support the rising user service capacity and the transmission efficiency of user services under different load conditions.In addition,software definition technology is also used in satellite networks,which could achieve effective management of satellite resources to improve the utilization of satellite resources.In this paper,the principle of software-defined cognitive hopping multi-beam broadband satellite network system is discussed.The spectrum utilization efficiency and throughput are analyzed,which can be used as a reference for broadband satellite network system design.

Design and implementation of multi-tenant oriented lightweight virtual HPC cluster based on supercomputing environment

Yu-song TAN,Rong-zhen LI,Qing-bo WU,Jian-feng ZHANG,Yao-xue ZHANG

2017, 38(Z2):  56-66.  doi:10.11959/j.issn.1000-436x.2017279

Asbtract ( 427 )   HTML ( 13)   PDF (899KB) ( 1075 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

In order to deliver on-demand HPC services for multi-tenants and solve software dependency problems which rigidly restrict the usage of HPC applications.On the premise of not destroying the existing large scale high performance cluster management,a lightweight HPC cluster delivery model was designed and implement with the Fat-Tree network topology and the virtual HPC cluster model.So as to supply a cloud based HPC rental service for the high performance applications of the multi-tenants and solve the rigid dependable problem of HPC application software stack for more HPC tenants in an easy way,the experimental results show that the proposed method is similar to the performance of the physical nodes,so that the HPC has the on-demand characteristics of the cloud and has a better embedding efficiency.

Trusted solution monitoring system based on ZigBee wireless sensor network

Tian-shu WANG,Gong-xuan ZHANG,Xi-chen YANG

2017, 38(Z2):  67-77.  doi:10.11959/j.issn.1000-436x.2017258

Asbtract ( 264 )   HTML ( 5)   PDF (2136KB) ( 653 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

In the greenhouse environment,the concentration of nutrient solution was always ignored.At the same time,the core node coordinator of the ZigBee wireless monitoring system was easily compromised.A trusted solution monitoring system based on ZigBee wireless sensor network was designed and implemented.Firstly the hardware structure of sensor,router and coordinator was introduced.Then the processes of external voltage measurement and data cleaning were illustrated.For realizing the trusted boot of the coordinator,trusted u-boot to measure the integrity of some key files was designed.Experimental results show that the solution concentration data could be sent from each node to the monitoring computer and the remote terminal.At the same time,the trusted module of the coordinator can detect the captured coordinator node and warning.Therefore,the designed system has strong function and security.

Research on space information representation model based on formal concept analysis

Shao-bo YU,Ling-da WU,Xi-tao ZHANG

2017, 38(Z2):  78-85.  doi:10.11959/j.issn.1000-436x.2017278

Asbtract ( 331 )   HTML ( 9)   PDF (1869KB) ( 708 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

By discussing the concept,connotation and development situation of SI and its representation,and combining the content of FCA,the SIR model was established from there aspects including SI formal context,SI concept lattice and SI implication relation.Taking part of space entity sets and space data sets as input,the specific application of representation model on SIN was realized based on concept explorer platform,meanwhile,the effectiveness and feasibility of the SIR model was also validated by this example.Of course,the aim of the model was to provide information representation support for the basic theory and key technical research of SIN.

Homology analysis of malware based on graph

Bing-lin ZHAO,Xi MENG,Jin HAN,Jing WANG,Fu-dong LIU

2017, 38(Z2):  86-93.  doi:10.11959/j.issn.1000-436x.2017259

Asbtract ( 929 )   HTML ( 89)   PDF (1174KB) ( 2081 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Malware detection and homology analysis has been the hotspot of malware analysis.API call graph of malware can represent the behavior of it.Because of the subgraph isomorphism algorithm has high complexity,the analysis of malware based on the graph structure with low efficiency.Therefore,this studies a homology analysis method of API graph of malware that use convolutional neural network.By selecting the key nodes,and construct neighborhood receptive field,the convolution neural network can handle graph structure data.Experimental results on 8 real-world malware family,shows that the accuracy rate of homology malware analysis achieves 93%,and the accuracy rate of the detection of malicious code to 96%.

Fault analysis and research of wireless sensor network based on kernel partial least squares

Guang-hai ZHOU,Zhao-long NING,Zhi-kui CHEN,Hua ZHONG,Yue-ming HU

2017, 38(Z2):  94-98.  doi:10.11959/j.issn.1000-436x.2017265

Asbtract ( 346 )   HTML ( 9)   PDF (724KB) ( 451 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

With the development of intelligent and networked sensor technology,wireless sensor networks were widely used in human life and commercial fields,because wireless sensor network nodes usually only carry limited resources,it is prone to failures due to insufficient resources,the accurate and timely fault diagnosis of WSN nodes can ensure the reliability of information,thus improving the maintainability of WSN and prolonging the service life of WSN.A method of using kernel partial least squares has been proposed to predict the fault reasons,the method overcomes the defects of traditional linear regression method and the nonlinear high dimensional space for data analysis.Through many experiments,the method can absorb the characteristics of canonical correlation analysis and principal component analysis method,provide a more thorough and rich content analysi,that the reason of the fault can be predicted effectively.

Big data based metro crowd delivery system

Kun-fang ZHANG,Ming-ming LU,Lin ZHENG

2017, 38(Z2):  99-112.  doi:10.11959/j.issn.1000-436x.2017256

Asbtract ( 453 )   HTML ( 10)   PDF (1471KB) ( 651 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

In recent years,the demand for urban express delivery experiences fast growing.Several startup delivery corporations,such as iShansong,have provided the delivery services,which are promised to deliver a parcel within 1 or 3 hours in a city.However,since the quality of the promised services have not been fully quantitatively analyzed based on real data,it is quite often that couriers refuse to pick up delivery requests or the accepted parcels fail to be delivered on time.To address the above issue,a metro crowd-delivery system was proposed,which could utilize the historical records of metro passengers to analyze the quality of the delivery service and provide differential service within different period of time.The system not only meet the requirements of high delay requirements of the courier,but also as a city-wide express supplement.At the same time,a courier transit program was proposed.Experiments show that this transfer program is efficient.

Social network information diffusion model based on user’s influence and interesting

Rui WANG,Yong LIU,Jing-hua ZHU,Ping XUAN,Jin-bao LI

2017, 38(Z2):  113-121.  doi:10.11959/j.issn.1000-436x.2017264

Asbtract ( 308 )   HTML ( 10)   PDF (774KB) ( 532 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

A new non-topological information diffusion model of social network was proposed,called non-topological influence-interest diffusion model (NT-II).Representation learning was exploited to construct two hidden spaces for NT-II，called the user-influence space and the user-interest space,each user and each propagation item was mapped into a vector in space.The model predicted the probability of a user receiving a propagated item,considering not only the degree of influence from other users,but also the user's preference for propagated item.The experimental results show that the model can simulate the propagation process and predict the propagation results more accurately.

Inception neural network for human activity recognition using wearable sensor

Duo CHAI,Cheng XU,Jie HE,Shao-yang ZHANG,Shi-hong DUAN,Yue QI

2017, 38(Z2):  122-128.  doi:10.11959/j.issn.1000-436x.2017262

Asbtract ( 418 )   HTML ( 20)   PDF (794KB) ( 1048 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

The experience from computer vision was learned,an innovative neural network model called InnoHAR (inception neural network for human activity recognition) based on the inception neural network and recurrent neural network was put forward,which started from an end-to-end multi-channel sensor waveform data,followed by the 1×1 convolution for better combination of the multi-channel data,and the various scales of convolution to extract the waveform characteristics of different scales,the max-pooling layer to prevent the disturbance of tiny noise causing false positives,combined with the feature of GRU helped to time-sequential modeling,made full use of the characteristics of data classification task.Compared with the state-of-the-art neural network model,the InnoHAR model has a promotion of 3% in the recognition accuracy,which has reached the state-of-the-art on the dataset we used,at the same time it still can guarantee the real-time prediction of low-power embedded platform,also with more space for future exploration.

Intelligent forecasting and monitoring of air index based on big data and internet of things

Yan LIU,Yong-ping ZHANG,Cheng ZHU,Jun GAO,Qi-ming LIU

2017, 38(Z2):  129-138.  doi:10.11959/j.issn.1000-436x.2017266

Asbtract ( 595 )   HTML ( 54)   PDF (2435KB) ( 1505 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Air quality forecast has become an urgent need.However,numerical forecast of air quality is a complex systems engineering.The intelligent forecasting of air index was studied from the perspective of big data and intelligent decision making.For the index prediction of air quality,the popular classification algorithm was introduced to realize the intelligent analysis of historical data.To obtain air quality information in real time,a monitoring system based on Internet of Things was established,and intelligent processing of real-time data collected by the classification algorithm was achieved.Due to the large amount of historical air index data and real-time data collected,the technology of cloud computing and big data was introduced to speed up the data processing and improve the storage of data.In addition,the client based on Android was developed to allow users to query the air quality anytime,anywhere.

Design of tree topology based Byzantine fault tolerance system

Wei-dong LYU,Xue-guang ZHOU,Zhi-min YUAN

2017, 38(Z2):  139-146.  doi:10.11959/j.issn.1000-436x.2017261

Asbtract ( 456 )   HTML ( 20)   PDF (701KB) ( 747 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

A tree topology based Byzantine fault tolerance system was designed and the calculation of the Byzantine node bound based on depth and width was given.The consistency protocol was designed through divide the tree into groups by communication and work,which guaranteed the safety of system.The access of a node was determined with joint signature mechanism.The Byzantine nodes were placed to the leaves by view change protocol,which guaranteed the liveness of the system.On the foundation of traditional protocol,the new protocol support multiple tasks operating at the same time,and communication between nodes is largely decreased,as well as the space complexity,hence the feasibility being promoted,and the gap between Byzantine fault tolerance system and the non-Byzantine systems being narrowed.

Optimization strategy of virtual machine online migration with awareness of application characteristics and network bandwidth migration

Xiang LI,Ning-jiang CHEN,Shang-lin YANG,Hua LI

2017, 38(Z2):  147-155.  doi:10.11959/j.issn.1000-436x.2017268

Asbtract ( 249 )   HTML ( 4)   PDF (963KB) ( 468 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Firstly the experiments to verify the relationship between the number of dirty memory pages and application characteristics which exist in virtual machine migration was conducted.Then,different virtual machine application characteristics were perceived,with which the number of dirty memory pages produced during the migrations was predicted by the use of GM(1,N) grey prediction model.At the same time,using residual correction to adjust error makes results more reliable.According to the prediction of memory dirty pages,network bandwidth was adjusted and reserved.Compared with the traditional pre-copy strategy,the given experiments show that the optimized strategy proposed can improve the performance of network and reduce migratory cost for the memory-intensive and network-intensive applications.

Secure deduplication and integrity audit system based on convergent encryption for cloud storage

Xiao-yong GUO,An-min FU,Bo-yu KUANG,Wei-jia DING

2017, 38(Z2):  156-163.  doi:10.11959/j.issn.1000-436x.2017269

Asbtract ( 561 )   HTML ( 27)   PDF (1363KB) ( 1185 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Cloud storage applications quickly become the best choice of the personal user and enterprise storage with its convenience,scalability and other advantages,secure deduplication and integrity auditing are key issues for cloud storage.At first,convergent key encapsulation/decoupling algorithm based on blind signature was set up,which could securely store key and enable it to deduplicate.Besides,a BLS signature algorithm based on convergence key was provided and use TTP to store public key and proxy audit which enables signature and pubic key deduplication and reduces client storage and computing overhead.Finally,cloud-based secure deduplicaion and integrity audit system was designed and implemented.It offered user with data privacy protection,deduplication authentication,audit authentication services and lowered client and cloud computation overhead.

Improved LEACH protocol algorithm for WSN based on energy balance and high efficiency

Li-xiao HUANG,Hui WANG,Li-yong YUAN,Ling-guo ZENG

2017, 38(Z2):  164-169.  doi:10.11959/j.issn.1000-436x.2017270

Asbtract ( 584 )   HTML ( 57)   PDF (747KB) ( 675 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

The improved energy-saving algorithm for the energy efficient and balanced protocol (LEACH-improved) was proposed.In the algorithm,the residual energy factor,the distance factor and the node density factor was added to improve the threshold formula.Moreover,in the various application scenarios of WSN,based on the different effects on the selection of cluster heads caused by the residual energy and geographical location,the weighting factor was introduced.Results of simulation experiments show that the improved algorithm can effectively reduce the network energy consumption and prolong the network lifetime.

Link optimization strategy based on spectral moment in wireless sensor network

Qing-song CAI,Xi-hou CHEN

2017, 38(Z2):  170-177.  doi:10.11959/j.issn.1000-436x.2017280

Asbtract ( 234 )   HTML ( 2)   PDF (1039KB) ( 384 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

In order to prolong the life of the wireless sensor network (WSN),an adaptive local communication link optimization algorithm of nodes was proposed to optimize the network structure dynamically.By introducing the term of Laplacian spectrum moment,each node can make the decision of adding or deleting a link iteratively with their neighborsusinglimited local network structural information,and then the whole network can evolve to a predefined structure.The experimental results show that each node takes only finite iterations and then the network structure can converge quickly to the target,which prove the effectiveness of the algorithm.

Improvement Differential fault attack on TWINE

Yang GAO,Yong-juan WANG,Lei WANG,Tao WANG

2017, 38(Z2):  178-184.  doi:10.11959/j.issn.1000-436x.2017274

Asbtract ( 326 )   HTML ( 20)   PDF (778KB) ( 513 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

A new method of differential fault attack was proposed,which was based on the nibble-group differential diffusion property of the lightweight block cipher TWINE.On the basis of the statistical regularity of the S-box differential distribution,the lower bound of the probability of recovering round key was calculated.Then expectation of number of fault injections when restoring seed key can be estimated.Theoretical proof and experimental results both show that an average of nine times of fault injections in 33,34 and 35 rounds bring about the seed key recovered completely.Finally,the improvement of the fault injection location was proposed,which enhances the feasibility of the genuine attack.

Comprehensive Reviews

Research on vulnerability and security technology of industrial Ethernet protocol

Tao FENG,Ye LU,Jun-li FANG

2017, 38(Z2):  185-196.  doi:10.11959/j.issn.1000-436x.2017277

Asbtract ( 606 )   HTML ( 36)   PDF (1338KB) ( 2018 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

To solve the information security of industrial control system,the safety of industrial Ethernet protocol was studied deeply,the research status of protocol security protection was reported and the vulnerability of the five-major protocol was analyzed.Firstly,the architecture of industrial control system and industrial Ethernet protocol was discussed.Secondly,from the three aspects of external active defense technology,internal passive defense technology and protocol security improvement,a perfect industrial Ethernet protocol security protection model was put forward,and the main protection technology was discussed.Finally,the future development direction and research ideas of information security improvement of industrial Ethernet protocol were pointed out.

Overview of detection techniques for malicious social bots

Rong LIU,Bo CHEN,Ling YU,Ya-shang LIU,Si-yuan CHEN

2017, 38(Z2):  197-210.  doi:10.11959/j.issn.1000-436x.2017275

Asbtract ( 750 )   HTML ( 79)   PDF (817KB) ( 2452 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

The attackers use social bots to steal people’s privacy,propagate fraud messages and influent public opinions,which has brought a great threat for personal privacy security,social public security and even the security of the nation.The attackers are also introducing new techniques to carry out anti-detection.The detection of malicious social bots has become one of the most important problems in the research of online social network security and it is also a difficult problem.Firstly,development and application of social bots was reviewed and then a formulation description for the problem of detecting malicious social bots was made.Besides,main challenges in the detection of malicious social bots were analyzed.As for how to choose features for the detection,the development of choosing features that from static user features to dynamic propagation features and to relationship and evolution features were classified.As for choosing which method,approaches from the previous research based on features,machine learning,graph and crowd sourcing were summarized.Also,the limitation of these methods in detection accuracy,computation cost and so on was dissected.At last,a framework based on parallelizing machine learning methods to detect malicious social bots was proposed.