Current Issue

25 February 2021, Volume 42 Issue 2

Previous Issue    Next Issue

Papers

Privacy protection scheme of DBSCAN clustering based on homomorphic encryption

Chunfu JIA, Ruiqi LI, Yafei WANG

2021, 42(2):  1-11.  doi:10.11959/j.issn.1000-436x.2021026

Asbtract ( 612 )   HTML ( 159)   PDF (1284KB) ( 783 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

In order to reduce the risk of data privacy leakage in the process of outsourced clustering, a privacy protection scheme of DBSCAN clustering based on homomorphic encryption was proposed.In order to encrypt the float data in the actual scene, three data preprocessing methods for different data accuracy were given, and a policy for choosing a proper data preprocessing method based on data characteristics, accuracy and computational cost was also proposed.For the ciphertext comparison operation that was not supported by homomorphic encryption, a protocol between the client and the cloud server was designed to realize the function of ciphertext comparison.Theoretical analysis and experimental results show that the proposed scheme can ensure the security of data privacy, and has a higher clustering accuracy rate and lower time overhead.

Secure and efficient two-party ECDSA signature scheme

Jing WANG, Libing WU, Min LUO, Debiao HE

2021, 42(2):  12-25.  doi:10.11959/j.issn.1000-436x.2021019

Asbtract ( 811 )   HTML ( 113)   PDF (1652KB) ( 1335 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

To solve the easy disclosure of signature private key and excessive concentration of signature rights, a secure and efficient two-party ECDSA signature scheme was proposed for the blockchain based network trading systems.By pre-computing one-time pad Beaver’s triple, and utilizing the Beaver’s triple based secure two-party multiplication technology, some computationally intensive homomorphic encryption operations and oblivious transfer operations with high communication overhead were effectively avoided, and thereby an efficient two-party ECDSA signing was realized, which could ensure that the two signing parties output valid ECDSA signature without reconstructing the complete private key.The proposed scheme was proved to be provably secure under the hybrid model of the universally composable framework.Theoretical analysis and simulation results demonstrate that the proposed scheme has significant advantages in terms of signing efficiency and bandwidth requirements when compared with the existing two two-party ECDSA signature schemes.

V2X collaborative caching and resource allocation in MEC-based IoV

Fangwei LI, Haibo ZHANG, Zixin WANG

2021, 42(2):  26-36.  doi:10.11959/j.issn.1000-436x.2021007

Asbtract ( 563 )   HTML ( 114)   PDF (898KB) ( 752 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Aiming at the problem that with the rapidly increasing of multimedia services in IoV, a large amount of data change has bought a heavy burden on the mobile networks, a V2X collaborative caching and resource allocation framework in MEC-based IoV was constructed.A V2X cooperative caching and resource allocation mechanism was proposed to achieve the effective allocation of computing resources, caching resources, and communication resources in the network.The graph coloring model was used to allocate channels to the offloading users.Lagrange multiplier method was used to allocate power and computing resources.The simulation results show that the proposed mechanism can effectively reduce system overhead and reduce task completion delay under different system parameters.

Research on forecast and recommendation technology of taxi passengers based on time-varying Markov decision process

Tong WANG, Shan GAO, Huiwen GONG, Bo SUN

2021, 42(2):  37-51.  doi:10.11959/j.issn.1000-436x.2021002

Asbtract ( 308 )   HTML ( 50)   PDF (2611KB) ( 532 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

To solve the problems of unloading rate caused by blind passenger search of taxis, the hotspot recommendation strategy of taxi passengers was proposed.The proposed strategy could optimize the process of matching passengers to the greatest extent to increase the efficiency of passenger search.Based on the historical trajectory data of taxis and the time series characteristics of hotspot passenger information, a segment prediction method was proposed based on recurrent neural network (SPBR) and a passenger recommendation model was proposed based on time-varying Markov decision process (TMDP).Experimental results show that the RMSE predicted by SPBR algorithm is 67.6%, 71.1% and 64.5% lower than the SVR, CART and BPNN algorithms.The expected return of taxis based on the TMDP algorithm is 35.9% higher than historical expectations.

Distributed data trading algorithm based on multi-objective utility optimization

Xiaohong HUANG, Yong ZHANG, Desheng SHAN, Yekui QIAN, Lu HAN, Dandan LI, Qun CONG

2021, 42(2):  52-63.  doi:10.11959/j.issn.1000-436x.2021034

Asbtract ( 455 )   HTML ( 82)   PDF (1158KB) ( 507 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

The traditional centralized data trading models are not well applicable to the current intelligent era where everything is interconnected and real-time data is generated, and in order to maximize the use of collected data, it is essential to design an effective data trading framework.Therefore, a distributed data trading framework based on consortium blockchain was proposed, which realized P2P data trading without relying on a third party.Aiming at the problem that existing data trading models only consider the factors of the data itself and ignore the factors related to user tasks, a bi-level multi-objective optimization model was constructed based on multi-dimensional factors, such as data quality, data attributes, attribute relevance and consumer competition, to optimize the utilities of data provider (DP) and data consumer (DC).To solve the above model, an improved multi-objective genetic algorithm-collaborative NSGAII was proposed, calculated by the cooperation of DP, DC and data aggregator (AG).The simulation results show that the collaborative NSGAII achieves better performance in terms of the utilities of DP and DC, thus realizing more effective data trading.

Low-complexity and frequency-offset-robust synchronization algorithm based on CAZAC sequence

Fengkui GONG, Ni WEN, Guo LI, Yang GAO

2021, 42(2):  64-71.  doi:10.11959/j.issn.1000-436x.2021038

Asbtract ( 355 )   HTML ( 52)   PDF (850KB) ( 450 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

A low-complexity and frequency-offset-robust synchronization algorithm based on CAZAC sequence was proposed, which was suitable for burst OFDM systems.A new preamble sequence was constructed based on the CAZAC sequence, and its simplified timing metric function was derived by using the combination characteristics of the sequence, and then the delay correlation and symmetric correlation characteristics were used to achieve accurate and stable timing synchronization.Furthermore, based on the new constructed preamble sequence, a weighted frequency synchronization scheme combining cyclic prefix was proposed to obtain more accurate frequency offset estimation.Simulation results show that the proposed algorithm has good synchronization performance in both Gaussian channels and multipath fading channels regardless of frequency offset, and its computational complexity is lower than the most existing representative algorithms.

Depth first traversal algorithm for the back-off tree of distributed queuing

Wennai WANG, Yanhe ZHANG, Wei WU, Chen BAI, Bin WANG

2021, 42(2):  72-80.  doi:10.11959/j.issn.1000-436x.2021044

Asbtract ( 241 )   HTML ( 31)   PDF (864KB) ( 429 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

An analytic model was provided for the conventional distributed queueing (DQ) and its back-off tree operations, followed by a design of improving algorithm based on depth first traversal.Combing the specific analysis of complete binary tree with generalized extension by random tree reconstruction, the performance of proposed algorithm was evaluated on the throughput in both theory and simulation experiment.A theoretic optimal solution of contention slots of DQ frame and a brief description of simulation extension based on the open source NS-3 were presented.The simulation results show that the maximum stationary throughput by the proposed algorithm reaches 70% of the physical capacity of channel.

Digital modulation recognition based on discriminative restricted Boltzmann machine

Zhengquan LI, Yuan LIN, Mengya LI, Yang LIU, Qiong WU, Song XING

2021, 42(2):  81-91.  doi:10.11959/j.issn.1000-436x.2021012

Asbtract ( 262 )   HTML ( 29)   PDF (894KB) ( 530 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

In order to improve the performance of digital modulation recognition under high dynamic signal-to-noise ratio, a joint modulation recognition method based on high-order cumulant and discriminative restricted Boltzmann machine was proposed, which extracted the high-order cumulant of digital signals as signal features, comprehensively utilized the generation ability and classification ability of the discriminative restricted Boltzmann machine, analyzed the recognition rate of digital signals in environments containing Gaussian noise, time-varying phase offset or Rayleigh fading.Experimental results show that compared with traditional classification methods, the recognition performance of the proposed method is obviously improved.In addition, the use of the model’s generation ability to reconstruct the input features can effectively improve the signal recognition rate under low signal-to-noise ratio.

Hyperspectral image classification method based on multi-scale proximal feature concatenate network

Hongmin GAO, Xueying CAO, Zhonghao CHEN, Zaijun HUA, Chenming LI, Yue CHEN

2021, 42(2):  92-102.  doi:10.11959/j.issn.1000-436x.2021024

Asbtract ( 308 )   HTML ( 46)   PDF (1501KB) ( 348 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Aiming at the phenomenon that the hyperspectral classification algorithm based on traditional CNN model was not expressive enough in detail and the network structure was too complex, a hyperspectral image classification method based on multi-scale proximal feature concatenate network (MPFCN) was designed.By introducing multi-scale filter and cavity convolution, the model could be kept light and the discriminative features of the space spectrum could be obtained, and the correlation between the proximal features of the CNN was proposed to further enhance the detail expression.Experimental results on three benchmark hyperspectral image data sets show that the proposed method is superior to other classification models.

Multi-feature fusion classification method for communication specific emitter identification

Zunwen HE, Shuai HOU, Wancheng ZHANG, Yan ZHANG

2021, 42(2):  103-112.  doi:10.11959/j.issn.1000-436x.2021028

Asbtract ( 566 )   HTML ( 59)   PDF (1187KB) ( 566 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

A multi-feature fusion classification method based on multi-channel transform projection, integrated deep learning and generative adversarial network (GAN) was proposed for communication specific emitter identification.First, three-dimensional feature images were obtained by performing various transformations, the time and frequency domain projection of the signal was constructed to construct the feature datasets.GAN was used to expand the datasets.Then, a two-stage recognition and classification method based on multi-feature fusion was designed.Deep neural networks were used to learn the three feature datasets, and the initial classification results were obtained.Finally, through fusion and re-learning of the initial classification result, the final classification result was obtained.Based on the measurement and analysis of the actual signals, the experimental results show that the method has higher accuracy than the single feature extraction method.The multipath fading channel has been used to simulate the outdoor propagation environment, and the method has certain generalization performance to adapt to the complex wireless channel environments.

Shape similarity differential privacy trajectory protection mechanism based on relative entropy and K-means

Suxia ZHU, Shulun LIU, Guanglu SUN

2021, 42(2):  113-123.  doi:10.11959/j.issn.1000-436x.2021008

Asbtract ( 339 )   HTML ( 39)   PDF (927KB) ( 426 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

To solve the problem that most studies had not fully considered the sensitivity of location to privacy budget and the influence of trajectory shape, which made the usability of published trajectory poor, a shape similarity differential privacy trajectory protection mechanism based on relative entropy and K-means was proposed.Firstly, according to the topological relationship of geographic space, relative entropy was used to calculate the sensitivity of real location to privacy budget, a real-time calculation method of location sensitive privacy level was designed, and a new privacy model was built in combination with differential privacy budget.Secondly, K-means algorithm was used to cluster the release position to obtain the release position set that was most similar to the real position direction, and Fréchet distance was introduced to measure the similarity between the release track and the real track, so as to improve the availability of the release track.Experiments on real data sets show that the proposed trajectory protection mechanism has obvious advantages in trajectory availability compared with others.

UAV path intelligent planning in IoT data collection

Shu FU, Xiangyue YANG, Haijun ZHANG, Chen CHEN, Peng YU, Xin JIAN, Min LIU

2021, 42(2):  124-133.  doi:10.11959/j.issn.1000-436x.2021036

Asbtract ( 488 )   HTML ( 77)   PDF (1345KB) ( 669 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

To solve the problem of path planning of UAV data collection, it was generally be divided into global path planning and local path planning.For global path planning, it was modeled as an orientation problem, which was a combination of two classical optimization problems, the knapsack problem and the traveling salesman problem.The pointer network of deep learning was used to solve the model to obtain the service node set and service order under the energy constraint of the UAV.In terms of the local path planning, the reference signal strength (RSS) of the sensor node received by UAV was employed to learn the local flight path of UAV by deep Q network, which enabled the UAV to approach and serve the nodes.Simulation results show that the proposed scheme can effectively improve the revenue of UAV data collection under the energy constraint of UAV.

Comprehensive Reviews

Survey of application of machine learning in wireless channel modeling

Liu LIU, Jianhua ZHANG, Yuanyuan FAN, Li YU, Jiachi ZHANG

2021, 42(2):  134-153.  doi:10.11959/j.issn.1000-436x.2021001

Asbtract ( 1842 )   HTML ( 262)   PDF (1125KB) ( 2549 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Channel characterization is primary to the design of the wireless communication system.The conventional channel characterization method cannot learn the law of certain types of channels by itself, which limits its application in several special scenarios, such as Internet of things, millimeter wave communication and Internet of vehicles.Machine learning was able to process the big data and establish the model.Based on this, the cooperation between the machine learning and channel characterization was investigated.The channel multipath clustering, parameter estimation, model construction and wireless channel scene recognition were discussed, and recent significant research results in this field were provided.Finally, the future direction of the machine learning in wireless channel modeling was proposed.

Research prospects of user information detection from encrypted traffic of mobile devices

Tengfei ZHANG, Shunzheng YU

2021, 42(2):  154-167.  doi:10.11959/j.issn.1000-436x.2021040

Asbtract ( 467 )   HTML ( 68)   PDF (882KB) ( 739 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Encrypted traffic analysis of mobile devices can obtain multiple types of user information in an active or passive way, which provides protection for network security management and user privacy protection.The basic principles and key methods of data collection, feature selection, models and methods, and evaluation systems involved in these user information detection were analyzed and summarized.The problems in the existing projects were summarized, as well as the future research directions and challenges.

Correspondences

Secure transmission for NOMA downlink based on short packet communication

Gangcan SUN, Shaoke ZHAO, Wanming HAO, Zhengyu ZHU

2021, 42(2):  168-176.  doi:10.11959/j.issn.1000-436x.2021041

Asbtract ( 426 )   HTML ( 68)   PDF (801KB) ( 553 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

For the low-latency requirements of Internet of things (IoT) business, short packet communication (SPC) and non-orthogonal multiple access (NOMA) were combined to study the problem of secure transmission in the multi-user NOMA system with eavesdroppers.With the maximizing the secure throughput of weak uses as the objective, considering the user decoding error probability constraint, total power constraint and power allocation constraint, a low-complexity power allocation algorithm was proposed to realize secure transmission.In order to solve the problem caused by complex objective function formula and unreliable serial interference cancellation (SIC) technology, the proof that the compactness of the constraints was necessary to find the optimal solution.Under the constraint of maximum decoding error probability, the power constraint was transformed and calculated to obtain the strict limit of transmitting power for strong users, and the transmit power search set from base station to strong user was derived.Then, the one-dimensional search algorithm was used to allocate power resources to maximize the throughput of weak users.Simulation results prove that the proposed algorithm can effectively improve the security throughput of weak users in the system.

DWB-AES: an implementation of dynamic white-box based on AES

Bin WANG, Si CHEN, Jiadong CHEN, Xing WANG

2021, 42(2):  177-186.  doi:10.11959/j.issn.1000-436x.2021020

Asbtract ( 672 )   HTML ( 77)   PDF (26412KB) ( 455 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

The resources of IoT devices are limited.Therefore, security, flexibility and lightweight cryptographic modules are required.The idea of white-box cryptography can meet the needs of IoT devices.In common AES white-box implementations, keys are bound to look up tables.So the look up tables must be changed when the keys are changed.It is not flexible enough in practical applications.To solve this problem, a dynamic white-box implementation method for AES, which was called DWB-AES, was proposed.By changing the boundary between rounds, all operations of the encryption and decryption process were performed by looking up the tables, and the tables and the keys were respectively confused, so that the entire encryption and decryption process did not expose the key information.The look up tables need not to be changed every time when the keys changed, which was more practical.The security analysis of DWB-AES shows that the DWB-AES has higher white-box diversity and ambiguous, it can resist BGE attack and Mulder attack.