Content of Network·Security in our journal

        Published in last 1 year |  In last 2 years |  In last 3 years |  All
    Please wait a minute...
    For Selected: Toggle Thumbnails
    Source Localization of Network Information Propagation via Invertible Graph Diffusion
    ZHAI Wenshuo, ZHAO Xiang, CHEN Dong
    Journal of Frontiers of Computer Science and Technology    2024, 18 (5): 1348-1356.   DOI: 10.3778/j.issn.1673-9418.2308074
    With the development of society, security issues in various types of networks have become increasingly prominent, especially network propagation issues. Accurately locating the diffusion source points of network propagation is an important means to control network propagation. The research on the source location of network propagation also faces problems such as diverse network structure and complex dissemination mechanism. Therefore, this paper studies the problem of source location of network propagation based on graph neural networks, and an invertible graph diffusion model based on graph convolutional networks (GCNIGD) is proposed. In the stage of node susceptibility estimation, the graph convolutional neural network is combined to make full use of the structural information of the network considering the connection relationship between network nodes. In the stage of node feature construction, the graph diffusion theory is combined to spatially localize the information propagation in the network, so that the graph-based model can be enhanced by learning from multi-hop information. In the stage of source localization, the graph traceability problem is transformed into the inverse problem of graph diffusion, a reversible graph network is constructed to accurately estimate the source node, and the ill-posed problems in network traceability are solved. Finally, extensive experiments are conducted on six real-world datasets, and the results show that the proposed method outperforms the state-of-the-art methods. This study has important guiding significance for network security issues such as false information traceability, network attack traceability, etc.
    Reference | Related Articles | Metrics
    Abstract107
    PDF145
    Blockchain Transactions Using Attached Blocks and Discrete Token Negotiation for Delay-Tolerant Networks
    ZI Lingling, CONG Xin
    Journal of Frontiers of Computer Science and Technology    2024, 18 (5): 1357-1367.   DOI: 10.3778/j.issn.1673-9418.2307039
    The essence of blockchain is the competition for node mining rights, and the real-time connectivity of the network is prerequisite and guarantee. However, there exist delay-tolerant networks with data transmission latency, which puts nodes at a disadvantage in blockchain activities. To address this problem, a transaction architecture is constructed to provide service support for nodes in discontinuous connectivity networks to participate in blockchain activities. Firstly, an auxiliary block named as an attached block is constructed as a basis for distinguishing new blocks generated when the network is connected or disconnected. Secondly, with the miner’s identity and several random numbers as parameters, unique and verifiable discrete tokens are generated to obtain mining qualifications. On this basis, the mining qualification attribution algorithm and branch processing algorithm are designed. The former avoids the problem of the richest man doing evil in PoS (proof of stake) and DPoS (delegated proof of stake), and the latter ensures the attached chains are added to the main chain with an equal probability. Finally, a quadratic consensus scheme is designed to solve the problem of false and duplicate transactions that may exist in generated blocks when the network is disconnected. Theoretical proof and experimental analysis show that the architecture has advantages in indices such as reliability, acceptability, transaction throughput, confirmation time and branching rate.
    Reference | Related Articles | Metrics
    Abstract30
    PDF49
    Image-Text Retrieval Backdoor Attack with Diffusion-Based Image-Editing
    YANG Shun, LU Hengyang
    Journal of Frontiers of Computer Science and Technology    2024, 18 (4): 1068-1082.   DOI: 10.3778/j.issn.1673-9418.2305032
    Deep neural networks are susceptible to backdoor attacks during the training stage. When training an image-text retrieval model, if an attacker maliciously injects image-text pairs with a backdoor trigger into the training dataset, the backdoor will be embedded into the model. During the model inference stage, the infected model performs well on benign samples, whereas the secret trigger can activate the hidden backdoor and maliciously change the inference result to the result set by the attacker. The existing researches on backdoor attacks in image-text retrieval are based on the method of directly overlaying the trigger patterns on images, which has the disadvantages of low success rate, obvious abnormal features in poisoned image samples, and low visual concealment. This paper proposes a new backdoor attack method (Diffusion-MUBA) for image-text retrieval models based on diffusion models, designing trigger prompts for the diffusion model. Based on the correspondence between text keywords and regions of interest (ROI) in image-text pair samples, the ROI region in the image samples is edited to generate covert, smooth and natural poisoned training samples, to fine-tune through the pretrained model, establishing incorrect fine-grained word to region alignment in the image-text retrieval model, and embed hidden backdoors into the retrieval model. This paper designs the attack strategy of diffusion model image editing, proposes the backdoor attack model of bidirectional image-text retrieval, and achieves good results in the backdoor attack experiments of image-text retrieval and text-image retrieval. Compared with other backdoor attack methods, it improves the attack success rate, and avoids the impact of introducing specific characteristics of trigger patterns, watermarks, perturbations, local distortions and deformation in the poisoned samples. On this basis, this paper proposes a backdoor attack defense method based on object detection and text matching. It is hoped that the study on the feasibility, concealment, and implementation of backdoor attacks in image and text retrieval may contribute to the development of multimodal backdoor attack defense.
    Reference | Related Articles | Metrics
    Abstract222
    PDF170
    Cryptomining Malware Early Detection Method Based on AECD Embedding
    CAO Chuanbo, GUO Chun, LI Xianchao, SHEN Guowei
    Journal of Frontiers of Computer Science and Technology    2024, 18 (4): 1083-1093.   DOI: 10.3778/j.issn.1673-9418.2307023
    Cryptomining malware can compromise system security, reduce hardware lifetime, and cause significant power consumption. Therefore, implementing cryptomining malware early detection to stop its damage in time is critical to system security. The existing dynamic analysis-based cryptomining malware early detection methods are hard to balance the timeliness and accuracy of detection. To detect cryptomining malware accurately and timely, this paper integrates a certain length of API (application programming interface) names, API operation categories and DLLs (dynamic link libraries) called by cryptomining malware in the early stage of operation to more fully describe its behavioral information in this stage, and proposes the AECD (API embedding based on category and DLL) embedding and further proposes a cryptomining malware early detection method based on AECD embedding (CEDMA). CEDMA uses the API sequence called by software in the early stage of operation as the object of detection and uses AECD embedding and TextCNN (text convolutional neural network) to build a detection model to implement cryptomining malware early detection. Experimental results show that when CEDMA takes the 3000 API sequence called for the first time after the software runs as input, it can detect the known and unknown cryptomining malware samples in the experiment with 98.21% and 96.76% accuracy values, respectively.
    Reference | Related Articles | Metrics
    Abstract48
    PDF45
    Lightweight Routing Protocol for Named Data Networking
    ZOU Bowen, SONG Tian, LI Tianlong, YANG Yating
    Journal of Frontiers of Computer Science and Technology    2024, 18 (3): 795-804.   DOI: 10.3778/j.issn.1673-9418.2304001
    Named data networking (NDN) is a novel network architecture that revolves around information-centric principles. In order to acquire global routing information within the network, typical NDN routing protocols rely on data synchronization mechanisms to achieve global routing updates. However, such synchronization protocols operate at the application layer, which introduces challenges in dynamic network environments, such as high costs associated with network information updates and limited ability to perceive network dynamics. To address these issues, this paper designs and implements a lightweight named-binding routing protocol (NBRP). This approach binds routing node names with their published content names, enabling routing nodes to make targeted requests for routing update information without requiring additional synchronization protocols, thus achieving lightweight routing information exchange. To further reduce the transmission costs of routing updates, this paper introduces a reusable routing information packet naming format and a routing information incremental transmission mechanism. Additionally, to validate the effectiveness of the proposed solution, corresponding functional modules are implemented and tested within the Linux kernel. Experimental results demonstrate that, in dynamic network environments, compared with traditional NDN link-state routing protocols, the proposed method can reduce routing update synchronization costs by 72% and increase link-state awareness and response speed by nearly twofold.
    Reference | Related Articles | Metrics
    Abstract75
    PDF89
    Application Layer Protocol Recognition Incorporating SENet and Transformer
    CHEN Qian, HONG Zheng, SI Jianpeng
    Journal of Frontiers of Computer Science and Technology    2024, 18 (3): 805-817.   DOI: 10.3778/j.issn.1673-9418.2304045
    Protocol recognition technology assumes a crucial position and exerts significant influence in the domains of network communication and information security. Existing protocol recognition methods based on spatio-temporal features cannot adequately and comprehensively extract protocol features. An application layer protocol recognition method incorporating SENet channel attention and Transformer is proposed. The model focuses on spatio- temporal feature extraction of protocol data, and the model consists of a spatial feature extraction module and a time extraction module. SE blocks are added to the residual network to capture the associations between multiple channels and adaptively assign weights, so as to extract the key space features in different channels. The temporal feature extraction module is constructed by stacking the transformer encoders based on multi-head attention mechanism. This module is used to comprehensively capture temporal features of the protocol data by directly leveraging the positional information of the input data. After extracting and learning more detailed spatial features and more comprehensive temporal features, better protocol feature representation is obtained to improve protocol recognition performance. Experiments are conducted on the ISCX2012 and CSE_CIC_IDS2018 hybrid datasets, and the results show that the overall recognition accuracy of the proposed model reaches 99.20%, and the [F1] score reaches 98.99%, which are higher than those of the comparison models.
    Reference | Related Articles | Metrics
    Abstract133
    PDF139
    Research on Hybrid Trust Management Scheme for VANETs
    XIANG Dan, CHEN Zemao
    Journal of Frontiers of Computer Science and Technology    2024, 18 (2): 516-525.   DOI: 10.3778/j.issn.1673-9418.2211032
    With the rapid development of intelligent transportation, vehicular ad hoc networks (VANETs) has broad development prospects, but also faces a variety of security threats. A distributed hybrid trust management scheme called HTMS-V is proposed for insider attackers and false message detection in VANETs. Specifically, considering the characteristics of VANETs, node trust is evaluated based on the improved subjective logic model. The evaluation combines direct trust and indirect trust, and the trust relationship between nodes is established based on interaction records. Message trust is evaluated based on node trust and the distance between nodes. And the scheme identifies false messages and malicious nodes based on the trust evaluation results. This paper tests the performance of the   HTMS-V scheme under four attack scenarios. The simulation results show that the HTMS-V scheme effectively resists various attacks in VANETs and is able to identify most false messages and malicious nodes even if the malicious node ratio reaches 40%, and the performance of the HTMS-V scheme is obviously better than that of the baseline scheme which is composed of subjective logical model and distance based weighted voting.
    Reference | Related Articles | Metrics
    Abstract102
    PDF102
    Adversarial Examples with Unlimited Amount of Additions
    JIANG Zhoujie, CHEN Yi, XIONG Ziman, GUO Chun, SHEN Guowei
    Journal of Frontiers of Computer Science and Technology    2024, 18 (2): 526-537.   DOI: 10.3778/j.issn.1673-9418.2302070
    Malware detection methods based on gray images and deep learning have the characteristics of high detection accuracy and no need of feature engineering. Unfortunately, adversarial examples (AEs) can deceive such detection methods. However, it is difficult to reduce the detection accuracy of this kind of detection method greatly without destroying the functional integrity of the original file. By analyzing the structure and loading mechanism of portable executable (PE) files, this paper proposes an unrestricted add-amount bytecode attack (BAUAA). BAUAA generates adversarial samples by adding bytecode to a “section additional space” in the PE file that is scattered after each section and is not loaded into memory, and because of the unlimited amount of this space that can be added, the generated adversarial samples can be transformed into grayscale images that vary in size and texture, which can affect the discrimination accuracy of gray images and deep learning-based malware detection methods. The experimental results show that the detection accuracy of the malware detection method based on gray images and deep learning for the AEs generated by BAUAA is significantly lower than that for the non-AEs. To avoid the abuse of BAUAA in reality, it proposes a targeted AE detection method.
    Reference | Related Articles | Metrics
    Abstract66
    PDF62
    Protecting Face Privacy via Beautification
    WANG Tao, ZHANG Yushu, ZHAO Ruoyu, WEN Wenying, ZHU Youwen
    Journal of Frontiers of Computer Science and Technology    2024, 18 (1): 244-251.   DOI: 10.3778/j.issn.1673-9418.2210098
    Face images distributed widely on social networks are vulnerable to inferring sensitive information by unauthorized automatic identification systems, which poses a threat to user privacy. To protect face privacy, several methods have been proposed to generate highly transferable adversarial faces to remove identity information. However, the results generated by existing methods still suffer from obvious perturbations that make visual perception poor, which is not friendly for sharing on social networks. This paper proposes an adversarial face generation scheme via beautification, i.e., Adv-beauty. Adv-beauty utilizes a face matcher and a beautification discriminator to collaboratively supervise the training process of the generator, prompting the generator to produce a beauty-like perturbation on the original face to confront the face matcher. In other words, the pixel changes produced by the beauty mask the undesirable visual effects produced by the perturbations. In addition, this paper sets an adversarial threshold for identity loss to prevent face distortion due to excessive deviation of identity features. Sufficient experiments show that Adv-beauty maintains good visual results and is effectively against unknown face recognition classifiers and commercial APIs.
    Reference | Related Articles | Metrics
    Abstract193
    PDF177
    DAGGraph: Blockchain Suitable for Mobile Ad Hoc Networks
    ZHANG Wentao, HUANG Jianhua, GU Bin, NING Yuhao, GONG Zaiwei
    Journal of Frontiers of Computer Science and Technology    2024, 18 (1): 252-264.   DOI: 10.3778/j.issn.1673-9418.2211070
    Aiming at the challenges faced by the combination of blockchain and mobile ad hoc networks, a system model DAGGraph based on DAG (directed acyclic graph) is proposed, which adopts the DAG structure to adapt to the network split caused by mobility. Firstly, the clustering algorithm is optimized, and an algorithm for limiting the density of nodes in the cluster is proposed, which effectively solves the problem of throughput reduction and energy consumption increase caused by the uncontrolled increase of the number of nodes in the cluster. Secondly, for network splitting and merging caused by the rapid movement of the nodes, a block recovery algorithm based on data synchronization between cluster heads is proposed. The legal blocks generated by all nodes are preserved through DAG. When the network is merged, the original cluster heads exchange their generated block branches, realizing the recovery of the block branches. Finally, a simplified block appending algorithm is proposed, which simplifies the block appending process on the premise that the internal nodes are trusted, reduces the error caused by block propagation in the mobile environment, shortens the block confirmation time, and improves the system throughput. Security analysis shows that DAGGraph can resist common attacks against blockchain, and resist denial-of-service attacks against mobile ad hoc networks. Simulation results show that the latency and throughput of DAGGraph are better than existing IoT blockchain solutions in most cases.
    Reference | Related Articles | Metrics
    Abstract50
    PDF62
    16-bit S-box Design Method Based on L-M-NFSR Structure
    WU Xiaonian, SHU Rui, DOU Daorao, ZHANG Runlian, WEI Yongzhuang
    Journal of Frontiers of Computer Science and Technology    2023, 17 (10): 2511-2518.   DOI: 10.3778/j.issn.1673-9418.2207012
    S-box is an important component for non-linear transformation in symmetric cryptographic algorithm, and the security of S-box determines the security of the cryptographic algorithm. In order to construct 16-bit S-box with strong security, a new L-M-NFSR structure is designed based on the Lai-Massey structure and nonlinear feedback shift register (NFSR) component. In the new structure, 8-bit S-boxes with advanced encryption standard (AES) algorithm S-box affine equivalence are selected as the round function to reduce the complexity of designing and increase the variability of the structure. Two designed NFSR components that can conform to strict avalanche properties with a small number of iterations are placed into two branches of the structure to improve the diffusion effect of the structure. Then, 16-bit S-boxes are constructed by 3-round iteration and traversal search. Furthermore, based on this structure, a large number of new 16-bit S-boxes can be generated by replacing 8-bit S-boxes in the round function with 8-bit S-boxes which are affine equivalent to the AES algorithm S-box. To improve the effici-ency of the evaluation of the properties of the constructed 16-bit S-boxes, parallel computation is performed using graphics processing unit (GPU). The test results show that the newly constructed 16-bit S-boxes have good crypto-graphic properties, which satisfy bijectivity with optimal algebraic number 15, the highest nonlinearity 31992, the lowest differential uniformity 18, and minimum signal-to-noise ratio 146.712, with excellent security against mathematical attacks and differential power analysis.
    Reference | Related Articles | Metrics
    Abstract172
    PDF138
    SM9 Identity-Based Encryption Algorithm with Deniable Authentication
    ZHAO Chenyang, KE Pinhui, LIN Changlu
    Journal of Frontiers of Computer Science and Technology    2023, 17 (10): 2519-2528.   DOI: 10.3778/j.issn.1673-9418.2207013
    SM9 identity-based encryption algorithm is a commercial identity-based encryption algorithm independently designed by our country, which has become the standard of the domestic identity-based encryption algorithm industry, and is widely used in e-mail, electronic voting and online negotiation, etc. However, SM9 identity-based encryption algorithm can??t effectively protect the identity privacy of the sender. Based on SM9 identity-based encryption algorithm, combined with the deniable authentication protocol, this paper proposes SM9 identity-based encryption algorithm with deniable authentication. This algorithm allows the sender to deny its participation after the protocol runs, and only the intended receiver can identify the true source of the given message. At the same time, the receiver can??t convince any other third party that the message is sent by a specific sender. Under the assumption of DBDH??s difficult problem, the formal definition and security model of SM9 identity-based encryption algorithm with deniable authentication are given, and the security analysis of the algorithm is given under the random oracle model, which proves that the algorithm can satisfy denial, confidentiality and deniable authentication at the same time. Theoretical analysis and simulation experiments show that the proposed algorithm not only maintains the efficiency advantage of SM9 identity-based encryption algorithm, but also has a lower computational overhead than other identity-based encryption algorithms with deniable authentication.
    Reference | Related Articles | Metrics
    Abstract171
    PDF127
    Risk Management Policies and Analysis of Cross-Chain Digital Assets
    TIAN Haibo, YE Wan
    Journal of Frontiers of Computer Science and Technology    2023, 17 (9): 2219-2228.   DOI: 10.3778/j.issn.1673-9418.2204052
    Due to the complexity of smart contracts and the different security levels of blockchains on which various digital assets are issued, a series of security incidents about cross-chain digital assets emerge, highlighting the serious security threats faced by cross-chain digital assets. For the high risk of cross-chain digital assets, this paper analyzes two typical cross-chain digital asset security incidents of bEarn Fi and Poly Network, obtaining a conclusion that the digital assets of cross-chain service nodes need to be protected. Then considering the characteristics of blockchain, focusing on cross-chain asset exchange, this paper proposes seven strategies of cross-chain digital asset risk management. With the traditional anomaly detection techniques, algorithms are provided to detect the abnor-mality of cross-chain asset exchange requests of users from the aspects of single transaction, multiple transactions, K-means and external data sources, so as to identify and terminate the abnormal requests, reducing the security risks of service-node cross-chain digital assets. Finally, by the simulation method, this paper presents a simple cross-chain system based on the hashed time-lock technique, which shows a serial of test events, to confirm the effectiveness of the risk management policies and anomaly detection methods. The test results show that the strategies of risk management and the corresponding anomaly detection algorithms can lower the loss of service-node cross-chain digital assets and reduce the risks of digital assets belonging to cross-chain service nodes.
    Reference | Related Articles | Metrics
    Abstract178
    PDF183
    Fault-Tolerant Bidirectional Choice Attack Combining EDCA and CPA
    ZHANG Meiling, SHANG Lirong, ZHENG Dong
    Journal of Frontiers of Computer Science and Technology    2023, 17 (9): 2229-2240.   DOI: 10.3778/j.issn.1673-9418.2206108
    When the designed attack scheme is fault-tolerant, it is often necessary to pick out the correct key in a very large candidate space. How to effectively achieve this goal is a very important and challenging problem in side-channel attacks. Aiming at this problem, a fault-tolerant bidirectional choice attack combining EDCA (Euclidean distance enhanced collision attack) and CPA (correlation power analysis) is studied with AES-128 as the target. Firstly, in order to improve the success rate of collision detection, the EDCA is proposed. Compared with correlation enhanced collision attack (CCA), EDCA utilizes the Euclidean distance to distinguish the similarity between two sets of energy traces. Its collision detection has a higher success rate, making the optimization more practical and meaningful. In addition, combined with EDCA and CPA, the keys and the corresponding collision pairs are grouped, and then bidirectional screening is performed to obtain the optimal collision chain, which greatly reduces the candidate space and reduces the complexity of key enumeration, effectively recovering the key. Experimental results show that, under the conditions of low signal-to-noise ratio[SNR=-3 dB]and [SNR=-6 dB], setting the threshold of collision pair being 5, the success rate of the proposed scheme reaches 98.78% and 80.25% when there are 3000 energy traces, both of which are better than the existing schemes.
    Reference | Related Articles | Metrics
    Abstract91
    PDF62
    Linear Complexity of Hidden Weighted Bit Functions
    CHEN Zhiru, FENG Ligang, ZHU Youwen
    Journal of Frontiers of Computer Science and Technology    2023, 17 (8): 1974-1980.   DOI: 10.3778/j.issn.1673-9418.2203038
    Boolean functions are crucial primitive in block cipher and are also used to design pseudorandom sequences. They play a crucial role in the design of symmetric cryptography and its analysis, and the study on the cryptographic properties of Boolean functions is a hotspot in cryptography. The hidden weighted bit functions (HWBF) are paid attention since they have many “good” cryptographic measures. However, there are no results on their linear complexity in the literature. Therefore, this paper discusses a family of binary sequences of period [2n]built by using [n-]variable HWBF (hidden weighted bit functions). It is proven that such sequences are balanced with maximal linear complexity using mathematical method. The 2-error linear complexity of the sequences is also determined in terms of the Hasse derivative and Lucas congruence. When [n(mod4)∈{0,1,3}], the values of the 2-error linear complexity are maximal. Results indicate that such sequences are of “good” cryptographic measures.
    Reference | Related Articles | Metrics
    Abstract141
    PDF74
    Cascaded Two-Stream Attention Networks for Traceability Analysis of Copy-Move Images
    JI Yanqing, ZHANG Yujin
    Journal of Frontiers of Computer Science and Technology    2023, 17 (8): 1981-1994.   DOI: 10.3778/j.issn.1673-9418.2203118
    Copy-move is a common way of the image forgery. Traditional methods are committed to locating tam-pering regions of copy-move tampering images, but the accurate distinction between the source and target of the copy-move image has become a bottleneck in the field of image forensics. At present, algorithms which can locate the tampering source and target regions from the original copy-move forged images still have some disadvantages. Therefore, this paper proposes a cascaded two-stream attention network for traceability analysis of copy-move images. It is divided into two stages. The first stage of the network includes a coding network, a module to analyze features and a decoding network. In the coding part, lightweight MobileNetV2 is used as the backbone to extract low and deep features as the double outputs of the network. In the module of analyzing features, tampering regions in deep features are multi-dimensionally captured by the attention mechanism of similar features and atrous spatial pyramid pooling module. At the same time, the low feature is used to improve the model’s performance of segmenting edges and details of tampered regions. In the decoding part, the feature map is predicted pixel by pixel and sampled. In the second stage of the network, the tampering regions detected in the first stage network are distinguished between the source and target. It is also a two-stream network. The inputs of two-branch are the original image blocks including the source or target and image blocks after extracting noise. The multiscale features are used to predict category, and the final mask is output by the region mapping. Experimental results show that the proposed network can not only locate the tampering regions, but also distinguish the source and target. The performance compared with the latest algorithm of the first stage of the network in the test dataset and two public datasets is increased by 9.4, 2.6, and 2.5 percentage points respectively, and the end-to-end performance in the test dataset is improved by 12.03%. At the same time, it has better robustness to conventional image post-processing.
    Reference | Related Articles | Metrics
    Abstract115
    PDF78
    Research and Implementation of Multi-source Account Authentication for High-Performance Computing Environment
    HE Rong, XIAO Haili, WANG Xiaoning, CHI Xuebin
    Journal of Frontiers of Computer Science and Technology    2023, 17 (7): 1700-1707.   DOI: 10.3778/j.issn.1673-9418.2202037
    High-performance computing (HPC) environment provides researchers with high level high-performance computing application services with unified access entrance, unified use method and user technical support through shielding the heterogeneity of job management system, access mode, management system and so on. With the development of the environment, more and more supercomputing centers, application communities, and service platforms are connected. Accounts of the supercomputing centers, communities, and service platforms are required to log in to the HPC environment using their original ways. The existing high-performance computing environment supports only grid accounts authenticated by LDAP (lightweight directory access protocol). Application communities and service platforms have their own users and different authentication modes. In order to provide a unified authentication center for the environment, this paper studies the multi-source account authentication technology and develops the multi-source user authentication system. At present, the HPC environment has supported multiple super computing centers. The major communities and platforms in the “high-performance computing” special project also have realized the connection with the national HPC environment through the multi-source account authentication technology. After the connection, accounts of the communities and platforms can log in with the environment grid accounts, at the same time use relevant resources.
    Reference | Related Articles | Metrics
    Abstract116
    PDF139
    Research on Deep Reinforcement Learning Method for Throughput Optimization of Internet of Vehicles Blockchain
    ZHANG Li, DUAN Mingda, WAN Jianxiong, LI Leixiao, LIU Chuyi
    Journal of Frontiers of Computer Science and Technology    2023, 17 (7): 1708-1718.   DOI: 10.3778/j.issn.1673-9418.2205019
    The rapid development of Internet of vehicles (IoV) depends on the safe and reliable infrastructure for storing and sharing large amounts of data. Blockchain, a kind of distributed data storage technology that cannot be forged and tampered with, can solve the security and privacy issues of IoV. However, the low throughput of blockchain hinders its wide application in IoV. The current research on blockchain throughput optimization has poor scalability because of its action space explosion. Aiming at the above problems, a blockchain throughput optimi-zation method in IoV based on deep reinforcement learning (DRL) is proposed to maximize the transaction throughput, and optimize the throughput of the blockchain by choosing block producers and consensus algorithms, adjusting block size and block interval while ensuring the decentralization, low delay and high security of the underlying blockchain system. This method introduces the branching dueling Q-network (BDQ) framework in DRL, carries out fine-grained division for action space, and solves the problem of action space explosion of traditional deep reinforcement learning methods. Simulation results show that the proposed method can improve the throughput of blockchain in IoV effectively.
    Reference | Related Articles | Metrics
    Abstract192
    PDF161
    Research on Malicious Attack Model of Blockchain Multi-mining Pools
    LIU Qiang, SONG Baoyan, JI Wanting, WANG Junlu
    Journal of Frontiers of Computer Science and Technology    2023, 17 (7): 1719-1728.   DOI: 10.3778/j.issn.1673-9418.2111107
    The double-spending attack against transactions in the blockchain is the focus of blockchain security research. As a large collection of computing power, mining pools are potential for malicious behavior, and the combination of multiple mining pools can double-spend transactions through different attack forms, which brings huge security risks to the blockchain. Based on the combination of malicious attacks in multiple mining pools, this paper proposes two malicious attack models in multiple mining pools. Firstly, in the multi-mining pool centralized attack model, the computing power of multiple mining pools is aggregated into a malicious computing power set, which focuses on double-spending attacks on the blockchain. Secondly, in the multi-mining pool decentralized attack model, each mining pool is independent malicious computing power set, and the malicious computing power is distributed and synchronized among the blockchain computing power attacks. In addition, combining the number of mining pools, the percentage of computing power and other parameters, the internal operating mechanism of two multi-mining pool models is simulated and derived. Based on this mechanism, the security constraints of the two attack models are constructed. The experiment verifies the effectiveness of the two mining pool attack models, and digitization shows the attack laws in two multi-mining pool attack models, as well as the similarities and differences between the two models. Finally, combining with theoretical models and experimental results, this paper puts forward the prevention and control strategies.
    Reference | Related Articles | Metrics
    Abstract224
    PDF171