Table of Content

    23 March 2024, Volume 10 Issue 3
    Research on Privacy Protection Technology in Federated Learning
    2024, 10(3):  194. 
    Asbtract ( )   PDF (1252KB) ( )  
    References | Related Articles | Metrics
    In federated learning, multiple models are trained through parameter coordination without sharing raw data. However,  the extensive parameter exchange in this process renders the model vulnerable to threats not only from external users but also from internal participants. Therefore, research on privacy protection techniques in federated learning is crucial. This paper introduces the current research status on privacy protection in federated learning. It classifies the security threats of federated learning into external attacks and internal attacks.Based on this classification,  it summarizes external attack techniques such as model inversion attacks, external reconstruction attacks, and external inference attacks, as well as internal attack techniques such as poisoning attacks, internal reconstruction attacks, and internal inference attacks. From the perspective of attack and defense correspondence, this paper summarizes data perturbation techniques such as central differential privacy, local differential privacy, and distributed differential privacy, as well as process encryption techniques such as homomorphic encryption, secret sharing, and trusted execution environment. Finally, the paper analyzes the difficulties of federated learning privacy protection technology and identifies the key directions for its improvement.
    A Network Intrusion Detection Model Integrating CNN-BiGRU and  Attention Mechanism
    2024, 10(3):  202. 
    Asbtract ( )   PDF (2042KB) ( )  
    References | Related Articles | Metrics
    To enhance the feature extraction capabilities and classification accuracy of the network intrusion detection model, a network intrusion detection model integrating CNNBiGRU (Convolutional Neural NetworkBidirectional Gated Recurrent Unit) and attention mechanism is proposed. CNN is employed to effectively extract nonlinear features from traffic datasets,while BiGRU extracts timeseries features. The attention mechanism is then integrated to differentiate the importance of different types of traffic data through weighted means, thereby improvingthe overall performance of the model in feature extraction and classification. The experimental results indicate that the overall accuracy rate is 2.25% higher than that of the BiLSTM (Bidirectional Long ShortTerm Memory) model. Kfold crossvalidation results demonstrate that the proposed model's good generalization performance, avoiding the occurrence of overfitting phenomenon, and affirming its effectiveness and rationality.
    Malicious TLS Traffic Detection Based on Graph Representation#br#
    2024, 10(3):  209. 
    Asbtract ( )   PDF (1728KB) ( )  
    References | Related Articles | Metrics
    Owing to the need for privacy protection, encryption services online are becoming increasingly popular. However, this also provides an avenue for malicious traffic to hide itself. As a result, the identification of encrypted malicious traffic has become an important task for network management. Currently, some mainstream techniques based on machine learning and deep learning have achieved good results. However, most of these methods ignore the structure of traffic and do not provide indepth analysis of encryption protocols. To address this problem, this paper proposes a graph representation method for SSLTLS traffic, summarizes the key features of TLS traffic and considers traffic correlation from the perspective of multiple attributes such as source IP, destination port and packet count of the flow. Furthermore, this paper establishes a malicious traffic identification framework GCNRF based on graph convolutional neural network and random forest algorithm. This method transforms traffic into graph structure, integrates the structural information and node features of traffic for identification and classification. Experimental results on real public datasets show that the classification accuracy of this method is higher than that of current mainstream models.
    Malware Detection and Classification Based on GHM Visualization  and Deep Learning
    2024, 10(3):  216. 
    Asbtract ( )   PDF (2289KB) ( )  
    References | Related Articles | Metrics
    Malware detection is becoming more and more challenging due to the increasing complexity and variability of malicious code. Most mutated or unknown malicious programs are formed by improving or obfuscating the logic of existing malicious codes, so it is becoming more and more important to discover malicious code families and determine their malicious behaviors. In this paper, we proposed a novel malware visualization method based on GHM (Gray, HOG, Markov) for data preprocessing. Unlike the traditional visualization methods, this method extracts more effective data features through HOG and Markov in the visualization process, and constructs a threechannel color image. In addition, a VLMal classification model based on CNN and LSTM is constructed to realize the malware detection and classification of visual images. Experimental results show that this method can effectively detect and classify malicious code with good accuracy and stability.
    A Comparative Research on Hash Function in Blockchain in Post Quantum Era#br#
    2024, 10(3):  223. 
    Asbtract ( )   PDF (1514KB) ( )  
    References | Related Articles | Metrics
    Hash functions play an important role as the cornerstone of security in blockchain systems, playing an irreplaceable role in building consensus mechanisms and protecting data integrity. However, with the accelerated development of quantum technology, the emergence of quantum computers will pose a serious security threat to classical hash functions. Based on the parallel characteristics of quantum computing, Grover’s algorithm can provide squared acceleration compared with the classical counterpart in searching for hash conflicts. Quantum algorithms represented by the Grover’s algorithm can effectively implement quantum computing attacks against classical hash functions, such as mining attacks and forgery attacks. This paper explains the original image collision resistance, weak collision resistance and strong collision resistance of hash functions, and analyzes the main forms of quantum computing attacks against classical hash functions: preimage collision attacks and second image collision attacks. This paper conducts a comparative study on hash functions in blockchain from the perspective of antiquantum security, and five typical hash functions are analyzed and compared from the aspects of construction, input, output, advantages and disadvantages, and proposes the advice for designing hash functions in blockchain. Overall, this paper provides useful references for the design of hash functions in blockchain in the postquantum era.
    A Chinese Spam Detection Method Based on Content and ERNIE3.0-CapsNet
    2024, 10(3):  233. 
    Asbtract ( )   PDF (1248KB) ( )  
    References | Related Articles | Metrics
    In order to solve the problems of inadequate word vector representation and limited feature extraction richness in the current Chinese spam recognition methods based on deep learning, this paper proposes an improved recognition model by integrating the ERNIE3.0 pretraining model with the capsule neural network, referred to as ERNIE3.0CapsNet. For the Chinese spam content text, we leverage ERNIE3.0 to generate a word vector matrix with outstanding memory and reasoning capabilities, along with rich semantics. Subsequently, we employ the capsule neural network for feature extraction and classification. For the capsule neural network, we enhance its structure, adopting GELU as the activation function of its dynamic routing, and conduct a comparative experiment between five groups of similar models and four groups of activation functions. On the open source TREC06C Chinese email dataset, the proposed ERNIE3.0CapsNet model exhibits remarkable overall performance, achieving an accuracy rate of 99.45%. The experimental results demonstrate the superiority of ERNIE3.0CapsNet over methods such as ERNIE3.0TextCNN, ERNIE3.0RNN confirming the model’s effectiveness and superiority in Chinese spam recognition.
    Research on Zero Trust Access Control Model Based on Role and Attribute#br#
    2024, 10(3):  241. 
    Asbtract ( )   PDF (1562KB) ( )  
    References | Related Articles | Metrics
    In the face of many security threats in the network, the traditional access control model is increasingly exposed to the problems of poor dynamics of permission allocation, low sensitivity to new threats, and high complexity of resource allocation. This paper proposed a zero trust access control model based on role and attribute to address the above problems. The model used a logistic regression approach to trust assessment of access subjects to achieve access control with high sensitivity to access subject attribute, and adopted a new resource decision tree, which reduced the time complexity of resource permission assignment while achieving finergrained security for access control. Finally, verifying the model in this paper under typical application scenarios showed that the model was significantly better than the traditional access control model in terms of dynamic assignment of permissions.
    Survey of Research on Key Technologies of Internet Content Security
    2024, 10(3):  248. 
    Asbtract ( )   PDF (1234KB) ( )  
    References | Related Articles | Metrics
    The rapid development of the Internet and easy content creation and sharing have made Internet content security a top priority for Internet construction and supervision. The dramatic increase of information content with text, image, audio, and video as carriers has brought great challenges to Internet content security. Internet content security is rich in connotation, and we focused on four key applications including multimedia content filtering, fake information detection, public opinion perception, and data protection. Then, we summarized key technologies and main research work adopted in those applications. Finally, we discussed and prospected key issues of Internet content security in future research.
    Research on Data Circulation Legislative Process and the Evolution of Measures in European Union
    2024, 10(3):  256. 
    Asbtract ( )   PDF (948KB) ( )  
    References | Related Articles | Metrics
    As data emerges as a critical element in the digital economy, the European Union(EU) has issued a series of acts to promote the circulation and transaction of data factor, taking a leading position in the world at the legislative level.Through a comprehensive examination of the EU’s legislative processes and initiatives concerning data factors,this study seeks to extract insights applicable to our country. The research reveals that the EU, while rigorously safeguarding data security, promotes the circulation and transaction of data factors within its jurisdiction by advancing the flow of nonpersonal data, encouraging the reuse of public data, dismantling barriers between public and private sectors,and establishing a public data space. Through the comparison of the three major directions of public data open utilization, data security protection, and data circulation transactions with the EU legislation, we posit  that China can  adopt the relevant measures of the EU to continue to maintain data security, promote the opening and utilization of public data, and improve property rights and benefits distribution mechanism, thereby promoting the circulation and transaction of data factors in China.

    The Path and Choice of Improving Legislation of Personal Information Right
    2024, 10(3):  263. 
    Asbtract ( )   PDF (740KB) ( )  
    References | Related Articles | Metrics
    At present, information leakage has become a global problem, prompting many countries to enact legislation on safeguarding the right of personal information. Although the introduction of the Personal Information Protection Law of the People’s Republic of China has optimized the legislative protection of personal information right, the relevant normative documents are not effectively connected while the rank of some normative documents is unclear. Besides, the legislative protection of personal information right in special fields is not systematic, and some local legislatures lack legislative initiative, with loopholes in the supervision of public power. In contrast, the United States has accumulated much experience and lessons in the problems of personal information right protection. By conducting a comparative analysis, this paper identifies both the legislative advantages and disadvantages of the protection of the right to personal information in the United States. Subsequently, proposing some suggestions on the legislation of the personal information protection, aiming to improve the legislative protection of the right to personal information in China.

    Research on Location Attack Detection of VANET Based on Incremental Learning
    2024, 10(3):  277. 
    Asbtract ( )   PDF (1866KB) ( )  
    References | Related Articles | Metrics
    In recent years, deep learning has been widely employed in the detection of malicious position attacks on vehicles. However, deep learning models necessitate extensive training time and possess a large number of parameters. Detection methods based on deep learning lack scalability and cannot accommodate the needs of continuously generated new data in vehicular networks. To address these issues, this paper innovatively introduces incremental learning algorithms into the detection of malicious position attacks on vehicles to solve the above problems.This approach first extracts key features from the collected vehicle information data. Subsequently, a malicious position attack detection system is constructed, utilizing ridge regression to quickly approximate the vehicular network’s malicious position attack detection model. Finally, the incremental learning algorithm is applied to update and optimize the malicious position attack detection model to adapt to newly generated data in the vehicular network.Experimental results demonstrate that this method surpasses other methods such as SVM, KNN, and ANN in terms of performance. It can swiftly and progressively update and optimize the old model, thereby enhancing the system’s detection accuracy for malicious position attack behaviors.
    Research on Situational Awareness of 5G Joint Construction and Shared Network Visualization Security
    2024, 10(3):  277. 
    Asbtract ( )   PDF (1645KB) ( )  
    References | Related Articles | Metrics
    In the 5G coconstruction and sharing networks, the massive security logs and alarm audits have resulted in a lack of analysis of security data, and the disposal of security incidents is inefficient. In the face of many pain points such as lacking of threat perception ability and difficulty in security management, we propose a visualizing security situation awareness model and visualization solutions in 5G coconstruction and sharing networks. The comprehensive security situation awareness of the attack detection in the five fields of air port, operation and maintenance, system, transmission and configuration verification was tested and verified. Conduct correlation analysis of security incidents, identify crossdomain attack chain, discover the core intentions of attackers, to assist customers with achieving  active, dynamic, measurable ,visual 5G coconstruction and sharing network security, stable operation and the rapid development of the security industry.

    Research on the Application Authority of Face Recognition Technology Based on the Scene Congruence Theory
    2024, 10(3):  284. 
    Asbtract ( )   PDF (2275KB) ( )  
    References | Related Articles | Metrics
    As a biological identification technology, face recognition provides efficiency for various fields of society. However, in the existing legal regulation, there are still ambiguities in the legal regulation for the user authority of face recognition technology. On the basis of determining the nature of the right of face information, combining with the scene consistency theory, this paper determined the scope of use and right restrictions of face recognition technology in scenes such as public authority scene and profitmaking scene, constructed a personal protection system for the realistic use of artificial intelligence that integrates prevention beforehand and examination and punishment afterwards.