Abstract In recent years, cloud-edge-end collaborative federated learning frameworks have been widely used in many scenarios and achieved good results. However, with the complexity of application requirements, the problems of device heterogeneity and data heterogeneity become more prominent. Traditional frameworks often face challenges such as uneven allocation of computational resources and inefficient training when dealing with these problems. To address this problem, this paper proposes a novel federated learning framework for multi-domain collaborative analysis of networked encrypted flows. First, we split the model training tasks, intelligently assign part of the model training tasks to terminal devices based on their performance, while the remaining model training tasks that require more computational resources are handed over to edge servers. Second, we introduce a resource scheduling scheme among edge servers to reasonably allocate model training tasks and fully utilize resources. Finally, high quality global models are obtained through a weight-enabled global model aggregation scheme. Experiments show that our proposed scheme can effectively address the impact of device heterogeneity and data heterogeneity in encrypted traffic identification in cross-domain networks, and improve the training efficiency and model performance of the overall system while ensuring data privacy and security.

Abstract Although there are numerous advanced and well-established models and methods available for code auditing tasks, their interpretability remains a significant challenge. For machine learning models designed to address code auditing problems, we often know that they can identify vulnerable code but lack insight into their decision-making criteria or whether they have effectively captured the characteristics of vulnerable code. To evaluate the capability of such models in extracting vulnerability-related features, this paper proposes a method called Program-PLATE. By extending a single vulnerable file into a PLATE-dataset, this method enables a more objective assessment of the model’s performance on the PLATE-dataset. We applied this method to evaluate multiple models, conducted an in-depth analysis based on the results, and provided suggestions and expectations for future research directions.

Abstract With the development of electric vehicles, charging stations have garnered significant attention and progress. As a result, various charging platforms have emerged. However, due to the lack of shared charging information, issues such as low utilization of charging stations during peak hours and insufficient station availability have negatively impacted user experience. Inspired by the concept of an aggregation platform, we propose an aggregated charging platform for electric vehicles (EVs). However, challenges related to trust, such as single-point failures and the difficulty of achieving unified identity management, hinder the implementation of the aggregation model. To address these issues, we propose a blockchain-based EV aggregation charging platform model and a cross-platform identity authentication scheme. Leveraging the decentralized and highly secure characteristics of blockchain, we construct a decentralized aggregation platform and enhance the credibility and reliability of cross-platform authentication through smart contracts. Furthermore, we design a batch authentication scheme to efficiently handle a large number of authentication requests. Simulation results demonstrate that, compared to other schemes, our proposed approach improves authentication efficiency by 50–66.6%.

In the Industrial Internet of Things (IIoT), the identification resolution system enhances communication and overall efficiency between isolated work islands, ensuring the trustworthiness and effectiveness of secure resource sharing across domains through cross-domain authentication. However, traditional identity authentication methods fail to empower users with control over their identity information and face challenges such as difficulty in tracking anonymous users, high computational overhead, and insufficient cross-domain trust. To address these issues, this paper proposes a cross-domain identity authentication scheme based on self-sovereign identity. The scheme leverages aggregate signature technology to enhance authentication efficiency, integrates blockchain technology and smart contracts to achieve cross-domain trust, and designs a mechanism for threshold identity tracking and revocation, as well as an attribute credential update mechanism to enable secure and efficient cross-domain authentication in the identification resolution system. The paper provides formal security definitions and proofs and evaluates the computational and storage efficiency of the scheme through theoretical analysis and experimental simulations. The results demonstrate that the proposed scheme offers significant advantages in resource-constrained scenarios within the IIoT.

To solve the problem of climate warming, countries around the world have paid special attention to the construction of carbon governance. Carbon emission accounting is an important policy tool to control the vented CO2. But at present, there are third-party agencies in carbon emission accounting that cannot ensure the fairness and impartiality of accounting, and there may be risks such as illegal use and leakage of sensitive information in the process of carbon emission data transmission. Therefore, We design the blockchain-based carbon emission security accounting scheme (BCESAS) and propose cross-chain verification contract to ensure the efficiency of cross-chain information accounting. In addition, bilinear pairing is used to ensure data integrity, and we encrypt private data using an improved and more secure homomorphic encryption algorithm to ensure that privacy is not leaked during the transfer of carbon emission data, which is more efficent than other homomorphic encryption algorithms. We also use reputation mechanism to regulate the behavior of carbon emission auditors. The theoretical and experimental analysis demonstrates that BCESAS can verify the integrity, correctness and privacy of cross-chain data calculation result effectively, realizing secure and reliable expansion of blockchain.

Abstract With the development of deep learning technology, great progress has been made in the field of coverless steganography based on deep learning technology, including some selection-based steganography methods that use deep learning technology and all generation-based steganography methods, however both of which have their limitations. The former is difficult to meet actual communication requirements in terms of communication capacity and completeness due to the limit of the algorithm. Due to the irreversibility of the process of generating secret images from message codeword, the recovery accuracy of the latter is very poor. To this end, this paper designs a robust joint coverless image steganography scheme called Joint Coverless Image Steganography (JoCS). Firstly, this paper proposes the Semantic Factorization Fitting module (SeFF) and the Transform Domain Steganography module (TrDS). The former adds the secret message to the input vector of the low resolution layer in the StyleGAN generator network, which establishs a mapping rule between message codeword and the coarse feature of the generated image, and then the extractor is used to fit the above mapping rule, which has excellent robustness and completeness; the latter encodes the main content area of the image based on the encoder in VQGAN, and then adds secret message to the latent vector of the encoded image, which achieves the steganography in the latent domain of the image. Secondly, we demonstrate the independence between two modules and the advantages of connecting two modules. By using the image generated in the SeFF module as the cover image in the TrDS module, secondary steganography of a single image is achieved, based on which we design the JoCS scheme. The results show that our scheme breaks through the communication capacity limit in the selection-based coverless methods while guaranteeing 100% completeness, excellent image quality and outstanding robustness against various image attacks. Moreover, our scheme exhibits strong security against detection by multiple steganalysis tools and excellent practicality in practical communication. Finally, this paper also discusses the following three points as further elaboration of the scheme: (1) the advantages of the mapping rule in the SeFF module (2) the verification of the independence between the two modules (3) the flexibility of the joint steganography scheme.

Energy systems are currently undergoing a transformation towards new paradigms characterized by decarbonization, decentralization, democratization, and digitalization. In this evolving context, energy blockchain, aiming to enhance efficiency, transparency, and security, emerges as an integrated technological solution designed to address the diverse challenges in this field. Data security is essential for the reliable and efficient functioning of energy blockchain. The pressing need to address challenges related to secure data storage, effective data management, and efficient data utilization is increasingly vital. This paper offers a comprehensive survey of academic discourse on energy blockchain data security over the past five years, adopting an all-encompassing perspective that spans data storage, management, and utilization. Our work systematically evaluates and contrasts the strengths and weaknesses of various research methodologies. Additionally, this paper proposes an integrated hierarchical on-chain and off-chain security energy blockchain architecture, specifically designed to meet the complex security requirements of multi-blockchain business environments. Concludingly, this paper identifies key directions for future research, particularly in advancing the integration of storage, management, and utilization of energy blockchain data security.

To solve the current situation of low vehicle-to-pile ratio, charging pile(CP) operators incorporate private CPs into the shared charging system. However, the introduction of private CP has brought about the problem of poor service quality. Reputation is a common service evaluation scheme, in which the third-party reputation scheme has the issue of single point of failure; although the blockchain-based reputation scheme solves the single point of failure issue, it also brings the challenges of storage and query efficiency. It is a feasible solution to classify and store information on multiple chains, and at this time, reputation needs to be calculated in a cross-chain mode. Crosschain reputation calculation faces the problems of correctness verification, integrity verification and efficiency. Therefore, this paper proposes a verifiable and efficient cross-chain calculation model for CP reputation. Specially, in this model, we propose a verifiable cross-chain contract calculation scheme that adopts polynomial commitment to solve the problems of polynomial damage and tampering that may be encountered in the crosschain process of outsourced polynomials, so as to ensure the integrity and correctness of polynomial calculations. In addition, the miner selection and incentive mechanism algorithm in this scheme ensures the correctness of extracted information when the outsourced polynomial is calculated on the blockchain. The security analysis and experimental results demonstrate that this scheme is feasible in practice.

Developing a fair, efficient, and scalable data trading protocol in decentralized networks has attracted much research effort recently. Zero-knowledge contingent payments (ZKCP) allows sellers and buyers to complete their trade fairly over the blockchain using zero-knowledge proofs. However, it suffers from memory-intensive requirements and scalability limitations. In this paper, we propose a practical data trading protocol tailored for Sudoku solutions, which is fair, efficient, and scalable. The core component of our protocol is a zero-knowledge argument for the correctness of a Sudoku solution of homomorphic encryption. This argument achieves sublinear communication complexity and the number of group exponentiations for both proving and verification is linear in the size of Sudoku solutions. The security of our protocol can be proven in the random oracle model under the Decision Diffie-Hellman assumption. In addition, we devise a mechanism that allows buyers to recover the private key through two zero-knowledge proofs and prevents the direct exposure of the decryption key. Furthermore, we implement the proposed protocol on the Ethereum testnet, and the experimental results show a significant improvement in overall efficiency.

Abstract Deep learning models are well known to be susceptible to backdoor attack, where the attacker only needs to provide a tampered dataset on which the triggers are injected. Models trained on the dataset will passively implant the backdoor, and triggers on the input can mislead the models during testing. Our study shows that the model shows different learning behaviors in clean and poisoned subsets during training. Based on this observation, we propose a general training pipeline to defend against backdoor attacks actively. Benign models can be trained from the unreliable dataset by decoupling the learning process into three stages, i.e., supervised learning, active unlearning, and active semi-supervised fine-tuning. The effectiveness of our approach has been shown in numerous experiments across various backdoor attacks and datasets.

Abstract Recently, deep neural networks have been shown to be vulnerable to backdoor attacks. A backdoor is inserted into neural networks via this attack paradigm, thus compromising the integrity of the network. As soon as an attacker presents a trigger during the testing phase, the backdoor in the model is activated, allowing the network to make specific wrong predictions. It is extremely important to defend against backdoor attacks since they are very stealthy and dangerous. In this paper, we propose a novel defense mechanism, Neural Behavioral Alignment (NBA), for backdoor removal. NBA optimizes the distillation process in terms of knowledge form and distillation samples to improve defense performance according to the characteristics of backdoor defense. NBA builds high-level representations of neural behavior within networks in order to facilitate the transfer of knowledge. Additionally, NBA crafts pseudo samples to induce student models exhibit backdoor neural behavior. By aligning the backdoor neural behavior from the student network with the benign neural behavior from the teacher network, NBA enables the proactive removal of backdoors. Extensive experiments show that NBA can effectively defend against six different backdoor attacks and outperform five state-of-the-art defenses.

With the digital transformation of the energy industry, energy blockchain is playing an important role in application areas, such as energy data sharing and distributed power trading. In this process, the use of energy data is a top priority. Federated learning (FL) can enable the analysis and computation of energy data while protecting their privacy. However, traditional FL relies on a central server and parties involved are not fully trusted. In energy blockchain environment, FL also faces data poisoning attacks launched by energy departments, besides, the supervisory committee carrying out checking models can launch deception attacks. Therefore, we propose a game theory-based incentive mechanism for collaborative security of FL in energy blockchain environment, which can discourage nodes from taking malicious behaviors in iterative training of FL. First, we propose an FL model in energy blockchain environment, which can protect privacy and achieve collaborative security. Considering that game theory can be used to analyze the strategies of participants, we build a game model with energy departments and supervisory committee as players and design our incentive mechanism based on game theory, which is implemented by smart contracts. Even if the accuracy of model checking algorithm is low, malicious behaviors in FL can be reduced by using our incentive mechanism. In particular, we prove that our mechanism can lead game model to a Nash equilibrium (NE) that achieve collaborative security. Security analysis and experimental evaluation show that our incentive mechanism is feasible in energy blockchain with robustness, reliability, and low complexity.

As a key technology for the development of 5G networks, network slicing is developing rapidly. Although network slicing can realize the flexible division of 5G network resources and quickly customize virtual networks that meet the differentiated needs of customers, it is still difficult to determine the optimal service quality parameters in application scenarios. To solve the problem, this article designs a multichain 5G network slicing service quality computing model to calculate the service quality parameters of the network slicing. The calculated service quality parameters can be used as an adjustment basis in the negotiation of the SLA between the customer and the network operator. However, the traditional method of calculating information across chains will cause frequent information interactions and affect efficiency. Therefore, in this scheme, we deploy a smart contract on each blockchain to calculate the information, which can reduce the frequency of information transmission and improve efficiency. In addition, in order to make the calculation between smart contracts more fluent and the requirements more relevant, this article proposes to coordinate the development of smart contracts through multiple blockchains. Besides, to ensure the cross-chain security calculation, the signature by Cosi protocol and multisigncryption algorithms are used in the transmission of nonprivate information and private information in the cross-chain process, respectively. Security analysis and experimental results prove that the multichain 5G network slicing service quality computing model is feasible and efficient in practice.

With the development of electric vehicles, the shortage of charging piles (CPs) has gradually been exposed. In response to this situation, CP operators have taken private CPs into the shared charging system. Due to the lack of maintenance personnel for private CPs that join shared charging, users often face the problems of damaged CPs and poor service attitudes of CP owners. Reputation solutions based on third-party platforms face a problem of single-point failures and reputation solutions based on blockchain face problems of storage and query efficiency. To improve storage and query efficiency, this article proposes a multichain charging model that stores different types of information on different blockchains. However, it faces the problem of unreliable information called across chains, when calculating reputation across chains. Therefore, this article proposes a cross-chain trusted smart contract ($C_{2}T$smart contract) to ensure the authenticity, real-time, and interchain write mutual exclusion of cross-chain information, making reputation calculation in the multichain charging model more convenient and more accurate. Especially, we propose a data mutual trust mechanism based on Merkle proof to ensure the authenticity of cross-chain information and prevent forged information from participating in calculating reputation. Furthermore, we present a data structure composed of multiple counting Bloom filters (MCBFs) to verify the real time of information and filter out non-real-time information, thereby ensuring the real time of the calculated reputation. In addition, we put forward an algorithm to guarantee the interchain write mutual exclusion by hash mutexes, making the reputation calculation process more accurate and complete. The security analysis and experimental results demonstrate that$C_{2}T$smart contract is feasible in practice.

With the proliferation of location-aware devices, trajectory data have been used widely in real-life applications. However, trajectory data are often associated with sensitive labels, such as users’ purchase transactions and planned activities. As such, inappropriate sharing or publishing of these data could threaten users’ privacy, especially when an adversary has sufficient background knowledge about a trajectory through other data sources, such as social media (check-in tags). Though differential privacy has been used to address the privacy of trajectory data, no existing method can protect the privacy of both trajectory data and sensitive labels. In this article, we propose a comprehensive trajectory publishing algorithm with three effective procedures. First, we apply density-based clustering to determine hotspots and outliers and then blur their locations by generalization. Second, we propose a graph-based model to efficiently capture the relationship among sensitive labels and trajectory points in all records and leverage Laplace noise to achieve differential privacy. Finally, we generate and publish trajectories by traversing and updating this graph until we travel all vertexes. Our experiments on synthetic and real-life datasets demonstrate that our algorithm effectively protects the privacy of both sensitive labels and location data in trajectory publication. Compared with existing works on trajectory publishing, our algorithm can also achieve higher data utility.

By decoupling the control and data planes, Software-Defined Networking (SDN) can implement centralized manage-ment on the network. With the increasing scale of the network, the multi-controller SDN architecture is becoming more and more popular, because it can handle what SDN with a single controller is not able to address. However, the controller load imbalance may happen due to traffic dynamics in an SDN with multiple controllers, which results in congestion at a certain controller and seriously affects the scalability of the control plane. Though switch migration is an effective solution to this problem, how to migrate the traffic is an NP-hard problem. In this work, we propose a switch migration scheme based on deep Q-learning (DQN) by combining the powerful perception of deep learning with the decision-making ability of Q-learning. We first describe the SDN state formally. Then, the network state is represented by the two-dimensional array as the input of the Q network. The network features are extracted through the convolution layer, and the full connection layer is achieved. Finally, the output layer to predict the migration action in some states of a network is extracted. After the migration action is performed, we will get an instant reward or penalty. We implement our algorithm based on the keras deep learning framework and compare it with the classic Q-learning algorithm. The results show that our scheme is superior to the traditional method in terms of resource utilization and load balancing ability.

Security attacks in smart contracts have drawn extensive attention due to the financial loss and erosion of trust caused by vulnerabilities. Even worse, smart contract is a tamper proof digital agreement and fixing bugs in it is difficult, so it is necessary for developers to detect security vulnerabilities in smart contract before deployment. Researchers have proposed several methods on smart contract vulnerability detection. However, despite the numerous vulnerability-finding tools, few of them have ideal detection performance because most of them rely on fixed rules, which is inefficient. In this paper, a modularized and systematic Deep Learning-based framework is proposed to automatically detect smart contracts vulnerability, called DeeSCVHunter. Particularly, we focus on two types of smart contract vulnerabilities: reentrancy and time dependence. And we propose a novel notion of Vulnerability Candidate Slice (VCS) to help models capture the key point of vulnerability. We conduct experiments on real-world dataset and the results prove the effectiveness of VCS, which leads to 25.76% improvement in F1-score at most. And extensive experiments also show that our proposed framework significantly outperforms state-of-the-art methods.

With the development of the Energy Internet and the support of the subsidy policies of various countries, Electric Vehicles(EVs) have ushered in a golden development period. However, the development of EVs needs to solve the problems of insufficient charging piles(CPs) and difficulty in finding CPs. In order to solve the problem of difficult charging of EVs, the concept of shared charging came into being, in which idle CPs or private CPs are shared to meet the charging needs of more people and improve the utilization rate of CPs. However, the shared charging scheme implemented by third-party platforms faces the issue of trust lacking. This paper proposes a blockchain architecture for shared charging, which can use the blockchain to build a trust environment involving private pile owners, charging pile(CP) operators, Electric Vehicle(EV) users, etc.. The blockchain architecture also contains the block structure where pointer was added for quick search, contract content that can automatically execute multi-party contracts to achieve secure computing and reputation-based incentive mechanism to provide high-quality charging services in detail. This architecture establishes the multi-party trust environment for shared charging from three aspects: secure storage, secure computing, and secure incentives.

Comment spam is one of the great challenges faced by forum administrators. Detecting and blocking comment spam can relieve the load on servers, improve user experience and purify the network conditions. This paper focuses on the detection of comment spam. The behaviors of spammer and the content of spam were analyzed. According to analysis results, two types of effective features are extracted which can make a better description of spammer characteristics. Additionally, a gradient boosting tree algorithm was used to construct the comment spam detector based on the extracted features. Our proposed method is examined on a blog spam dataset which was published by previous research, and the result illustrates that our method performs better than the previous method on detection accuracy. Moreover, the CPU time is recorded to demonstrate that the time spent on both training and testing maintains a small value.

This paper aims at the problem of anomaly user detection, in which a novel, effective and comprehensive feature extraction method is proposed. Instead of existing keystroke timing information from the dataset, three types of new features are extracted in our approach for a better description of user keystroke characteristics. Moreover, an AdaBoost based algorithm is used to generate an optimized anomaly user detection model based on comprehensive keystroke features. This model contains the optimal weights of multiple alternative weak classifiers acquired through training. What's more, it effectively integrates several types of features in the form of combinations, each of which describes different inherent characteristics of keystroke processes. Both the proposed features and method are examined on CMU keystroke dynamics benchmark dataset which ensures the comparison on same grounds. Through extensive comparisons on the same dataset, the results illustrate that the proposed method has a great performance against the state-of-the-art techniques in terms of average equal error rate (EER). Furthermore, the CPU time during our optimized model generation and testing were recorded to demonstrate that the time spent is acceptable.