Leveraging the advanced capabilities of Deep Neural Networks (DNNs), attackers can precisely detect users' online activities through traffic monitoring, nullifying the efficacy of current encrypted communication tools/protocols and progressively resulting in privacy leakage. Several defensive methods against DNN-based traffic monitoring (DTM) have been proposed; however, these methods often rely excessively on prior knowledge and incur inevitable additional bandwidth overhead (BWO). Moreover, they frequently generate invalid packets that violate network transmission constraints. To address these drawbacks, in this paper, we propose BYTEFLIPPING, a byte-space grey-box defensive method, which perturbs vulnerable bytes in the transport layer payload to generate adversarial sample packets. We design a Payload Byte Vulnerability Ranking algorithm to pinpoint the most vulnerable bytes and based on this generate adversarial packets to defend DTM. Extensive experiments reveal that ByteFLIPPING performs well in protecting against three DTM methods across two benchmark datasets, significantly decreasing the accuracy of the state-of-the-art ET-BERT by 94%. Compared to baseline defensive methods, BYTEFLIPPING incurs no extra BWO, offers more dependable packet validity, and boasts greater feasibility.

To protect the confidentiality of communication data, internet users often use encryption protocols (e.g., TLS/SSL) or tools (e.g., VPN, Tor) for network access. Therefore, as a pivotal network management method, encrypted traffic classification technology is vital for guaranteeing the quality of service, the quality of experience, and network security. Researchers have already developed some end-to-end deep learning-based methods to realize encrypted traffic classification. However, given the constrained computational resources available in real-world network measurement scenarios, the existing approaches with high complexity and computation overhead are not appropriate. In this paper, we propose a lightweight model to tackle this issue. Firstly, we propose a base model based on the self-attention mechanism to obtain the key bytes in packets contributing to the classification. Secondly, we leverage these key bytes to reconstruct the input and then streamline the base model, carrying out a lightweight model, i.e.,$\mathcal{L}{-}$ETC. Finally, we implement experiments on three benchmark datasets.$\mathcal{L}-\mathbf{ETC}^{\prime}\mathrm{s}$macro Fl score of the three tasks exceeds 0.92 with only 0.076M (Million) parameters, and the throughput reaches 917 pps, which is also superior to state-of-the-art methods.

Covert communication technology serves as a crucial tool for safeguarding not only the content of communication but also the identities of the parties involved. In this regard, blockchain emerges as a promising solution due to its decentralized nature, flood propagation of data, and inherent anonymity features. This makes blockchain an ideal candidate for covert communication channels, effectively addressing the weaknesses associated with traditional covert communication methods susceptible to detection, tracing, and interruption. However, the current efforts encounter obstacles like limited practicality, constrained channel capacity, and insufficient concealment capabilities, impeding their broad adoption in real-world scenarios. To address these issues, we propose DarkTrans, a blockchain-based covert communication scheme consisting of an address binary tree and a novel embedding mechanism. The address binary tree as a dynamic label method enables rapid recognition of specific transactions by the recipient, rendering detection by third parties challenging. The embedding mechanism encodes secret messages into transaction values for transmission to augment channel capacity, which can be practically realized within an Ethereum private blockchain. Our experiments with three aspects demonstrate that, compared with the existing scheme, DarkTrans achieves a low embedding time and a high channel capacity. Additionally, Kolmogorov-Smirnov test and sample entropy analysis are conducted to validate the robust concealment of this scheme.

The privacy protection capabilities of open source proxy software (OSPS) while browsing the Internet have sparked great interest from both industry and academia, bringing forth pressing security concerns. Currently, using artificial intelligence for traffic identification is the most promising direction. Due to the wide variety and rich configuration of OSPS, it is not feasible to train models for all tasks and deploy them on the same network device. It is a novel idea to improve efficiency by leveraging multi-task learning. However, the related studies still have three shortcomings. First, improving the performance of the main task through auxiliary tasks does not apply to equally important OSPS identification tasks. Second, the model’s ability to characterize traffic is weak, resulting in performance gaps between different tasks. Finally, the influence of task difficulty on convergence speed is ignored, which is easy to cause overfitting and underfitting. Aiming at the shortcomings, we propose Cerberus, an OSPS traffic identification scheme based on multi-task learning. The main contributions of our work can be summarized in three aspects. Firstly, a high-quality dataset is constructed through traffic collection, and three OSPS traffic identification tasks are defined on it. Secondly, an identification model is designed by optimizing the ability to characterize traffic and balancing the convergence speed of multiple tasks. Finally, Cerberus is verified through comparative experiments. Its classification performance is better than both single-task and multi-task solutions. Besides, Cerberus runs fast and consumes few resources, making it suitable for deployment on network devices.

Consortium blockchains have been widely used in many industries such as medical care, finance, and so on. The business data on different blockchains are isolated from each other. In order to share the value data, it is necessary to achieve cross-chain access. Existing cross-chain technologies are mainly aimed at public blockchains, while consortium blockchain users hold identity credentials to participate in transactions, which has higher security requirements. In addition, most of the existing consortium blockchains use pluggable cryptography components. If all consortium blockchains participating in the cross-chain are required to be configured with the same cryptosystem, the cost will be too high. To solve the above privacy protection and compatibility issues, we propose a privacy-preserving complete cross-chain authentication scheme and name it ChainPass. Chain-Pass introduces paillier homomorphic encryption and pseudonym technologies, allowing users to use the original cryptosystem to generate public and private keys and participate in cross-chain transactions. At the same time, the user’s pseudonym is updated according to the user’s public key. ChainPass protects users’ privacy while ensuring complete compatibility. Security analysis and performance evaluation prove that ChainPass has better security and higher efficiency.

Spam has harassed Internet users for a long time, and how to detect spam accurately and efficiently is a critical problem. As yet, there are lots of research works proposed to detect spam, e.g., black and white lists, machine learning methods, and deep learning content-level measures, etc. Based on previous works, we find that most of methods' accuracy can reach 0.95 when they focus on one type and one language spam. Nevertheless, nowadays, people will receive spam messages of different types, different sources, and even different languages. Toward this, we develop a novel model, which is based on Google multilingual bidirectional encoder representations from transformers (M-BERT). Meanwhile, we design a brand new bilingual multi-type spam dataset to train our model. Particularly, we utilize optical character recognition (OCR) to extract text from image-based spam. Through the experiment, we find that the proposed model's accuracy can reach 0.9648, which outperforms the comparison models. In terms of time overhead, the proposed model only costs 0.3168 seconds per training step, which is an acceptable overhead. Therefore, these analysis results demonstrate that our approach can detect bilingual multi-type spam effectively.

The high-precision maps can provide additional information on roads and conditions, which plays an important role in autonomous vehicles (AVs) navigation. Compared with the existing map update methods, the real-time map updates based on crowdsensing have lower cost and higher accuracy. However, in the process of map update, the map service platform (MSP) cannot recruit enough vehicle users to obtain the sensing data due to a lack of incentive mechanism. Therefore, how to motivate more vehicle users to provide high-quality sensing data is the key for real-time map updates. In this article, we propose a secure and privacy-preserving incentive scheme for reliable real-time map updates, named SPIR. Specifically, under the condition of limited service platform budget and limited vehicle user's ability, an effective incentive mechanism based on reverse auction is presented, which can solve two core problems: i.e., payment control for MSP and completion quality for vehicle users. Meanwhile, a credit management and payment system based on the blockchain technique are designed. In addition, the partially blind signature technique is applied to guarantee the security of the incentive mechanism and protect the privacy of vehicle users. Both theoretical analysis and simulation results indicate that the proposed SPIR achieves near-optimal benefits, which can provide the fair reward for vehicle users and reasonable budget for the MSP. In the real-time map update services, SPIR can guarantee the computational efficiency and data reliability.

Traffic-related messages exchange (TME) is considered as a powerful approach to improve traffic safety and efficiency in vehicular networks. However, TME assumes all vehicles always are honest, and thus offering opportunities for attackers to fake traffic-related messages. To combat such threat, recent efforts have been made to trust mechanism. In this article, a vulnerability for trust mechanism is found, that is, the ratings from initiator vehicles (IVs) are generally unchecked. Such ratings corresponding to the truth of traffic-related events can be exploited by attackers to disturb trust mechanism. Specially, attackers would form a clique to help with each other in an inside-and-outside collusive (IOC) manner. One of the IOC attackers can disguise as an IV who sends the rating in accordance with the traffic-related messages of his conspirators, result in promoting their trust value quickly. With high trust value, attackers can escape the detection of trust mechanism. We conduct an in-depth investigation on IOC attack and propose a defense scheme called TFAA from the design ideas of trust fluctuation association analysis. In addition, the trust data management of central and distributed trust mechanism may be unsuitable for vehicular networks. To support the trust data management for the TFAA scheme, we also design a semi-distributed trust data storage scheme called TruChain with the combination of consortium blockchain and vehicular regions partition. The simulation results show that the TFAA scheme can enhance the accuracy of trust value evaluation, and thus successfully reducing the power of IOC attack against TME.