UAV-based steppe rathole monitoring faces severe challenges due to the tiny scale and extreme spatial sparsity of targets. Conventional label assignment methods, relying on local dense sampling assumptions, often struggle to distinguish isolated ratholes from high-confidence clustered background noise, leading to high false-positive rates. To address this, we propose a dual-enhanced label assignment framework. First, a Centripetal-Scale Coupling Distance (CSCD) is developed to reinforce centripetal constraints via a multiplicative gating mechanism, enhancing sensitivity to minute positional offsets. Second, a Sparsity-Guided Dynamic Top-K (SGTOP) strategy is introduced to suppress clustered texture-induced false alarms by incorporating environmental sparsity priors. Experimental results on the RatHolev2 dataset demonstrate that our framework significantly boosts detection performance across various mainstream models while maintaining high computational efficiency.

Task offloading can significantly enhance task processing efficiency and reduce latency. However, current research often relies on fixed offloading devices, which struggle to respond quickly to sudden events or changing demands in dynamic scenarios such as disaster relief or emergency response. The mobility of Unmanned Aerial Vehicles (UAVs) enables rapid deployment to various locations to assist with task offloading; however, limitations in battery life and computational power, along with their inherently flexible nature, complicate the effective formulation of offloading strategies in dynamic environments, resulting in persistent latency issues. To address this challenge, we propose an improved algorithm–an enhanced method combining Long Short-Term Memory (LSTM) networks with Deep Deterministic Policy Gradient (DDPG) (LSDDPG). This algorithm achieves real-time optimization of task offloading decisions by continuously learning the dynamic features of system states and action variations. Specifically, LSDDPG generates superior action strategies based on the current state and historical data, learning from past states and actions to provide a more objective basis for decision-making. This approach effectively alleviates the constraints of UAVs in terms of power, computational capability, and mobility, significantly reducing task processing latency. Experimental results demonstrate that our proposed LSDDPG algorithm outperforms existing solutions in terms of task offloading latency, highlighting its potential for application in dynamic environments.

Heterogeneity among nodes is very common in blockchain networks. Most of the existing methods ignore the positive impact of heterogeneous information. However, this paper argues that heterogeneous information plays a key role in capturing complex interaction patterns between different account types and improving the classification effect of the model. Therefore, this paper proposes a Heterogeneous adversarial feature enhancement model (HAFEM), which screens out heterogeneous neighbor nodes through the similarity node ranking strategy, and combines the adversarial feature enhancement module to optimize the feature generation. Experiments show that HAFEM exhibits comparable or even better performance than baseline methods using only homogeneous neighbor nodes on the Ethereum account classification task.

As a representative of the public blockchain, Ethereum has been applied in various industries. However, the vast number of transactions on the platform has also brought a number of illegal activities, such as phishing scams, which have caused significant damage to the Ethereum ecosystem. Due to anonymity of the blockchain, it is difficult for detectors to extract features that can be directly applied to phishing scams detection. Existing studies mainly model Ethereum transaction records as a network and mine key information from them to identify phishing addresses. However, these methods usually employ traditional feature engineering or network embedding, ignoring the fine-grained features in the transaction network. In addition, since the original network is too large to make learning difficult, existing work usually uses random walk (RW) to sample a part of nodes for training, thus ignoring the multiplicity of the network. To address these issues, in this article, we propose a three-stream feature fusion (TSFF) approach to enhance the feature representation of nodes. Specifically, we construct node states to guide RW sampling, and manually extracted 8-D features from the resulting dataset as basic features. Temporal features are jointly learned through long short-term memory network and contrastive learning. We combine residual blocks and graph convolutional network to extract fine-grained structural features from transactional networks. Finally, we fuse these three types of features and input them into a downstream classifier. Experiments show that our TSFF (85.3% Precision) outperforms the state-of-the-art methods, and the effectiveness of each feature is demonstrated.

Cryptocurrency crime incidents in Ethereum are continuously rising, with phishing scams accounting for 50% of all criminal activities. The severe data imbalance significantly impacts the performance of Ethereum phishing detection models. The current solution may introduce redundant information or lead to the loss of important data. In this paper, we propose an Ethereum phishing detection method based on Graph Contrastive Learning with augmentations. This approach addresses the issue of insufficient learning of phishing node features, thus alleviating the influence of data imbalance on the model’s detection performance without disrupting the original data distribution. To enhance the representation of structural features, we employ two data augmentation methods: feature masking and edge perturbation. We conducted extensive experiments on a real Ethereum phishing dataset to evaluate the performance of our method. Compared to alternative methods, our approach not only significantly improves Precision, ranging from 12% to 30%, but also achieves noticeable enhancements in Recall, Auc, and F1-score. The experimental results provide ample evidence of the effectiveness of the proposed method.

The rise of Ethereum in various economic and social domains has made it a prime target for illegal activities, particularly phishing scams, which has caused substantial financial losses. Existing methods mainly model transaction records as networks and classify nodes. However, a particular challenge is that not all transactions involved in phishing nodes are illegal, which makes phishing detection very difficult. In order to address the problem, in this paper, we propose a Curriculum Learning-based method for Ethereum phishing detection. We incorporate the Local Outlier Factor to measure the difficulty of nodes, considering the significant feature differences among nodes of the same class. By assigning lower difficulty values to easily identifiable nodes and higher difficulty values to nodes involved in mixed-class transactions, we ensure an effective difficulty measure. Then we gradually increase the number of training nodes input into Graph Convolutional Network in each epoch based on a certain ratio determined by the sorted difficulty scores. Finally, we employ LightGBM as the classifier for identifying phishing nodes. Experimental evaluations on a real-world Ethereum phishing scam dataset demonstrate the superiority of our method over baseline approaches, as evidenced by several evaluation metrics.

Data lineage technology is a method of describing the relationships between data, which plays a crucial role in solving many challenges in the data marketplace, such as unauthorized data redistribution, data tampering, and false data transactions. However, the complexity of the multi-chain data marketplace, including factors such as data credibility, trust environment, data ownership, and relationship complexity, poses significant challenges to construct data lineages. To effectively address these challenges, we propose a data lineage construction method for multi-chain data marketplaces. This method mainly involves the following three steps: Firstly, map data assets to Data Non-Fungible Token (DataNFT) and use referable NFT (rNFT) to record data lineage. Secondly, when data assets require cross-chain transfer, the transfer message of DataNFT is broadcasted through the interchain NFT protocol, without the need for actual cross-chain transfer. Finally, we add weights to the data lineage link, enabling us to quickly locate problematic data based on weight sorting during data auditing. We have implemented a system prototype of this method and verified its effectiveness through experiments. The experimental results show that our proposed scheme not only ensures the correctness and completeness of data lineages, but also effectively reduces audit costs.

Phishing scams have become one of the primary frauds on Ethereum, leading to substantial financial losses for users. Therefore, developing an effective phishing detection method for Ethereum holds significant importance for the blockchain ecosystem. However, the heterogeneity of Ethereum poses challenges when it comes to detecting phishing scams. In this paper, we propose a Curriculum Learning-based approach (CLPSD) for Ethereum phishing detection. We collect transaction records to build a graph. In order to distinguish the varying degrees of contribution among nodes, we design a difficulty measurer combining Local Outlier Factor and information entropy. Thus, CLPSD utilizes Graph Convolutional Network to initially learn from easy samples and progressively advance to more complex ones. The experimental results demonstrate that CLPSD outperforms existing methods, highlighting the superior performance of our approach and its significant relevance in enhancing the Ethereum ecosystem.

In blockchain-based data marketplaces, the cross-chain solution based on relay chains has become an effective way to support digital asset transfer between heterogeneous blockchains. If a digital asset is popular, it will be frequently transferred between multiple data marketplaces. Digital assets are typically represented as (Non-fungible Tokens) NFTs. However, when utilizing a cross-chain transfer scheme based on a relay chain that involves locking, unlocking and burning of the NFT, it can lead to significant transaction costs. Due to the frequent minting and burning of popular NFTs, they tend to consume a significant amount of gas, resulting in high transaction costs. To address this issue, we propose the DART scheme, a secure and low-cost cross-chain transaction scheme for popular digital assets based on (dynamic NFT) DNFT and relay chain. The scheme mainly includes three aspects: Firstly, the initial owner of (Data NFT) DataNFT locks it onto the gateway contract on SChain, mints a (Cross-chain NFT) CNFT on the relay chain that references the DataNFT for resale, and minted DataNFT on the destination blockchain until the CNFT is no longer being resold. Secondly, the DNFT standard is utilized to implement CNFTs. The DNFT standard records the transaction hash of each CNFT resale, ensuring the secure reselling of DataNFTs. Thirdly, the metadata of DataNFT is encrypted to ensure its security during cross-chain transmission. Finally, we selected Axelar Network as the relay network and successfully implemented a prototype of the system. The experimental results demonstrate the feasibility of our proposed scheme.

With the development of 5G networks, quantum networks, IoT, and smart cities, network information security and trust appear new challenges. False messages, malicious devices, communication vulnerabilities, etc., which threaten network security, the trust status of nodes need to be identified, and trust needs to be measured. Access control, data encryption, identity authentication, electronic signature, etc., are a series of network security technologies, however, they are powerless in the face of the large and complex network information. This paper proposes a kind of model based on machine learning, which utilizes the miscellaneous network information of nodes, to be applied to the trustworthiness evaluation, trust management of nodes, and detection of malicious nodes. First of all, this paper introduces the designed principles-assumptions model, and its parameter optimization mechanism, i.e., the gradient descent algorithm. Secondly, this paper carries on the simulation experiment and analysis of the model, including the three trust evaluation mechanisms, the generation of the data set, the training, testing, and evaluation of the model, the resistance analysis of malicious attacks, and the comparison-based performance evaluation. The experimental results show that under malicious network attacks, when the number of malicious nodes is 45%, the F-score of the principles-assumptions model is more than 95%. The ability of the model to detect malicious nodes is excellent, and can be applied to the trust evaluation and management of network nodes.

Trust management of network nodes can protect the security of IoT, e.g., detection of false messages and malicious nodes, optimizing IoT services, etc. Trust evaluation and trust prediction are the core contents of trust management, which are used to evaluate the trustworthiness of nodes. However, the lack of trust-related data will lead to the failure of the activation of trust evaluation mechanisms. The update of trust needs to be solved by trust prediction schemes. This paper designs a trust prediction scheme, utilizing neural networks and mathematical calculations to realize the classification and prediction of trust iterations. First of all, this paper gives the detailed architecture of the designed scheme, including the neural network model and data preprocessing. Then analyze the theoretical calculations of related variables in trust iterations, e.g., mean value, variance, etc. Finally, utilizing simulation experiments to verify the performance of the designed trust prediction scheme, including conserved quantity in trust iterations, dataset preprocessing, testing of the scheme, malicious attack resistance analysis, etc. The experiments prove that the scheme can resist related malicious attacks, e.g., data tampering attacks, etc. Moreover, the residual error of trust prediction is not more than 0.0075, which is better than the existing trust prediction schemes.

Blockchain has facilitated the growth of cryptocurrencies but has also provided new ideas for illegals to commit fraud. Research on malicious accounts detection shows that the number of malicious accounts is much smaller than that of benign accounts, leading to imbalanced dataset samples. Most researchers adopt the under-sampling method to help deal with this issue, but this method does not correspond to the actual scale. So, we propose an anomaly detection method based on community discovery. Firstly, we use the transaction information in the Ethereum public chain to build a transaction network and use the Louvain algorithm to divide the transaction network into communities. Secondly, we use the LightGBM algorithm to classify the community. Finally, based on the classification results, we use HBOS, LOF, K-Means, KNN and iForest algorithms as benchmark algorithms for anomaly detection and compare the experimental results using the methods in this paper with the results of anomaly detection using the original transaction network. Experimental show that our method can reduce the amount of data by 35.53% and increase the AUC values of the five algorithms by 7.52%, 8.41%, 14.88%, 0.83% and 27.95%.

The Vehicle Named Data Networking(V-NDN) is a vehicular ad-hoc network with the Named Data Networking(NDN) as the architecture, and the most advantageous feature is the in-network cache, which caches the content in the intermediate nodes of the network and can quickly satisfy the requests of subsequent consumers for the same content. Since the cache space of nodes is limited, the cached content should be the popular content frequently requested by users in the network, so the most important problem is accurately finding out the future popular content in the network. This paper designs a multi-feature content popularity prediction algorithm to address this problem based on the attention mechanism and GRU (GRU-Attention). According to the characteristics of multiple historical requests for content, the GRU-Attention model is used to predict the future popularity of content. Through experimental verification, the content popularity prediction algorithm proposed in this paper effectively improves the accuracy of prediction.

With the explosive development of technology and Internet communication, it has become an inevitable trend to realize the secure storage and sharing of electronic medical data among hospitals. In recent researches, there are also many problems in realizing secure storage and sharing of electronic medical data, such as "data silos", leakage of patient sensitive information due to data sharing and having no reliability about the original data uploaded by patients. To solve the above problems, we propose a blockchain-based medical data storage and secure sharing scheme. In the scheme, we utilize IPFS-based Web3.Storage for medical data storage, propose a sensitivity classification and access control strategy for sensitive data leakage and present a blockchain-based original data reliability checking strategy to check the reliability of the original data. Our scheme is explained in detail in the paper, and the performance analysis of this scheme is carried out to prove the feasibility of this scheme.

With the rapid development and popularization of block chain technology, some social platforms based on block chain technology have emerged. Due to the characteristics of block chain technology, the social behaviors and ways of users on the network have been affected, thus changing the process of information transmission. Based on the characteristics of information transmission in blockchain social networks, this paper proposes a new information transmission model by comprehensively considering the influence of factors such as blockchain incentive mechanism, node intimacy in traditional social networks and individual differences on the state transition probability. A new voting node has been added to the model, representing users on the blockchain network who are affected by the incentive mechanism. In the simulation experiment, by adjusting the weight of each factor, the influence on the number of spreaders in the information transmission model is analyzed. Moreover, by comparing the model proposed in this paper with the traditional social network model, the number of propagation nodes is reduced by 17.32%. The experimental results show that the blockchain environment can effectively curb the spread and spread of false news, and achieve a good ecological network public opinion communication environment.

V-NDN (Vehicular Named Data Networking) applies completely new communication mechanism to the Internet of Vehicles, which can effectively improve the efficiency of vehicle communication and reduce communication overhead. However, the dynamic characteristics of the vehicle network topology cause malicious nodes to join the network and send fake messages to normal nodes, which brings many security challenges to the reliable communication between vehicles in the V-NDN. In this paper, we propose a method based on machine learning to detect fake messages and malicious nodes. The proposed method simplifies the detection process into a classification process and combines with a variety of machine learning algorithms such as logistic regression, KNN. It can intelligently identify fake messages and malicious nodes in the V-NDN. The simulation results show that the method can effectively increase the accuracy and efficiency of identifying fake messages and malicious nodes and improve the security in the V-NDN.

The Vehicular Named Data Networking (V-NDN) improves the speed of message acquisition between vehicles and reduces network overhead by using a in-network caching mechanism. The vehicles in V-NDN have the capability of built-in caching, in other words, they can cache contents passing by and provide content services for users. However, malicious nodes in V-NDN might apply fake messages for malicious purposes, which is one of the major risks of network security. In this paper, we build a trust management mechanism based on blockchain to solve the above problems. In the proposed mechanism, vehicles first judge the credibility of the received message based on the vehicle reputation value and the feature of the message itself. Then the vehicle reputation value is updated according to the message credibility. Finally, the blockchain is used to realize the consensus of the message credibility and the vehicle reputation value. We conduct experiments on the simulation platform and simulation results show that the proposed mechanism can effectively improve the accuracy of message credibility judgment and malicious vehicles detection, thereby improving the security of the V-NDN.

Nowadays, online shopping has gradually become the mainstream, and the importance of user comments has gradually increased. Due to the operation of “playing” and “controlling comments”, the fake “brainless blowing” comments that overtout the products endorsed by stars have poured out in large numbers. These fake reviews are easy to lead users to make wrong purchase decisions, damage the interests of users, and also easily cause nonfan group users’ aversion, affecting the promotion and sales of merchants. Therefore, it becomes very important to identify fake comments from a large number of comments, but it is very difficult for users to identify fake comments quickly and accurately in a large number of comments. In response to this problem, this paper collects the semantic and emotional characteristics of the text content of user reviews, combines the behavior characteristics of users when making purchase decisions, and builds a false recognition model for user reviews based on LSTM (Long Short-term Memory) to assist users in identifying false reviews. So that users can make accurate decisions and purchase products that meet their needs. Based on the F1 value, accuracy rate, and recall rate as evaluation indicators, after comparison experiments with eight commonly used models, the algorithm in this paper has the best effect.

The miscellaneous users, devices, and information in the network bring a great threat to the security of network information. Compared with access control, data encryption and so on, trust management is a technology that can utilize many factors to measure the dynamic trust relationship between nodes. Trust propagation and trust evaluation are the contents of trust management, and trust propagation refers to the spread of trust relationships between nodes. This paper designs a trust propagation mechanism based on information entropy theory. First of all, this paper describes the trust relationship between nodes, including the normalization of trust value, series propagation, and parallel propagation, as well as the corresponding change principles of information entropy in the process of propagation. Then, this paper designs the increasing algorithm and decreasing algorithm of information entropy, and utilizes experiments to determine the values of related parameters. Finally, through the simulation experiment of the dynamic relationship between nodes, this paper verifies the effectiveness of the designed trust propagation mechanism. The experimental results show that the designed information entropy increasing algorithm and decreasing algorithm can be applied to trust propagation to measure the trust relationship between any node. Utilizing the trust propagation mechanism, to quantify the relationships between any non-adjacent nodes in the knowledge graph, and enhance the traversal and search of knowledge and their relationships.

Nowadays, data are regarded as an intangible asset. In the data marketplace, data are treated as a commodity or service, data owners can gain economic revenue by selling their data ownership or data usage rights. However, the data marketplace is facing various threats and challenges, such as unauthorized data reselling, trade of bogus data, dishonest data ownership claims, and unreasonable revenue allocation. Particularly, in the process of data resale, the revenue allocation remains a challenge when the data are processed and resold in another format. To solve this problem, we propose a revenue allocation mechanism based on the referable Non-Fungible Token (rNFT) and Shapley value method. Firstly, we tokenize the data to NFT to ensure the data ownership is traceable; Secondly, we use rNFT to record the data lineage that ensures the data owner can participate in the revenue allocation when the data are resold; Finally, we calculate the contribution of each party by the Shapley value method to ensure fairness in revenue allocation. We implement a prototype of our scheme on Ethereum and evaluate it comprehensively. The test results indicate that our scheme can meet the performance requirements of the data marketplace and improve the revenue of data owners effectively.

Compared with other security technologies of the Internet of Things (IoT), trust management (TM) can accommodate sudden or unpredictable changes brought about by ethical and moral standards. First of all, this article introduces the basic contents of TM, including objectives, overall architecture, threat model, related indicators, application scenarios, and so on. Then, this article develops four dimensions: 1) trust-related data (TRD); 2) trust evaluation mechanisms; 3) trust propagation mechanisms (TPMs); and 4) trust influence mechanisms (TIMs), and describes the main contents. There are 4 kinds of TRD and 12 kinds of trust evaluation mechanisms, including trust evaluation mechanisms meeting reinforcement learning. There are five kinds of TPMs and six kinds of TIMs. Based on four dimensions, the existing TM is classified and summarized. Finally, this article gives the current research challenges and future research directions in each dimension. In addition, the modeling process based on 4-D TM is given in the conclusion. In this article, the TM model is based on the mass points in the 4-D space, and the mapping on each dimension is specific data or methods, which can be selected. If the specific content of each dimension is selected, the TM model will run according to a specific process in a specific IoT environment. The 4-D trust management model helps to select more appropriate data or methods to effectively adapt to the dynamic and complex environment of the IoT.

The blockchain technology in the Internet of Vehicles (IoV) can be applied to solve security problems. Due to the high dynamics and resource constraints of the IoV, the traditional single-chain blockchain system is not suitable for the practical scenarios. In our previous work, we proposed a lightweight blockchain architecture based on the directed acyclic graph lattice (DAG-lattice) structure for the IoV. However, the current consensus algorithms are not suitable for the previous architecture mentioned above. Therefore, this article proposes a parallel consensus mechanism based on the DAG-lattice structure in the IoV using practical Byzantine fault tolerance (PBFT) as consensus algorithm. This mechanism solves the problem of low efficiency of PBFT caused by too many consensus nodes and the mobility of nodes by adopting the method of network sharding, and meanwhile to ensure that transactions can be reached on consensus parallelly. To avoid the selection of primary nodes from being predicted in advance, this article proposes a primary node selection method based on verifiable random function (VRF). Moreover, a method is proposed of selecting transactions based on weighted round robin (WRR) scheduling to motivate nodes for the purpose of maintaining a high reputation score. Finally, the performance of the proposed scheme is verified by experiments whose results show that the proposed consensus mechanism shows better performance than other comparable schemes in terms of the time added to the blockchain, transaction per second (TPS), consensus success rate, and the time of obtaining transaction. And the proposed mechanism is less dependent on roadside unit (RSU).

User preference modeling from heterogeneous implicit feedback (i.e. various user behaviors) is a key issue in recommender systems, especially when homogeneous implicit feedback is sparse or even not available. But different types of implicit feedback have different confidence in user preference indication, which bring difficulties in building accurate user preference models in recommender system.In this article, we propose to model user preference profiles based on heterogeneous implicit feedback from a new perspective, in which preference confidence from different user implicit feedback should be measured in a unified way. Specifically, a preference confidence transfer learning algorithm based on multiple types of user behavior is designed. We assume that preference confidence can be transferred between target behavior and auxiliary behavior based on the features of user behavior. Then we integrated the assumption into a Latent Factor Model with confidence. Furthermore, we conduct extensive empirical studies with various baseline methods on two real public datasets and find that our algorithm can perform significantly more accurate than the other methods.

As the scale of online shopping users continues expanding, comments and feedback to users of their opinions after purchase and use are of great significance to assist users in purchasing decisions. In order to encourage users to actively comment, some websites provide comment tags for users to choose. These labels only summarize the common features of the products, but do not consider the differences among different products, it is even more unable to meet the personalized expression needs of user comments.. Based on the above reasons, the paper proposes an automatic generation algorithm of comment tags oriented to the personalized needs of users. This algorithm takes a single user as the research object, combines the Seq2Seq (Sequence to sequence learning) model and the part-of-speech-syntax features analysis algorithm to comment on users analyze, capture the description style and product characteristics of a single user in the comment language, combine the user's personalized language with the specific product characteristics, and generate comment tags from the two perspectives of user needs and product characteristics. The paper verifies the effectiveness of the algorithm through real data, compared with the existing algorithm, the algorithm in this paper is more suitable for user needs.

With the development of blockchain technology, the data on the blockchain represented by Bitcoin and Ethereum is exploding. Data represents people's activities, indicating that blockchain has been paid more and more attention. However, many problems are hidden behind so much data, such as security and privacy. Analyzing the data can help to find out the problems and propose methods to improve blockchain. Therefore, in order to make the blockchain well applied to various walks of life, data analysis is quite essential. In this paper, we collected the relevant literature in the field of blockchain data analysis, summarized the current analysis methods, and analyzed the research status from four aspects: security, privacy, performance and prediction of price. Finally, we discussed those research work and forecasted the future trend and challenges in this field, providing some reference for related researchers. We thought that applying machine learning technology to blockchain data analysis will become mainstream in the future.

In recent years, Vehicular Ad Hoc Network (VANET) has developed significantly. Coordination between vehicles can enhance driving safety and improve traffic efficiency. Due to the high dynamic characteristic of VANET, security has become one of the challenging problems. Trust of the message is a key element of security in VANET. This paper proposes a Manhattan Distance Based Trust Management model (MDBTM) in VANET environment which solves the problem in existing trust management research that considers the distance between the sending vehicle and event location. In this model, the Manhattan distance and the number of building obstacles are calculated by considering the movement relationship between the sending vehicle and event location. The Dijkstra algorithm is used to predict the path with the maximum probability, when the vehicle is driving toward the event location. The message scores are then calculated based on the Manhattan distance and the number of building obstacles. Finally, the scores are fused to determine whether to trust the message. The experimental results show that the proposed method has better performance than similar methods in terms of correct decision probability under different proportions of malicious vehicles, different numbers of vehicles, and different reference ranges.

With the development of wireless communication technology and the automobile industry, the Vehicular Ad Hoc Networks bring many conveniences to humans in terms of safety and entertainment. In the process of communication between the nodes, security problems are the main concerns. Blockchain is a decentralized distributed technology used in nonsecure environments. Using blockchain technology in the VANETs can solve the security problems. However, the characteristics of highly dynamic and resource-constrained VANETs make the traditional chain blockchain system not suitable for actual VANETs scenarios. Therefore, this paper proposes a lightweight blockchain architecture using DAG-lattice structure for VANETs, called V-Lattice. In V-Lattice, each node (vehicle or roadside unit) has its own account chain. The transactions they generated can be added to the blockchain asynchronously and parallelly, and resource-constrained vehicles can store the pruned blockchain and execute blockchain related operations normally. At the same time, in order to encourage more nodes to participate in the blockchain, a reputation-based incentive mechanism is introduced in V-Lattice. This paper uses Colored Petri Nets to verify the security of the architecture and verifies the feasibility of PoW anti-spam through experiment. The validation results show that the architecture proposed in this paper is security, and it is feasible to prevent nodes from generating malicious behaviors by using PoW anti-spam.

With a novel content-based communication model that is designed for data sharing, Named Data Networking (NDN) borns with innate security by per package encryption and achieves better scalability and mobility. Security attacks in the current Internet, e.g., IP and/or DNS spoofing, can hardly hazard NDN since there is no address and the embedded security in the data packet. However, some new kinds of attacks (e.g., interest flooding attack and content poisoning attack) damage NDN if not investigated thoroughly. In NDN, there exists a variant of content poisoning attack (CPA) that the attack is launched by some compromised core routers and this kind of attack causes severe damage to the network. In this paper, this variant of CPA has named the mole content poisoning attack (MCPA), and a Kalman filter based light-weight mitigation scheme is proposed. Besides, random sampling is imposed on the data traffic and tunes the state parameters in the Kalman filter to achieve faster convergence in a router. Compared with the mitigation scheme by the probabilistically checking of signatures on the data traffic, the proposed light-weight mitigation scheme recovers interest satisfaction rates (ISRs) of consumers while introducing less computation overhead.

The Internet of Things (IoT) is a large-scale complex network composed of a large number of heterogeneous devices, and due to the complexity of permission management in some scenarios of IoT, such as smart cities and smart healthcare, access control systems need to support strong flexibility. Permission delegation is an effective means to improve flexibility, but the delegated permission token in the current permission delegation scheme is a whole, and the recipient cannot re-delegate part of the permission in the received token. In this paper, a blockchain-based permission token segmentation scheme is proposed, which splits out part of the permissions in a permission token owned by the subject to generate a new token, so that the subject can control the fine-grained permissions in the token, enabling the subject to manage the permissions more flexibly. The permission combination scheme is then provided, and the permission invalidation problem in the token segmentation scheme is analyzed and discussed, and a token invalidation scheme is given. The security analysis shows that the scheme can reliably guarantee the security of authorized access.

Named Data Networking (NDN) is one of the next generation network architecture that focus on content rather than its location. However, there is a new type of security threat to NDN named Interest Flooding Attack. This paper proposes AH-IFAC: a countermeasure based on autoregressive integrated (ARI) model and Hidden Markov Model (HMM) to fight Interest Flooding Attack. The AH-IFAC runs on routers that connected to user and monitors external characteristic parameters of user. The router charges user when providing forward service and rewards user if the user is legal. Considering the dependency between sequential external characteristic parameters of user, the ARI model is used to predict the external characteristic parameters of user and the HMM utilizes predicted external characteristic parameters to detect the evil user. Finally, the effectiveness of the AH-IFAC is verified through experiment. The simulation results show that the AH-IFAC can effectively mitigate IFA, the Interest packet satisfaction ratio can be increased up to 84% and 87% at tree and net-like topology respectively.

In this paper we propose a new user purchase preference model based on their implicit feedback behavior. We analyze user behavior data to seek their purchase preference signals. We find that if a user has more purchase preference on a certain item he would tend to browse it for more times. It gives us an important inspiration that, not only purchasing behavior but also other types of implicit feedback like browsing behavior, can indicate user purchase preference. We further find that user purchase preference signals also exist in the browsing behavior of item categories. Therefore, when we want to predict user purchase preference for certain items, we can integrate these behavior types into our user preference model by converting such preference signals into numerical values. We evaluate our model on a real-world dataset from a shopping site in China. Results further validate that user purchase preference model in our paper can capture more and accurate user purchase preference information from implicit feedback and greatly improves the performance of user purchase prediction.

Named Data Networking (NDN) is one of the future Internet architectures and has recently attracted a great deal of attention. NDN adopts named routing and caches all contents passing by. So it helps to reduce network traffic, accelerate content distribution and reduces content retrieval latency. However, the default caching strategy named Leave Copies Everywhere (LCE) also brings a large number of redundant data and the higher number of content replacements in NDN. In this paper, we present a Dynamic Popularity-Based Caching Permission strategy (DPCP). DPCP takes advantage of interest packet and data packet to carry content popularity, so routers in the path can obtain the information about the content popularity and uses dynamic popularity threshold to make cache permission policy. It also introduces cache control flag to avoid caching the same redundant copies in adjacent routers. The simulation results show that the DPCP strategy can effectively reduce the amount of redundant data in the network as well as reduce the number of the content replacement while acquiring the similar cache hit ratio compared to LCE.