This paper presents LAPA (Look Around and Pay Attention), a novel end-to-end transformer-based architecture for multi-camera point tracking that integrates appearancebased matching with geometric constraints. Traditional pipelines decouple detection, association, and tracking, leading to error propagation and temporal inconsistency in challenging scenarios. LAPA addresses these limitations by leveraging attention mechanisms to jointly reason across views and time, establishing soft correspondences through a cross-view attention mechanism enhanced with geometric priors. Instead of relying on classical triangulation, we construct 3D point representations via attention-weighted aggregation, inherently accommodating uncertainty and partial observations. Temporal consistency is further maintained through a transformer decoder that models long-range dependencies, preserving identities through extended occlusions. Extensive experiments on challenging datasets, including our newly created multi-camera (MC) versions of TAPVid-3D panoptic and PointOdyssey, demonstrate that our unified approach significantly outperforms existing methods, achieving 37.5% APD on TAPVid-3D-MC and 90.3% APD on PointOdyssey-MC, particularly excelling in scenarios with complex motions and occlusions. Code is available at https://github.com/ostadabbas/Look-Around-and-Pay-Attention-LAPA-.

Underwater target detection faces great challenges due to low visibility, low contrast and light absorption and scattering. The existing YOLO series has obvious defects when used underwater: the weight of convolutional kernel parameters is fixed, and it is difficult to adapt to complex underwater background and multi-scale targets; The traditional multi-scale feature extraction mechanism can not effectively deal with the problem of fuzzy and dense distribution of small underwater targets. Lack of robust attention mechanism for underwater low-quality images, susceptible to background interference and sensitive to occluded targets. Therefore, the improved model YOLO-RDM based on YOLOv8 is proposed in this study. The RFAConv module, which can dynamically adjust convolution ker-nel parameters, is introduced to strengthen the feature extraction capability of complex underwater scenes. The Dual_ C2f structure with$3\times 3$convolution kernel and IxI convolution kernel is used to optimize the feature interaction, which reduces the computational cost while maintaining multi-scale information fusion. A new multi-scale spatial attention enhancement mechanism, MultiSEAM, is designed to improve the robustness of detection of small targets and occluded scenes through depth-separable convolution and multi-scale feature weighting. In the URPC2020 dataset, [email protected] and [email protected]:0.95 of YOLO-RDM reach 84.0% and 49.8%, respectively, the reasoning speed is 49.9, and the parameter number is only 3.0M. Significantly better than the mainstream YOLO series and RT-DETR and other latest models. The ablation experiments further confirm the effectiveness of the improved modules. The YOLO-RDM achieves a good balance between precision and lightweight, providing an efficient and reliable solution for small underwater target detection.

With the development of ocean exploration and monitoring technology, underwater acoustic sensor networks play an increasingly important role in underwater environmental monitoring, disaster prevention, resource exploration and other fields. However, the particularities of the underwater environment, such as signal attenuation, noise interference, and energy constraints, make efficient data transmission a great challenge. Therefore, we propose a depth-first clustering routing algorithm (DFCR) for underwater acoustic sensor networks, and optimize the cluster head selection and routing of the algorithm, aiming to improve the efficiency of data transmission and prolong the network lifetime. We divide the network into multiple layers, and select the cluster head in each layer except the bottom layer according to the depth and residual energy. The cluster head receives the messages from the sensor nodes at the lower layer than itself, which can avoid the transmission of data from the nodes at the shallower depth to the nodes at the deeper layer, and reduce the number of network hops. Finally, we conduct simulation experiments to verify the effectiveness of the proposed algorithm. The experimental results show that compared with some existing routing algorithms, our algorithm has better performance in terms of packet delivery ratio and network lifetime.

In the digital age, the security and privacy of col-laborative systems are increasingly critical. This paper examines the asynchronous interactive learning environment within Delay Tolerant Networks (DTN), which leverages satellite broadcasting in remote areas. However, this environment encounters challenges such as weak infrastructure, unstable connections, and security vulnerabilities associated with open networks. Additionally, traditional public key cryptosystems are susceptible to threats posed by quantum computing. To mitigate these issues, we propose an anonymous communication scheme based on lattice theory and pseudonyms, designed to enhance secure network communication for remote and impoverished areas through collaboration among communication entities in the DTN environment. Our security analysis and performance comparison demonstrate that the pro-posed scheme outperforms existing lattice-based solutions while maintaining essential security characteristics.

Federated learning (FL) is a privacy-preserving distributed machine learning approach that enables training by exchanging model parameters without uploading local private data. However, the data heterogeneity across clients poses significant challenges in achieving personalized privacy protection. Existing methods often struggle to balance privacy and model utility, especially with inconsistent data distributions. Personalized differential privacy (PDP) is a commonly used technique to provide differential privacy (DP) by introducing varying levels of noise for each client. The noise level directly affects the model's utility, making it crucial to precisely determine the appropriate noise for each client. To address this challenge, we propose a federated knowledge distillation method based on PDP, named PDP-FD. PDP-FD dynamically adjusts the network architecture of both base and personalized layers to align with the data characteristics of different clients, enhancing model personalization. Moreover, it allocates an appropriate privacy budget to each client based on the similarity between local and global models, thereby meeting diverse privacy needs. Experimental results show that PDP-FD significantly outperforms existing FL methods in accuracy, effectively balancing privacy protection and model utility.

Vertical Federated Learning (VFL) offers a promising approach to collaborative model training, allowing participants to share identical data samples with distinct attributes. This approach avoids direct raw data sharing, enhancing data privacy, though it may sometimes reduce model accuracy. In this paper, we introduce DPTB-VFL, a differential privacy-based vertical federated learning framework that leverages boosting trees. Boosting trees are chosen for their exceptional ability to handle heterogeneous features, deliver robust performance with minimal preprocessing, and provide interpretability-all crucial in privacy-sensitive scenarios. Additionally, most current methods apply a uniform privacy budget across all training steps, overlooking variations in local gradients that could better balance privacy and utility. DPTB-VFL addresses this gap by introducing an adaptive differential privacy protocol, enabling dynamic privacy budget allocation based on the model's learning progress. Experimental evaluations on three public datasets demonstrate that DPTB-VFL not only achieves higher accuracy than existing approaches but also significantly reduces computational latency, aligning data privacy needs with model performance

Federated Learning (FL) enables multiple clients to collaboratively train models without sharing data. However, it commonly faces dual challenges of security and high communication costs. Differential Privacy (DP) offers protection by adding noise to model parameters based on strict privacy standards, but excessive noise can compromise model accuracy. Additionally, the communication cost associated with training large-scale models in FL can be both slow and expensive. In this paper, we propose CPDP-FL, an efficient and privacy-preserving federated learning algorithm that combines model pruning with differential privacy to address these issues. By pruning the model based on neuron correlation before client training, we reduce redundant parameters, which not only improves communication efficiency but also reduces the amount of noise needed for DP, thereby preserving model accuracy. During training, we apply differential privacy to the remaining parameters and introduce a novel layer-wise privacy budget allocation strategy. This approach assigns different privacy budgets to different layers to balance privacy protection with model accuracy. Extensive experiments demonstrate that our method achieves high communication efficiency and robust privacy protection while minimizing unnecessary privacy budget expenditure.

Education plays a crucial role in driving personal development and societal progress. In the digital age, satellite-based remote open education for remote areas has encountered significant opportunities for development. However, with limited satellite network bandwidth, resource allocation efficiency is insufficient to handle high-concurrency task requests, especially with repeated requests for the same tasks, preventing satellite remote education from fully realizing its potential. To address this issue, this paper proposes a resource scheduling method for satellite-based remote education networks, including a task merging mechanism and a multivariate-initialized particle swarm-genetic hybrid algorithm for optimized scheduling. Experimental results show that, compared to baseline algorithms, the proposed method improves task completion rate by at least 10% per unit time and achieves lower resource consumption, with maximum bandwidth utilization efficiency. Therefore, this algorithm enhances the performance of satellite network educational resource scheduling, optimizes task scheduling and resource allocation strategies, and provides effective support for remote education in underserved areas.

Diffusion models have proven effective at generating high-quality images from learned distributions, but their application to the temporal domain, especially for driving scenarios, remains underexplored. Our work addresses key challenges in existing simulations, such as limited data quality, diversity, and high costs, by extending diffusion models to generate realistic long driving videos. We introduce the Dualconditioned Temporal Diffusion Model (DcTDM), an opensource method that incorporates dual conditioning to enforce temporal consistency by guiding frame transitions. Alongside DcTDM, we present DriveSceneDDM, a comprehensive driving video dataset featuring textual scene descriptions, dense depth maps, and canny edge data. We evaluate DcTDM using common video quality metrics, demonstrating its superior performance over other video diffusion models by producing long, temporally consistent driving videos up to 40s, achieving over 25% improvement in consistency and frame quality.

With the advent of the 5G era, emerging industries such as cloud computing and big data are rapidly developing. Data centers, as resource management platforms and critical transmission hubs, are experiencing a sharp increase in data traffic, necessitating efficient routing optimization algorithms. Advances in software-defined networking and programmable network devices enable different flows to be transmitted across multiple network paths, allowing fine-grained network performance optimization. Based on a knowledge-defined networking architecture, this paper proposes a multipath routing algorithm called DRL-MR, utilizing multi-agent deep reinforcement learning. The algorithm organizes agents in a hop-by-hop manner to generate routing paths, taking into account flow latency, transmission rate, and packet loss rate to devise optimal routing strategies. Additionally, we introduce an auxiliary learning mechanism to enhance reliability and accelerate the online learning process. Experimental results show that DRL-MR significantly outperforms baseline methods in terms of network throughput, link bandwidth utilization, and latency, while demonstrating strong adaptability, robustness, and reliability in scenarios with traffic variation, node failures, unknown large-scale topologies, and partial deployment.

Federated learning enables multiple clients to collaboratively train a model without sharing their data, but it faces two main challenges: privacy concerns during the upload process and high communication costs. Differential privacy protects privacy in federated learning by adding noise to model parameters; however, the amount of noise is directly related to the size of the model parameters, and excessive noise can degrade the model’s accuracy. Additionally, training large-scale models increases communication costs and slows down training speed. To address these issues, this paper proposes a privacy-efficient federated learning algorithm that combines model pruning, differential privacy, and knowledge distillation to enhance communication efficiency while ensuring privacy protection. The algorithm first prunes the global model based on neuron correlations to reduce the size of the model parameters, lower the noise introduced by differential privacy, and significantly reduce communication costs. By pruning irrelevant or less impactful neurons, the algorithm effectively reduces the model’s complexity. Next, knowledge distillation is performed on the server side, where the pruned model is distilled with the original global model. This process transfers knowledge from the original model to the pruned one, alleviating the accuracy loss caused by pruning. Finally, the distilled student model is sent to each client for local training. During training, differential privacy is applied to the model parameters to ensure privacy protection. Additionally, a hierarchical privacy budget allocation scheme is introduced, dynamically distributing the privacy budget across different layers of the model. This optimizes accuracy while maintaining privacy for each layer. Extensive experiments on three datasets validate the effectiveness of the proposed method, demonstrating its significant advantages in terms of communication efficiency, privacy protection, and accuracy preservation.

Routing algorithms in underwater acoustic sensor networks are the primary solution to address issues such as low bandwidth, long transmission delays, high bit error rates, and limited energy in underwater environments. However, many routing algorithms face challenges such as suboptimal path selection, low energy efficiency, and poor overall performance. To address these issues, this paper proposes an intelligent routing algorithm for underwater acoustic sensor networks (QLMR-PO) that combines learning automaton and Q-learning for efficient routing. First, we optimize the transmission power of each node using learning automaton. Then, we introduce multiple influencing factors into the reward function design and select the best relay node by choosing the maximum Q-value. Finally, we ensure the forward propagation of data by introducing depth information and a memory mechanism. Simulation results show that the proposed algorithm outperforms several Q-learning-based algorithms in terms of energy efficiency, end-to-end delay, network lifetime, and other performance metrics.

Unlike traditional networks, Delay Tolerant Network (DTN) does not have a stable end-to-end connection, so it adopts the store-carry-forward transmission mode. In DTN routing, the context information of nodes and messages, such as the social attributes between nodes, the buffer occupancy of nodes and the hop count of messages, is very important for the design of routing algorithms. Considering the above factors, this paper proposes an improved spray and wait routing algorithm based on Q-learning (ANSAW-Q). Firstly, this paper proposes a social circle construction method to characterize the similarity of social circles between nodes. Secondly, exploiting the distinctive features of the DTN routing process, we introduce connection and transmission reward functions that leverage context information of nodes and messages to update the Q-value. Finally, an adaptive method for initial message replication is introduced based on the relationship between the source and the destination. In the routing process, the node selects the appropriate relay node and manages the buffer according to the Q value. Simulation results show that ANSAW-Q can improve the message delivery ratio and reduce the network load. Q-learning can effectively improve spray and wait algorithm.

In response to the challenge of users lacking access to accurate network attributes, this paper proposes a heterogeneous network vertical handover algorithm that combines a fuzzy logic system with reinforcement learning. It begins by introducing the three modules of the fuzzy logic system and then delves into the Q-learning method in reinforcement learning. The paper proceeds to compare the proposed algorithm with three other methods using five metrics: bandwidth, latency, jitter, packet loss, and cost. Finally, it discusses the current issues and challenges faced by existing vertical handover algorithms. Simulation results indicate that this approach outperforms the other three methods in terms of switching frequency and the specified metrics, including bandwidth, jitter, latency, packet loss, and cost.

With the advent of the 5G era and the rapid development of emerging industries such as cloud computing and big data, data centers, as resource management platforms and key transmission hubs, are experiencing a dramatic increase in business data volume. Therefore, efficient routing optimization algorithms are needed. Improvements in Software Defined Networks (SDN) and programmable network devices allow different data traffic to be transmitted through various network paths, enabling fine-grained network performance optimization. In this paper, we utilize the emerging Hybrid Knowledge-Defined Network (KDN) architecture to propose a Multi-Agent Reinforcement Learning (MARL)-based multipath routing algorithm, termed RL-MR. RL-MR organizes intelligent agents to generate routes in a hop-by-hop manner, offering excellent scalability. The algorithm comprehensively considers flow latency, flow transmission rate, and packet loss rate to formulate routing strategies. To ensure reliability and accelerate the learning process, we introduce an auxiliary learning mechanism into RL-MR. Experiments conducted using Ryu and Mininet demonstrate that the RL-MR algorithm significantly outperforms baseline methods in terms of network throughput, link bandwidth utilization, and latency. Additionally, it exhibits excellent adaptability and reliability in scenarios involving flow changes, unknown large topologies, and partial deployments.

As the most widely distributed vegetation type on Earth, grasslands are the second largest ecosystem after forests, and are known as the “skin of the earth”. However, the frequency of rodent infestation has been increasing year by year in recent years due to climate change, loose soil and other factors. As the current traditional monitoring methods are time-consuming and laborious, while the deep learning method has the advantages of low cost, high efficiency and high accuracy. Therefore, this paper proposes a combination of UAV remote sensing and deep learning methods to monitor grassland mouse hole. First, because there is no publicly available grassland mouse hole dataset, this study used a UAV to capture images to create a grassland mouse hole dataset. Second, an improved YOLOv7 model, YOLOAPM, is proposed with the goal of improving the accuracy of small target recognition. Through ablation and comparison experiments, the validity of the improved model as well as the method of this paper are verified to be improved compared to other models. Finally, in order to facilitate the monitoring task of the researchers, a monitoring system is established and the mouse hole identification of this paper is embedded into the system.

In recent years, remote sensing technology has been widely applied in various industries, and semantic segmentation of remote sensing images has attracted much attention. Due to the complexity and special characteristics of remote sensing images, multi-scale object detection and accurate object localization are important challenges in remote sensing image semantic segmentation. Therefore, this paper proposes a context aggregation network (CANet). The design of CANet is influenced by advanced technologies such as attention mechanisms and feature fusion and enhancement. This network first introduces nested dilated residual module (NDRM), which can fully utilize the features extracted by the backbone network. Then, improved integrated successive dilation module (IISD) is proposed to effectively aggregate a series of contextual information scales. Next, Swim Transformer module is embedded to provide global contextual information. Finally, multi-resolution fusion module (MRFM) is proposed, allowing the comprehensive fusion of feature layers from different stages of the encoder, preserving more semantic and detailed information. The experimental results show that CANet outperforms other advanced models on the Potsdam and Vaihingen datasets.

Power infrastructure and its connectivity are central to building a smart grid. The infrastructure regional subsidence caused by environmental factors or geological hazards may devastate grid systems. Therefore, it is critical to forecasting the infrastructure regional subsidence in smart grids. In this study, we used an InSAR time series subsidence dataset based on satellite remote sensing images to train, test and compare four deep learning-based Transformer series prediction models using transfer learning to achieve subsidence crisis monitoring and prediction in smart grid infrastructure areas, considering the influence of environmental factors on infrastructure regional subsidence. Meanwhile, an GIS for subsidence crisis prediction in smart grid infrastructure areas was developed based on the Autoformer model. It helps the power industry maintain the smart grid more efficiently and accurately while also saving a lot of money and labor.

Evaluating autonomous ground vehicles requires evaluating their mobility performance. Since autonomous vehicles are envisioned to make decisions in a variety of situations and environments too diverse to practically assess with only physical testing, their development, and evaluation will necessarily include the use of simulations. These simulations must represent reality sufficiently to represent the decisions that the vehicles would make in real-world. In this paper we present our Scoping Autonomous Vehicle Simulation (SAVeS) platform for benchmarking the performance of simulated environments for autonomous ground vehicle testing1.

Delay Tolerant Network (DTN) is a network model designed for special environments. It is designed to be used in challenging network environments with high latency levels, bandwidth constraints, and unstable data transmission. It plays an important role in extremely special environments such as disaster rescue, maritime communication, and remote areas. Currently, research on DTN mainly focuses on innovative routing protocols, with limited research of the security issues and solutions. In response to the above problems, this paper analyzes and compares the security problems faced by delay tolerance networks and their solutions and security schemes.

Due to the different coverage of different access technologies and network properties, it is one of the fundamental problems how to ensure that users can choose the most suitable network in the case of multi network selection, and can switch between different networks seamlessly, so as to get high-quality user service. This paper focuses on the heterogeneous Internet of Vehicles environment composed of LTE, WiFi, and DSRC, and combines single-user and group handover schemes according to the different states of the vehicle. When the vehicle is in an isolated state, the single-user vertical switching algorithm is adopted; when the vehicle is in the cluster head or cluster member state, the vertical switching algorithm of “fleet” users is adopted. The simulation experiment is carried out by MATLAB, and the better performance of switching is obtained.

In sparse opportunistic networks with social at-tributes and limited cache space, routing algorithms based on long-term and stable social relations have low message delivery rates in a short period of time, and do not consider the impact of delivered message copies on communication efficiency. To solve the problem of low delivery rates, the paper integrates social attributes and space-time constrained characteristics and proposes social closeness and spatio-temporal interaction degree based on historical information. Then we design a routing algorithm (SC-STID) based on these two metrics. SC-STID takes into account the social closeness and spatio-temporal interaction of nodes to select relay nodes comprehensively and uses the delivered message deletion mechanism to remove the copies of delivered messages from the network. Experimental data shows that SC-STID improves the success rate of message delivery and reduces network resource consumption compared to existing routing algorithms (Prophet, Epidemic, ORRF) in a sparse opportunistic network of nodes with social attributes.

Traffic flow prediction plays an important role in intelligent traffic management. In order to solve the problem that the prediction model has low accuracy in traffic flow prediction when the amount of data is small. Considering the chaotic nature of the traffic flow time series data, the phase space reconstruction method is adopted to process the data, and then the SVR model is used to predict the processed data. The two parameters of phase space reconstruction, including embedding dimension and delay time, have a great impact on the final prediction accuracy. In this paper, the improved KNN method is used to select the parameters of phase space reconstruction and construct the KNN-GA-SVR model. Compared with the CC-GA-SVR model of phase space parameter selection based on C-C method, this model improves the prediction accuracy, is effective and feasible for the prediction of short-term traffic flow, and has strong applicability.

In an opportunistic network with social attributes, the selection of relay nodes is mainly done according to community division. In the social opportunity network, community partition is the key to reducing data forwarding delay and improving the success rate of data delivery. To solve the problem that existing opportunistic network routing algorithms do not analyze the impact of overlap and transitivity of community on opportunistic routing, a 2-hop routing strategy for overlapping communities based on social intimacy(TOCSI) is proposed. The algorithm first introduces the overlapping community division method PercoMCV to reasonably divide the node community structure and designs the opportunity route based on the result of the community division. Experiments show that compared with Epidemic, Prophet, and ORRF algorithms, TOCSI can effectively improve the success rate of message delivery and reduce the average data forwarding delay and routing overhead.

Energy has always been a key bottleneck restricting the large-scale deployment and long-term operation of wireless sensor networks (WSNs). Wireless rechargeable sensor networks (WRSNs) can effectively alleviate the energy-constrained problem of sensor nodes. However, due to the constraints of the service capabilities of mobile charging equipment, how to efficiently replenish energy and maintain the long-term operation of the network is still very challenging. In this paper, we propose an UAV-based energy replenishment mechanism in WRSN, which aims to maximize the replenished energy benefit of the network from the energy expended by UAVs. This paper firstly constructs the network model and defines the optimization problem of energy replenishment efficiency maximization. Then, in order to improve the performance of WRSN in terms of energy replenishment efficiency, the node clustering, anchor node selection and flight path planning problems are respectively studied, and a data collection and energy replenishment mechanism is proposed for the above problems. The experimental results show that the proposed scheme can effectively improve the energy replenishment efficiency of the system, prolong the life of the network and balance the energy consumption of the network.

In recent years, drone technology has developed rapidly. Drone's low cost, fast and flexible deployment, as well as strong mobility have made it possible to use drone-assisted sensor networks for data collection tasks. In this way, data collection nodes can break through the movement path restriction of traditional nodes, broaden the spatial movement range of nodes, and it is more suitable for data collection in complex environments. In this paper, we proposed a data collection strategy based on drone technology in Wireless Sensor Networks. Kmeans++ clustering method is used for auxiliary clustering and cluster head election in the initial state, which significantly improves the final error of the clustering result. Then, we used drone to assist cluster head election and data collection, which comprehensively considering the relative distance of every sensor node in the cluster and their relative remaining energy. In addition, for some nodes that have not been elected in the previous specified round, a reasonable priority is set to make the energy consumption of sensor nodes in the entire network more balanced. At the same time, we excluded the influence of dead nodes. Compared with many new methods proposed in recent years, the data collection strategy proposed delays the death time of the sensor nodes, reduces the overall energy consumption of the sensor nodes, and has a better performance. This work provides new ideas for the future work.

Frequent itemset mining has been a focused theme in the field of data mining, which is widely used in business decision making, economics, medicine, bioinformatics and other fields. Frequent itemset mining can provide a lot of valuable information when making decisions, but it may bring the risk of privacy disclosure when mining and publishing frequent itemsets. In order to solve the privacy leakage problem, most of the existing solutions are using horizontal mining method to mine frequent itemsets under differential privacy. However, these solutions generally suffer from complex support computation and poor accuracy due to large candidate sets. In this paper, we propose a new vertical frequent itemset mining algorithm based on differential privacy, which is referred to as DP-Eclat. In DP-Eclat, a new privacy budget allocation strategy is proposed to rationalize the privacy budget allocation, which allows privacy budget to be used more fully. In addition, we devise a multiple pruning strategy to further improve the data utility by prune before and after the generation of candidate itemsets. Through privacy analysis, we prove that DP-Eclat satisfies E -differential privacy. Extensive experiment results on multiple real datasets show that DP-Eclat significantly outperforms state-of-the-art algorithms in terms of data utility.

Privacy-preserving data publishing is a hot issue in the field of privacy protection. Differential privacy is a burgeoning technology of privacy protection which provides a powerful privacy mechanism and does not make restrictive assumptions on the attacker's background knowledge. At present, there does not have an effective way to generate Synthetic high-dimensional data with differential privacy technology. Aiming at the issue of high-dimensional privacy data publishing, this paper proposed a method called APrivBayes, which altered the structure of Bayesian network to make it adapting differential privacy mechanism. Then proposed a first node selection mechanism based on attribute correlation degree for the new structure of Bayesian network. Through theoretical analysis and experimental evaluation, this method improves the effect of network and reduces Laplace noise effectively, while protecting personal privacy and improving the usability of published data.

In recent years, as image data are widely used in data analysis tasks, the problem of privacy disclosure is becoming more and more serious. However, the privacy protection technology of image data is still immature. In this paper, we propose a privacy protection framework named dp-WGAN for image data. This framework uses differential privacy and generative adversarial network to train a generative model with privacy protection function. Using this generative model, synthetic data with similar characteristics to sensitive data can be obtained, and synthetic data is published instead sensitive data to complete all kinds of data analysis tasks. Through extensive empirical evaluation on benchmark datasets, we demonstrate that dp-WGAN can provide strong privacy protection for sensitive data and produce high-quality synthetic data.

VANET is the application of traditional mobile Ad hoc network on traffic roads, which is a special mobile Ad hoc network. In recent years, a lot of research work has been done on the VANET, but most of these studies fail to consider the selfishness of vehicle nodes in the VANET. However, each vehicle node in VANET has its own private property in practice. This thesis introduces the interest packet and data packet transmission method of CCN and proposes a routing incentive strategy based on virtual credit. First, the system model is established for the proposed incentive strategy, and then introduces the implementation process of the incentive strategy CDI based on virtual credit in detail. Then the VANET message transmission security mechanism is proposed. Finally, the simulation experiment analysis of the CDI incentive strategy is carried out. It proves the effectiveness of the incentive strategy.