Federated learning coordinates model training in a distributed manner within Internet of Things (IoT) systems and ensures the privacy of local client data simultaneously. Nonetheless, traditional federated learning relies primarily on a unified global model and focuses on local feature extraction, failing to accommodate the diversity and personalized needs of clients in non-independent and identically distributed (non-IID) environments. To mitigate the decline in model accuracy posed by these challenges, we propose a discrepancy-aware federated learning method based on multi-branch feature fusion (FedDM). Firstly, we design a differential-aware aggregation strategy (DA), which adjusts the contribution of each client during model aggregation using Gaussian distribution statistics, to generate personalized local models. Next, we propose a multi-branch feature fusion mechanism (MFF) that integrates diverse feature representations through multi-scale pooling and feature enhancement, enabling the incorporation of features across both spatial and channel dimensions for a more holistic representation. Experimental results demonstrate that FedDM enhances model accuracy and robustness, while exhibiting adaptability when facing challenges posed by data distribution heterogeneity.

Vehicle trajectory prediction plays a crucial role in IoT-based intelligent transportation systems, which can effectively address key issues, such as driving safety and multivehicle collaboration. However, the sensitivity of trajectory data and the reluctance of data holders to share it constrain the prediction model’s ability to capture vehicle behavior patterns in different scenarios. To address the above problems, we propose a blockchain-enabled asynchronous federated proximal policy optimization framework (BE-AFPPO) for the trajectory prediction of self-driving vehicles. First, we propose a curiosity proximal policy optimization (C-PPO) algorithm. The method utilizes a driven exploration strategy to actively motivate the intelligent agent to explore the unknown state space. The avoidance policy model reaches a local optimum when processing trajectory data. In addition, we design historical gated recurrent unit (GRU) and future GRU as input layers. The target’s historical motion features and future trajectory features are extracted, respectively. Then, various data is received through asynchronous federated learning. This model can fully learn the vehicle’s behavior patterns in different scenarios, which improves prediction accuracy. Based on this, we develop a blockchain-based dynamic group practical Byzantine fault tolerance (DG-PBFT) consensus algorithm. This enhances the credibility and integrity of the data while enriching the sources of trajectory data. Finally, we perform the experiments on the publicly available dataset nuScenes. The results demonstrate that the proposed method improves the robustness and accuracy of trajectory prediction.

Accurate traffic flow prediction is crucial for alleviating traffic congestion and optimizing intelligent transportation systems. However, traffic flow is subject to uncertainties and exhibits complex spatial and temporal dependence and dynamic change characteristics. Moreover, many efforts rely on a single view, which makes it difficult to comprehensively capture multiple levels of spatial and temporal correlations, thus limiting the accuracy of predictions. Therefore, we propose the multi-view spatio-temporal dynamic graph convolution framework MVSTDG for more comprehensively exploring and fusing the multi-view spatio-temporal features. Firstly, we design a dual-path Time-Patch Convolution (TPConv) module to separately model short-term fluctuations and long-term periodic trends, enabling effective extraction of dynamic features at multiple temporal scales. Secondly, we construct a data-driven traffic pattern library to generate dynamic adjacency matrices and integrate them with static topologies view. An Adaptive Diffusion Graph Convolutional Network (ADGCN) is then employed to model both local and global spatial correlations. In addition, we design a cross-gated spatio-temporal fusion mechanism that adaptively adjusts the contribution of short-term and long-term information, enhances the interaction of spatio-temporal information, and improves the model’s adaptive capability under different time scales. The experimental results show that MVSTDG outperforms the state-of-the-art baselines in several evaluation metrics and demonstrates higher prediction accuracy and stability on the four real datasets.

Mobile ad hoc networks (MANETs) have attracted much attention due to their mobility and ease of deployment. However, the wireless and dynamic natures render them more vulnerable to various types of security attacks than the wired networks. The major challenge is to guarantee secure network services. To meet this challenge, certificate revocation is an important integral component to secure network communications. In this paper, we focus on the issue of certificate revocation to isolate attackers from further participating in network activities. For quick and accurate certificate revocation, we propose the Cluster-based Certificate Revocation with Vindication Capability (CCRVC) scheme. In particular, to improve the reliability of the scheme, we recover the warned nodes to take part in the certificate revocation process; to enhance the accuracy, we propose the threshold-based mechanism to assess and vindicate warned nodes as legitimate nodes or not, before recovering them. The performances of our scheme are evaluated by both numerical and simulation analysis. Extensive results demonstrate that the proposed certificate revocation scheme is effective and efficient to guarantee secure communications in mobile ad hoc networks.

Since Wireless Mesh Networks (WMNs) can be easily deployed without wirelines among wireless mesh routers, they allow us to quickly recover network access services in disaster areas even if the existing network infrastructures have been enormously destroyed by terrible earthquake, tsunami, and so on. However, the performance of wireless mesh networks is largely affected by many factors, e.g., wireless mesh routers' locations, channel assignment, transmission scheduling, etc. In particular, the method of selecting gateways which has a connection to external networks significantly impacts on the network performance when the topology and routing have been fixed in the wireless mesh network. In this paper, we suppose a wireless mesh network which consists of wireless mesh routers and a base station directly connected to external networks. The base station is located at the center of the wireless mesh network chooses a certain number of wireless mesh routers as gateways, and establishes a connection with each of them. Our goal is to easily and quickly find the candidate gateways that maximize the system throughput without solving a complex optimization problem which includes a large number of parameters and involves heavy computation load. The performance of the proposed scheme is evaluated by numerical analysis, and demonstrated through computer simulations. The results show that our proposed scheme can determine the appropriate candidate gateway with high accuracy when there is a certain variance in the amount of traffic generated by users at each wireless mesh router.

Certificate revocation is an important security component in mobile ad hoc networks (MANETs). Owing to their wireless and dynamic nature, MANETs are vulnerable to security attacks from malicious nodes. Certificate revocation mechanisms play an important role in securing a network. When the certificate of a malicious node is revoked, it is denied from all activities and isolated from the network. The main challenge for certificate revocation is to revoke the certificates of malicious nodes promptly and accurately. In this paper, we build upon our previously proposed scheme, a clustering-based certificate revocation scheme, which outperforms other techniques in terms of being able to quickly revoke attackers' certificates and recover falsely accused certificates. However, owing to a limitation in the scheme's certificate accusation and recovery mechanism, the number of nodes capable of accusing malicious nodes decreases over time. This can eventually lead to the case where malicious nodes can no longer be revoked in a timely manner. To solve this problem, we propose a new method to enhance the effectiveness and efficiency of the scheme by employing a threshold based approach to restore a node's accusation ability and to ensure sufficient normal nodes to accuse malicious nodes in MANETs. Extensive simulations show that the new method can effectively improve the performance of certificate revocation.