With the advancement of autonomous driving and intelligent transportation systems, there has been a notable increase in the demand for multi-modal task processing by mobile edge terminals. However, intelligent vehicles need to offload complex tasks to servers due to their constrained processing capabilities and limited storage. In this context, the multi-access edge servers which are closer to vehicle terminals, emerge as a superior alternative. Nonetheless, the heterogeneous distribution of traffic flows across time and space can lead to significant communication delays due to potential overloading of some edge servers when offloading schemes employ random resource allocation. To address these challenges, this paper proposes a region attention-based edge resource allocation (RAllo) model to allocate the computing resource by predict regional demand. A regional attention-based traffic prediction model (RATra) extracts crucial spatio-temporal information using a temporal module and a graph attention network to forecast the regional demand. As for the resource allocation, after getting the predicted demand of tasks based on RATra, the radiating-breadth-first-allocation (RBFA) algorithm is designed for task scheduling within and across regions based on task demand. Finally, the results of experiments demonstrate the superiority of RAllo in reducing time consumption and enhancing MEC server utilization compared to existing methods.

In the Internet of Connected Vehicles, caching content frequently requested by users on edge devices can reduce access latency. Particularly in high-traffic density areas, Unmanned Aerial Vehicles (UAVs) can integrate into future cellular networks to enhance the network capacity and meet increased requests. Therefore, we formulate a joint optimization problem of cooperative caching of Base Station (BS) and UAVs and UAV trajectory planning to minimize network latency while considering the limited energy and storage capacity and dynamic vehicles. First, we propose a Temporal-evolving Bipartite Graph Neural Networks (TBGN) model for traveling areas prediction of vehicles. Then, regarding the coupling of optimization variables, we propose an Energy-aware Monte-Carlo Tree Search algorithm to optimize the UAV’s service trajectory by predicted spatio-temporal vehicle density. Finally, the optimization problem degenerates into a monotonic submodular function to optimize caching decisions. We utilize real vehicle trajectories for simulations. The results show that the TBGN outperforms other advanced models in terms of mobility prediction accuracy by 7.4%, and the proposed scheme reduces average latency by 16% compared to other schemes.

With the development of the 6th generation communication technology, the service traffic of mobile communication is rapidly growing. Many new types of services usually have high requirements for computing resources and low latency constraints. They need to be offloaded to a base station (BS) with computing resources for processing. In some disaster areas, the communication system will go down due to damage to the ground infrastructure. High altitude platforms (HAPs) with extensive coverage and unmanned aerial vehicles (UAVs) with simple deployment can provide various emergency services as aerial BS. UAVs and HAP carry servers and other equipment to serve users. It is a promising technology for communication and computing services. Due to UAVs’ limited computing resources and energy, it is a challenge to deploy them effectively and fully use network resources. Therefore, a task-dynamic processing through multi-UAV cooperation (TDPUC) strategy is proposed. A improved K-means algorithm is proposed to realize the dynamic deployment, which optimizes the number of UAVs dispatched and reduces the overall energy consumption. In addition, the multi-UAV cooperation for task offloading can realize dynamic task processing under constrained energy and resources. When UAVs cooperate, the multi-agent reinforcement learning (MARL) algorithm is used to optimize resource allocation and learns online. By numerical results, the proposed TDPUC strategy can improve the service capacity of tasks by 11% on average with less energy consumption.

By shifting the requested content to the edge in the Internet of Vehicles (IoV), edge caching is expected to be an effective solution to satisfy the low latency and high-reliability requirements of IoV users for multimedia services. However, the edge node’s coverage area and storage space are limited. Moreover, since vehicles have high mobility and in-vehicle multimedia applications require sequential delivery for contents, we need to address two main issues: 1) how to optimize the proactive content caching decision (i.e., the placement of cached content chunks) among edge nodes (ENs) to provide better Quality of Services (QoS) for IoV users and 2) how to ensure that vehicles can download the required contents sequentially to improve Quality of Experience (QoE). In this article, we propose a mobility-aware proactive edge caching scheme (MSTPS), where the spatial and temporal prediction of vehicles are taken into account for content deployment and scheduling. Specifically, we optimize the caching decision based on predicting the vehicle’s driving trajectory and travel preference. The scheme learns the vehicle’s travel preferences to cope with mobility uncertainty by combining users with similar travel patterns. Meanwhile, the proposed scheme can support the sequential downloading of content chunks. Furthermore, in order to deal with the dynamic characteristics and unpredictable challenges of the IoV, we design a system recovery strategy, which can avoid the degradation of the proposed scheme due to the failure of prediction. Finally, by using real mobility data sets and scenarios, we explore the impact of the number of ENs deployed in advance for each vehicle’s request when the cache needs to be updated on system performance. In addition, we evaluate the effectiveness of the proposed scheme. Our proposed scheme can achieve the best cache hit ratio and decrease caching costs compared to the existing mobility-aware in-order caching schemes.

With the development of wireless communication technology, the amount of data in the Internet of Things has increased rapidly, and the application mode has become ubiquitous. The content sharing mode has the characteristics of diversified services, diversified contents, and diversified scenarios. Content sharing from device to device (D2D) can cope with the increasing traffic pressure on cellular networks. Users can use mobile devices to transmit and share content across spaces by opportunities. However, in the content sharing process, the available cache space for users is limited, and content may delay delivery in searching cooperative sharing users. Researching effective content sharing and forwarding algorithms can improve such a transmission environment. In the content sharing process in the mobile Internet of Things, users can analyze and judge the surrounding fields according to the shared content attributes and personalized preferences and search for suitable sharing targets. This paper proposes a content sharing algorithm based on dynamic behavior and collaborative prediction (DBCPNF). It establishes a user preference model based on the user’s historical behavior, cooperation users, and content attributes. By calculating the matching degree between content attributes and user preferences, users with matching higher are adopted as target users to share content cooperatively. Through experimental analysis and comparison with other algorithms, our algorithm has the best performance on the content delivery ratio and improves the network’s overall efficiency.

Abstract The rapid development of the Internet of things has led to the explosive development of data in various fields. Traditional routing protocols cannot effectively handle the reception and transmission of data. This makes it difficult to exchange and transmit information in the Internet of things. Therefore, the choice of data transmission methods is particularly important. In order to solve this problem, this paper proposes a data transmission mechanism based on social relationships, namely, the mobile node social relationship quantification (MNSRQ) algorithm, which analyses the social relationship characteristics of mobile nodes in the Internet of things, extracts decision‐making features to study the dynamics of social relationship, then combines information entropy and fuzzy clustering theory to quantify the social relationship, and then selects the relay node with strong social relationship for data transmission. Theoretical analysis and experimental results show that the performance of the MNSRQ algorithm is better than previous studies. Compared with epidemic algorithm, ICMT algorithm, spray and wait algorithm, and EIMST algorithm, the MNSRQ algorithm can reduce end‐to‐end transmission delay and routing overhead while maintaining the life of the network, effectively reduce energy consumption during transmission, and maintain a high data transmission success rate, with a transmission success rate of 0.7–0.9.

OBJECTIVE: Accurate segmentation and partitioning of lesions in PET images provide computer-aided procedures and doctors with parameters for tumour diagnosis, staging and prognosis. Currently, PET segmentation and lesion partitioning are manually measured by radiologists, which is time consuming and laborious, and tedious manual procedures might lead to inaccurate measurement results. Therefore, we designed a new automatic multiprocessing scheme for PET image pre-screening, noise reduction, segmentation and lesion partitioning in this study. PET image pre-screening can reduce the time cost of noise reduction, segmentation and lesion partitioning methods, and denoising can enhance both quantitative metrics and visual quality for better segmentation accuracy. For pre-screening, we propose a new differential activation filter (DAF) to screen the lesion images from whole-body scanning. For noise reduction, neural network inverse (NN inverse) as the inverse transformation of generalized Anscombe transformation (GAT), which does not depend on the distribution of residual noise, was presented to improve the SNR of images. For segmentation and lesion partitioning, definition density peak clustering (DDPC) was proposed to realize instance segmentation of lesion and normal tissue with unsupervised images, which helped reduce the cost of density calculation and completely deleted the cluster halo. The experimental results of clinical data demonstrate that our proposed methods have good results and better performance in noise reduction, segmentation and lesion partitioning compared with state-of-the-art methods.