Due to the expanding scale of vehicles and the new demands of multimedia services, current vehicular networks face challenges to increase capacity, support mobility, and improve QoE. An innovative ...design of next generation vehicular networks based on the content-centric architecture has been advocated recently. However, the details of the framework and related algorithms have not been sufficiently studied. In this article, we present a novel framework of a content-centric vehicular network (CCVN). By introducing a content-centric unit, contents exchanged between vehicles can be managed based on their naming information. Vehicles can send interests to obtain wanted contents instead of sending conventional information requests. Then we present an integrated algorithm to deliver contents to vehicles with the help of content-centric units. Contents can be stored according to their priorities determined by vehicle density and content popularity. Pending interests are updated based on the analysis of transmission ratio and network topology. The location of a content-centric unit to provide content during the moving of vehicles is determined by the forwarding information. Finally, simulation experiments are carried out to show the efficiency of the proposed framework. Results indicate that the proposed framework outperforms the existing method and is able to deliver contents more efficiently.
Vehicular content networks (VCNs), which distribute medium-volume contents to vehicles in a fully distributed manner, represent the key enabling technology of vehicular infotainment applications. In ...VCNs, the road-side units (RSUs) cache replicas of contents on the edge of networks to facilitate the timely content delivery to driving-through vehicles when requested. However, due to the limited storage at RSUs and soaring content size for distribution, RSUs can only selectively cache content replicas. The edge caching scheme in RSUs, therefore, becomes a fundamental issue in VCNs. This paper addresses the issue by developing an edge caching scheme in RSUs. Specifically, we first analyze the features of vehicular content requests based on the content access pattern, vehicle's velocity, and road traffic density. A model is then proposed to determine whether and where to obtain the replica of content when the moving vehicle requests it. After this, a cross-entropy-based dynamic content caching scheme is proposed accordingly to cache the contents at the edge of VCNs based on the requests of vehicles and the cooperation among RSUs. Finally, the performance of the proposed scheme is evaluated by extensive simulation experiments.
With the rapid advances in vehicular technologies and social multimedia applications, VSNs have emerged and gained significant attention from both industry and academia. However, due to the low ...communication ability between vehicles, heavy network traffic load, and limited storage capacity, VSNs face the challenge to improve the performance of content delivery to provide a pleasant and safe driving experience. Therefore, in this article, first we present a novel framework to deliver content in VSNs with D2D communication. In the proposed framework, moving vehicles can exchange content directly with each other according to D2D communication. All contents are managed in a content-centric mode, where moving vehicles can send their interests to obtain content with naming information, resulting in a reduction of network traffic load. Based on the D2D communication, parked vehicles around the street can form vehicular social communities with the moving vehicles passing along the road, where the storage capacity of VSNs can be increased by using the contents in parked vehicles. Then we present the detailed process of content delivery in VSNs including interest sending, content distribution, and content replacement. Finally, experiment results prove the efficiency of the proposal.
Recently, parked vehicles have been shown to be useful to deliver content in vehicular ad hoc networks, where the parked vehicles can form social communities to share and exchange content with other ...moving vehicles and road side units (RSUs). However, as it takes resource such as bandwidth and power for parked vehicles and RSUs to deliver content, the incentive scheme with the optimal pricing strategy needs to be studied. Furthermore, because multiple places including RSUs and parked vehicles can deliver content to moving vehicles, the optimal algorithm to determine where to obtain the requested content should also be discussed. Therefore, in this paper, we first propose a framework of content delivery with parked vehicles, where moving vehicles can obtain content from both the RSU and parked vehicles according to the competition and cooperation among them. Then, based on a Stackelberg game, we develop a pricing model where each of the three players, including moving vehicles, RSU, and parked vehicles, can obtain their maximum utilities. Next, a gradient based iteration algorithm is presented to obtain the Stackelberg equilibrium. Finally, the simulation results prove that the proposal can outperform other conventional methods and that each player in the game can obtain its optimal strategy during the content delivery.
Content dissemination, in particular, small-volume localized content dissemination, represents a killer application in vehicular networks, such as advertising distribution and road traffic alerts. ...The dissemination of contents in vehicular networks typically relies on the roadside infrastructure and moving vehicles to relay and propagate contents. Due to instinct challenges posed by the features of vehicles (mobility, selfishness, and routes) and limited communication ability of infrastructures, to efficiently motivate vehicles to join in the content dissemination process and appropriately select the relay vehicles to satisfy different transmission requirements is a challenging task. This paper develops a novel edge-computing-based content dissemination framework to address the issue, composed of two phases. In the first phase, the contents are uploaded to an edge computing device (ECD), which is an edge caching and communication infrastructure deployed by the content provider. By jointly considering the selfishness and the transmission capability of vehicles, a two-stage relay selection algorithm is designed to help the ECD selectively deliver the content through vehicle-to-infrastructure (V2I) communications to satisfy its requirements. In the second phase, the vehicles selected by the ECD relay the content to the vehicles that are interested in the content during the trip to destinations via vehicle-to-vehicle (V2V) communications, where the efficiency of content delivery is analyzed according to the probability that vehicles encounter on the path. Using extensive simulations, we show that our framework disseminates contents to vehicles more efficiently and brings more payoffs to the content provider than the conventional methods.
The heterogeneous vehicular networks (HetVNETs), which apply the heterogeneous access technologies (e.g., cellular networks and WiFi) complementarily to provide seamless and ubiquitous connections to ...vehicles, have emerged as a promising and practical paradigm to enable vehicular service applications on the road. However, with different costs in terms of latency time and price, how to optimize the connection along the vehicle's trip toward the lowest cost represents fundamental challenges. This paper investigates the issue by proposing an optimal access control scheme for vehicles in HetVNETs. In specific, with different access networks, we first model the cost of each vehicle to download the requested content by jointly considering the vehicle's requirements of the requested content and the features of the available access networks, including conventional vehicle to vehicle communication and the heterogeneous access technologies. A coalition formation game is then introduced to formulate the cooperation among vehicles based on their different interests (contents cached in vehicles) and requests (contents to be downloaded). After forming the coalitions, vehicles in the same coalition can download their requested contents cooperatively by selecting the optimal access network to achieve the minimum costs. The simulation results demonstrate that the proposed game approach can lead to the optimal strategy for the vehicle.
Recently, the internet-of-things (IoT) has emerged as a new paradigm with an ever-increasing number of things to be connected to the internet. Different from the conventional paradigms, in the IoT ...the data computing scheme is needed to efficiently collect and offer data to provide sensing service. However, the existing data computing schemes are unfriendly which lack the integrated and incentive consideration to reduce the cost of data collection and encourage more participants for cooperation. Therefore, in this paper we propose a utility-based data computing scheme which allows vehicles to collect mobile data in the urban area, in order to provide sensing service in the IoT. First, we present an integrated architecture by introducing roadside buffers where each buffer can have a sink node to collect sensor data from vehicles. Next, by considering both the time cost and power cost during the data collection, we make the analysis of utilities in data computing process. Then, with a bargaining game to model the interaction among participants, a utility based data computing scheme is proposed with incentives where the optimal price can be determined for sensing service. Finally, extensive simulation experiments prove that the proposed scheme can efficiently improve the sensing service in IoT with a low cost.
To the best of our knowledge, applying adaptive three-dimensional lookup tables (3D LUTs) to underwater image enhancement is an unprecedented attempt. It can achieve excellent enhancement results ...compared to some other methods. However, in the image weight prediction process, the model uses the normalization method of Instance Normalization, which will significantly reduce the standard deviation of the features, thus degrading the performance of the network. To address this issue, we propose an Instance Normalization Adaptive Modulator (INAM) that amplifies the pixel bias by adaptively predicting modulation factors and introduce the INAM into the learning image-adaptive 3D LUTs for underwater image enhancement. The bias amplification strategy in INAM makes the edge information in the features more distinguishable. Therefore, the adaptive 3D LUTs with INAM can substantially improve the performance on underwater image enhancement. Extensive experiments are undertaken to demonstrate the effectiveness of the proposed method.
Building extraction is a popular topic in remote sensing image processing. Efficient building extraction algorithms can identify and segment building areas to provide informative data for downstream ...tasks. Currently, building extraction is mainly achieved by deep convolutional neural networks (CNNs) based on the U-shaped encoder–decoder architecture. However, the local perceptive field of the convolutional operation poses a challenge for CNNs to fully capture the semantic information of large buildings, especially in high-resolution remote sensing images. Considering the recent success of the Transformer in computer vision tasks, in this paper, first we propose a shifted-window (swin) Transformer-based encoding booster. The proposed encoding booster includes a swin Transformer pyramid containing patch merging layers for down-sampling, which enables our encoding booster to extract semantics from multi-level features at different scales. Most importantly, the receptive field is significantly expanded by the global self-attention mechanism of the swin Transformer, allowing the encoding booster to capture the large-scale semantic information effectively and transcend the limitations of CNNs. Furthermore, we integrate the encoding booster in a specially designed U-shaped network through a novel manner, named the Swin Transformer-based Encoding Booster- U-shaped Network (STEB-UNet), to achieve the feature-level fusion of local and large-scale semantics. Remarkably, compared with other Transformer-included networks, the computational complexity and memory requirement of the STEB-UNet are significantly reduced due to the swin design, making the network training much easier. Experimental results show that the STEB-UNet can effectively discriminate and extract buildings of different scales and demonstrate higher accuracy than the state-of-the-art networks on public datasets.
With the development of sensors and of the Internet of Things (IoT), smart cities can provide people with a variety of information for a more convenient life. Effective on-street parking availability ...prediction can improve parking efficiency and, at times, alleviate city congestion. Conventional methods of parking availability prediction often do not consider the spatial–temporal features of parking duration distributions. To this end, we propose a parking space prediction scheme called the hybrid spatial–temporal graph convolution networks (HST-GCNs). We use graph convolutional networks and gated linear units (GLUs) with a 1D convolutional neural network to obtain the spatial features and the temporal features, respectively. Then, we construct a spatial–temporal convolutional block to obtain the instantaneous spatial–temporal correlations. Based on the similarity of the parking duration distributions, we propose an attention mechanism called distAtt to measure the similarity of parking duration distributions. Through the distAtt mechanism, we add the long-term spatial–temporal correlations to our spatial–temporal convolutional block, and thus, we can capture complex hybrid spatial–temporal correlations to achieve a higher accuracy of parking availability prediction. Based on real-world datasets, we compare the proposed scheme with the benchmark models. The experimental results show that the proposed scheme has the best performance in predicting the parking occupancy rate.
PDF