The abundance of data provided by mobile edge devices enables a wide range of mobile edge computing (MEC) applications. Numerous studies have investigated efficient offloading methods for bandwidth savings in MEC. However, they focus on trading the device's computational cost for a reduction in communication, while edge devices can be rather resource-limited and must handle several jobs simultaneously. In this paper, the computation overhead on the device is pushed to its absolute minimum (almost no overhead), and consideration is given to enhancing the accuracy of the image recognition task within the constraints of the transmission volume limitation. We propose a mask-reconstruct system called MOT to mask images on the device side and recover images with the Masked Autoencoders (MAE)-based model on the server side. We further design a feedback-driven scheme to achieve content-aware transmission. Extensive experiments have been conducted to verify the effectiveness of the MOT.

Controlling drones with natural language instructions is an important topic in Vision-and-Language Navigation (VLN). However, previous models can not effectively guide drones with the integration of multimodal features, as few of them exploit the correlations between instructions and the environmental contexts and consider the model’s capacity to understand natural languages. Therefore, we propose a novel language-enhanced cross-modal model that has a conditional Transformer to effectively integrate the multimodal features, i.e., the textual instructions and visual contexts. To enhance the ability of language representation, we also employ SentenceBERT. In addition, to address the issue that users could provide various textual instructions even for the same navigation task, we propose a prompt-based approach by introducing an LLM-based intermediary component (LLMIR) for rephrasing users’ instructions. We evaluate our approaches with a quadcopter simulator. Our model improves the absolute task completion rate by 1.39%. To evaluate LLMIR, we create a new test set by extracting the essential and minimal instructions from the original test set. By using the LLM, the task completion rate improves by 1.51%. And it narrows the performance gap between new and original test set by 34.83%.

Sign language serves as a bridge between the hearing-impaired and other people. Existing sensor-based approaches tend to only collect data from the dominant hand. Does this signal collection method affect the accuracy of gesture recognition, especially gestures where the dominant hand has the same movement while the non-dominant hand has different movements? The specific gestures are called same-right-and-different-left (SRDL) where the right hand is dominant. This article is the first to propose an SRDL-aware sign language recognition system. First, an SRDL discriminator based on an autoencoder and range classifier is designed to determine whether the gesture is SRDL. Second, an SRDL feature selector based on clustering relationship is presented. Multivariate variational mode decomposition and fast fourier transform are used to obtain the feature expression. Moreover, a clustering relationship algorithm is proposed to dynamically select features for every group of SRDL gestures in the feature expression. Finally, the experimental results show that the average word error rate is 14.3% and decreases by 8.5% and 12.1% compared with Signspeaker and MyoSign, respectively.

The dramatically growing trend of vehicles equipped with driving camera recorders has allowed realizing real-time crowdsourced video sharing in vehicular edge computing (VEC). Such cameras can assist in monitoring objects directly in front of and behind the vehicles, enabling them to provide important visual information through real-time video streaming in case of possible accidents. Exploiting the on-board units (OBUs) for VEC can allow drivers and passengers to share and access on-road video surveillance services. However, data security and privacy concerns of video generators (owners) are two key challenges that can severely limit video sharing in a VEC environment. In this article, we propose a blockchain empowered publish/subscribe (P/S) scheme to enable one-to-many secure video sharing in the VEC scenario. Then, we design an attribute-based encryption algorithm with static and dynamic attributes (ABE-SD) to achieve fine-grained access control in a mobile environment. Finally, We utilize permissioned blockchain and smart contracts to record access policy and publish and subscribe events, thus resulting in user self-certification and event traceability. The numerical results indicate that our proposed scheme ABE-SD outperforms traditional centralized CP-ABE methods in terms of encryption and decryption performance. The simulation experiments demonstrated that the proposed video-sharing scheme is secure and efficient.

Unlike a traditional centralized and producer-controlled power grid, the smart grid is a more complicated distributed power system consisted of many resources and applications. In smart grid, huge amounts of data generated by edge devices are collected by different parties. To achieve high operation efficiency, it is important to enable the data sharing and cooperative computation among different parties. How to protect the security and privacy of the utility data and the identities of their owners has become a major concern. There have been some studies on this issue. However, most of these works failed to consider the privacy protection in the dual sides of the data owner and receiver. In this article, we propose BPM4SG, a blockchain-based dual-side privacy-preserving multiparty computation (MPC) scheme for edge-enabled smart grid. In BPM4SG, the data segmentation method is adopted to ensure the security of MPC (e.g., summation) in edge nodes. The consortium blockchain and smart contract are used to further increase the system security and avoid the dependency on trusted third parties. Additionally, a data obfuscation method based on the ring signatures and a new one-time address scheme are proposed to protect the privacy of both the data owner and data receiver. The analysis shows that BPM4SG can meet the security and privacy requirements of smart grid. The experimental evaluation results demonstrate that our scheme has a better performance compared with other popular schemes.

Environments built from the edge-based Industrial Internet of Things (IIoT) are maelstroms of information that continuously flows between heterogeneous data objects, such as sensors and devices, and edge nodes. However, the explosive growth in the number of IIoT data objects connected to edge nodes generally results in significant computation and storage requirements, which exceed those of resource-constrained edge nodes. The problem mentioned above causes a major concern related to efficient memory usage and faster communication processing in data acquisition. Therefore, in this article, we propose a multibuffers multiobjects (MBMOs) architecture to support the parallel delivery of multiple data objects to multiple variable-length buffer blocks. Furthermore, a mathematical model is established, aiming to find the optimum buffer according to the size of each communication data packet. For the aforementioned matching problem, a spatiotemporal resource allocation algorithm is designed to maximize memory usage while minimizing communication processing time. We implement our MBMO in a monitoring system in which embedded programmable logic controllers (ePLCs) serve as the edge nodes. The analyses illustrate that memory usage and time efficiency are improved greatly with the utilization of MBMO.

With the development of microelectronics and sensor technologies, there are more and more researchers applying them to human action recognition, most of which are professional motion and rely on specific-designed sensors and wearable de-vices. Meanwhile, the need of rehabilitation training is increasing due to occupational diseases, bad life-style and incorrect exercise habit. However, it is costly and inconvenient to train in clinics and hospitals. To buy or borrow a set of medical training equipment is also unpractical. In this paper, we propose to use smart phones, which have larger computing power and are equipped with richer sensors ever than before, to run artificial intelligence based models and algorithms for identification of rehabilitation actions. Beyond doubt, it will be more convenient to use smart phones instead of professional equipments. Nevertheless, there are still some challenges which prevent it from being put into practice, such as phone deployment, data collection, and model training. We initially conceptualize and implement a smart phone-based accuracy judgment system for rehabilitation action. According to the characteristics of the system, e.g., sensor difference, position variation, and computing power limitation, a supervised and data-sharing learning algorithm is proposed, the operation framework, loss function and regular expression function are carefully selected. The experiment on a prototype of the system verifies that the proposed method precisely identifies the rehabilitation actions of testees.

Facial micro-expressions are characterized by their extremely short duration and low intensity, can provide an important basis for judging people's emotions, and therefore have promising potential applications in numerous fields. This paper puts forward a novel method for recognizing microexpressions by using a quaternion-based sparse representation (QSR) model combined with the integral projection of difference energy image (IP-DEI) to extract features from color images of human faces . Using the quaternion model to jointly process color images can obtain greater feature information than gray or RGB images, and the QSR model helps reduce feature dimensions and enables greater discriminative representation. First, each microexpression sample undergoes IP-DEI to allow the features of all samples to be displayed in the form of a quaternion matrix Y. Next we find overcomplete dictionary matrix D and sparse coefficient matrix X such that Y = DX in ideal scenarios, and consider X̂̃̅̅̆̆̇ to be the features contained within the microexpression samples. Finally, we apply our method to the SMIC, CAMSE I and CAMSE II micro-expression databases while using SVM as classifier. The results of the experiment demonstrate that our method outperformed the currently most advanced methods in terms of micro-expression recognition accuracy.

Multi-robot services are widely used to improve the efficiency of industry Internet applications, especially in smart factories. Under the situation that the tasks are becoming more and more intensive, how can smart factories use limited robot resources to complete tasks more efficiently? In order to solve this problem, we transform it into a resource-constrained multiproject scheduling problem, and consider using a combinatorial auction method to get the solution. In order to ensure the security of the system and solve the transaction cost of the robot system, we adopt Blockchain technology and smart contracts to organize the work of the robots. We finally conducted performance analysis of our proposed method, and the results show that smart contracts and combined auction algorithms are safe and effective.

Facing globalized competition, there have been increasing requirements for safety and efficiency in smart factories, where the industrial Internet of Things can enable the monitoring of equipment's status and the detecting of faults before they go critical. Regarding cloud computing, data-driven methods running at clouds are adopted to train the model with a large amount of raw data at the beginning, then end machines upload their real-time readings to the cloud center for processing. However, this incurs considerable computational costs and may sometimes bear a severe delay. In this article, we consider a hierarchical structure where edge-PLCs are employed to gather sensed data locally and reduce communication costs. Since a single fault may be related to multiple influencing features, we want to first minimize the number of features that need to determine a fault, then try to find out the minimal set of edge-PLCs which can cover all key features so as to save the deployment cost. We propose a random-forest-based method to handle the features selection problem, and then the selection of edge-PLCs by solving the set coverage problem. Through the simulation on real data trace, we compare our method with other artificial-intelligence-based methods, such as the logistics regression model and its extensions. The results prove the efficiency and performance of the proposed method, which reaches or even exceeds the accuracy of methods using the full set of data.

Nowadays we are witnessing the formation of a massive Internet-of-Things (IoT) ecosystem that integrates a variety of wireless-enabled devices ranging from smartphones, wearables, and virtual reality facilities to sensors, drones, and connected vehicles. As IoT is penetrating every aspect of people’s life, work, and entertainment, an increasing number of IoT devices and the emerging IoT applications are driving exponential growth in wireless traffic in the foreseeable future. As a result, current IoT system architectures are facing significant challenges to handle millions of devices; thousands of servers; the transmission and processing of large volume of data, etc.

Nowadays, the availability of parking spaces is far behind the quick rising number of cars. Rather than building more lots, a better way is to share private-owned parking spaces. However, this faces the challenge that users are not willing to expose their privacy to the public. To solve this problem, we propose a new architecture for parking space sharing, integrating homomorphic cryptography into the design of a secure protocol for parking space searching and booking. The proposed privacy-preserving matching scheme (PPMS) is constructed in an untrusted third-party service system including two independent entities, namely, a server and an intermediary platform. Via the participant comparison protocol (PCP), a driver can choose from the matching result and be navigated to the parking space near his destination, without knowing any information of the provider and vice versa. In the meanwhile, in order to further improve the efficiency of matching, we also propose a block algorithm based on the longitude and latitude (BABLL), which utilizes a novel partitioning scheme. The feasibility of the architecture is validated through the detailed theoretical analysis and extensive performance evaluations, including the assessment of the resilience to attacks.

To fulfil the increasing charging requests of Electric Vehicles (EVs), various means have been proposed such as improving charging efficiency at charging stations, optimal locating of charging stations, enabling V2V trading and so on. Most existing work requires vehicles to be stationary while being charged. Mobile wireless charging is a promising trend to fix this problem since vehicles can perform moving and charging simultaneously. In this paper, a new concept called platoon-based charging is presented, which combines the energy-aware driving and mobile charging. In the design, an EV may have to detour to follow the platoon. To minimize the additional energy cost and delay brought by that, we studied the optimal assignment problem between EVs and charging platoons and converted it to a dynamic weight bipartite matching. The experiment results show that our algorithm outperforms the exiting ones.

In smart grid, large quantities of data is collected from various applications, such as smart metering substation state monitoring, electric energy data acquisition, and smart home. Big data acquired in smart grid applications is usually sensitive. For instance, in order to dispatch accurately and support the dynamic price, lots of smart meters are installed at user's house to collect the real-time data, but all these collected data are related to user privacy. In this paper, we propose a data aggregation scheme based on secret sharing with fault tolerance in smart grid, which ensures that control center gets the integrated data without revealing user's privacy. Meanwhile, we also consider fault tolerance during the data aggregation. At last, we analyze the security of our scheme and carry out experiments to validate the results.

Traditional vision based vehicular advertising methods can only support planar advertisements. If there are obstacles blocking the line-of-sight between the advertiser and customers, the efficiency drops quickly. With the proliferation of vehicular ad-hoc networks (VANETs), advertisements can be disseminated through wireless means. However, the utility of advertisements still decay over time so that the advertiser will require fast delivery to achieve higher rewards. In this paper, we consider a taxi based multimedia advertisements broadcasting scenario where the taxis act as the advertising sources. To solve the contradiction between limited communication capacity and big data size, we propose TaxiCast which can achieve very good performance. It first applies signal strength based coding and decoding to obtain the demandings of the surrounding vehicles. Then it solves the advertisements selection process as a knapsack problem. We also consider the reward decay of advertisements and conflicts between the taxis. We conduct simulations in both fixed reward case and decayed reward case. The result shows that our scheme achieves better performance than the existing strategies.

In Mobile Opportunistic Networks, there are special cases of multicast in which destinations (except the total number) are not predetermined. E.g., a person tries to find another three players to play poker without knowing them. Achieving efficient data routing in this scenario (target any k destinations among m of them) can be very challenge due to (1) no predetermined destinations, (2) extra delay cost by the destinations collision, and (3) difficulty of the receiver quantity control. We define this as a K-Anycast problem, the goal is to route copies of the message to any k destinations with the shortest average delay. In this paper, we first propose a matching method to solve the problem in a centralized way. We then bring forward a layered hierarchical structure where nodes are organized according to their degree of activities. Based on the structure, two routing algorithms are proposed where K-Cast initializes exactly k copies of the message without replication in the middle, K-Epidemic performs epidemic routing only in a controlled range. Both algorithms will first forward copies upwards along the structure and then downwards to the destinations. Experiments on real data trace show that the proposed algorithms achieve much better delay, delivery ratio and lower forwarding numbers.

In the largely populated city of a developing country, the ambulance service usually sends an emergent patient to the available hospital with shortest pre-consultation delay. The problem is, a life-critical patient may encounter the lack of treatment resource, such as sickbed, in desired hospitals, and the delay to a next appropriate hospital would cause his death, because non-critical patients already occupied the resources. In the worst case, the service encountering a catastrophe may hold hundreds of people on their way to the hospital and require sickbeds be reserved in advance. In this paper, we propose a resource allocation to balance delay in sending patients to hospitals. We extend the scheme to consider sickbed reservation along the time scale by estimating from the past records in history. As a result, the occupancy is balanced in order to reduce the risk of life-critical patients being delayed. Then we develop an in-hospital waiting queue to keep serious patients waiting locally, when it costs more to reach another available hospital. Simulation results show the substantial improvement of our approach in average delay and number of failure-of-assignment.

A wireless sensor network usually has a large amount of nodes deployed in a area to report ambient reading, to detect abnormal events, or to monitor the region. Each sensor performs several tasks, such as computing, sensing and communicating. A node equipped with multiple sensors is able to participate in many sensing tasks to improve measurement accuracy. However, the contribution a new node can make to the corresponding sensing accuracy depends on the number of existing nodes while the energy cost will increase in a consistent way. Therefore, there is a tradeoff between the task accuracy and the energy cost. In this paper, we consider the dynamic fading fact of contribution of redundant sensor nodes and bring forward a Contribution Aware Task Allocation method to maximize the total accuracy efficiency of sensor network. The method is derived from max-weight resource allocation algorithm (i.e., KM algorithm) and can guarantee the optimal of the solution. We compare our method with two other greedy and optimizing method. The experimental result shows our method outperform the competitors and are more efficiency.

The purpose of this paper is to represent a wind powered wireless sensor network system and introduce a novel transmission power control scheme based on remaining energy level and energy harvesting status to extend the lifetime of WSNs. Energy constraint has always been one of the most significant problems of wireless sensor networks along with the development. Many methods have been introduced to solve this problem, basically in two aspects: energy management and energy harvesting. In this paper, a sensor network system has been developed which uses wind power as energy harvesting resource and ultra-capacitor as energy storage. By analyzing the power recharging, leakage and energy consumption rate, a novel Energy Level based Transmission Power Control scheme (EL-TPC) is produced. In EL-TPC scheme, the transmission power is classified into three levels which correspond to specified communication requirements. By adapting the nodes' operation pattern, hierarchical network architecture can be formed, which prioritizes the use of high energy level, fast charging and leaking nodes to save the energy of uncharged nodes. The scheme is implemented in a Building Surface mounted, Wind Power collected Wireless Sensor Network system called BSWPWSN, which aims to monitoring the usage pattern of air conditioners and the outdoor temperature. The results show that EL-TPC scheme can significantly balance the energy consumption in different nodes and extend the entire network lifetime. The overall energy level of the network keeps a dynamic balance during the experiment, which indicates that the network will not lose effect due to energy constraint.

Wireless sensor networks(WSNs) have been increasingly available for large-scale applications in which energy efficiency is an important performance measure. These applications include environmental monitoring and structure monitoring which demand multifarious data. Driven by the energy limitation nature of WSNs lots of research works have been done in aspects such as nodes deployment, routing protocol, topology control, data reduction, sleep scheduling, etc. However, heterogeneous, i.e. hybrid sensor nodes are combined together into semantic sensor networks to provide large-scale applications with content rich information. In this paper, we discuss the potential of energy efficiency that semantization could bring to sensor networks. First we have an overview of some related work and then address current approaches of energy conservation in WSNs as well as how semantization can contribute in each aspect of saving energy. Finally a recommendatory architecture of semantic sensor network is proposed. Semantization will be a promising solution to improve energy efficiency together with system performance.

In this paper, we introduce a novel Web service based cooperative workflow model which not only greatly improves the efficiency of concurrent processing but also makes results combination easier. Web service based workflows provide support for aggregating Web services into new higher-level Web services by means of process composition. We add states and actions to the activity, and realize the cooperation between different activities. Petri-nets are used to analyze and verify the proposed model. A cooperative businesses process is also modeled by Petri-nets as a demonstration. In addition, we realize the model in an open source cooperative workflow software-Bonita and test it with the ActiveBPEL engine. Then we apply the model to a cooperative car design scenario. The results indicate that the proposed model is valid and effective.