Search

Attention mechanism has been a successful method for multimodal affective analysis in recent years. Despite the advances, several significant challenges remain in fusing language and its nonverbal context information. One is to generate sparse attention coefficients associated with acoustic and visual modalities, which helps locate critical emotional semantics. The other is fusing complementary cross‐modal representation to construct optimal salient feature combinations of multiple modalities. A Conditional Transformer Fusion Network is proposed to handle these problems. Firstly, the authors equip the transformer module with CNN layers to enhance the detection of subtle signal patterns in nonverbal sequences. Secondly, sentiment words are utilised as context conditions to guide the computation of cross‐modal attention. As a result, the located nonverbal features are not only salient but also complementary to sentiment words directly. Experimental results show that the authors' method achieves state‐of‐the‐art performance on several multimodal affective analysis datasets. [ABSTRACT FROM AUTHOR]

Summary: Deep learning has shown its great potential in real‐world applications. However, users(clients) who want to use deep learning applications need to send their data to the deep learning service provider (server), which can make the client's data leak to the server, resulting in serious privacy concerns. To address this issue, we propose a protocol named EPIDL to perform efficient and secure inference tasks on neural networks. This protocol enables the client and server to complete inference tasks by performing secure multi‐party computation (MPC) and the client's private data is kept secret from the server. The work in EPIDL can be summarized as follows: First, we optimized the convolution operation and matrix multiplication, such that the total communication can be reduced; Second, we proposed a new method for truncation following secure multiplication based on oblivious transfer and garbled circuits, which will not fail and can be executed together with the ReLU activation function; Finally, we replace complex activation function with MPC‐friendly approximation function. We implement our work in C++ and accelerate the local matrix computation with CUDA support. We evaluate the efficiency of EPIDL in privacy‐preserving deep learning inference tasks, such as the time to execute a secure inference on the MNIST dataset in the LeNet model is about 0.14 s. Compared with the state‐ofthe‐art work, our work is 1.8×$$ \times $$–98×$$ \times $$ faster over LAN and WAN, respectively. The experimental results show that our EPIDL is efficient and privacy‐preserving. [ABSTRACT FROM AUTHOR]

Summary: While data is distributed in multiple edge devices, federated learning (FL) is attracting more and more attention to collaboratively train a machine learning model without transferring raw data. FL generally exploits a parameter server and a large number of edge devices during the whole process of the model training, while several devices are selected in each round. However, straggler devices may slow down the training process or even make the system crash during training. Meanwhile, other idle edge devices remain unused. As the bandwidth between the devices and the server is relatively low, the communication of intermediate data becomes a bottleneck. In this article, we propose time‐efficient asynchronous federated learning with sparsification and quantization, that is, TEASQ‐Fed. TEASQ‐Fed can fully exploit edge devices to asynchronously participate in the training process by actively applying for tasks. We utilize control parameters to choose an appropriate number of parallel edge devices, which simultaneously execute the training tasks. In addition, we introduce a caching mechanism and weighted averaging with respect to model staleness to further improve the accuracy. Furthermore, we propose a sparsification and quantitation approach to compress the intermediate data to accelerate the training. The experimental results reveal that TEASQ‐Fed improves the accuracy (up to 16.67% higher) while accelerating the convergence of model training (up to twice faster). [ABSTRACT FROM AUTHOR]

Semi‐quantum secure summation allows multiple parties with limited quantum capabilities to perform collective calculations under the premise of protecting privacy. However, current approaches are confined to low‐dimensional quantum states as information carriers and are restricted to scenarios involving two or three parties. This limitation hinders their universality and ability to fulfill the requirements of practical secure multiparty computations. To address these challenges, a novel semi‐quantum secure multiparty summation (SQSMS) protocol is proposed based on high‐dimensional entangled states and single‐particle states. The protocol leverages the entanglement correlation to guarantee the security of the transmission, enabling resistance against various potential attacks. Then, by utilizing the IBM Qiskit platform, corresponding quantum circuits and simulation results are presented to validate the feasibility and correctness of the protocol. Furthermore, building upon the foundation of the SQSMS protocol, semi‐quantum anonymous auction and anonymous ranking protocols are further constructed. Compared with previous approaches, the designed protocols cater to the requirements of their respective counterparts. Consequently, the approach may have broad applications in semi‐quantum secure multiparty computation scenarios. [ABSTRACT FROM AUTHOR]

The development of emerging information technologies, such as the Internet of Things (IoT), edge computing, and blockchain, has triggered a significant increase in IoT application services and data volume. Ensuring satisfactory service quality for diverse IoT application services based on limited network resources has become an urgent issue. Generalized processor sharing (GPS), functioning as a central resource scheduling mechanism guiding differentiated services, stands as a key technology for implementing on-demand resource allocation. The performance prediction of GPS is a crucial step that aims to capture the actual allocated resources using various queue metrics. Some methods (mainly analytical methods) have attempted to establish upper and lower bounds or approximate solutions. Recently, artificial intelligence (AI) methods, such as deep learning, have been designed to assess performance under self-similar traffic. However, the proposed methods in the literature have been developed for specific traffic scenarios with predefined constraints, thus limiting their real-world applicability. Furthermore, the absence of a benchmark in the literature leads to an unfair performance prediction comparison. To address the drawbacks in the literature, an AI-enabled performance benchmark with comprehensive traffic-oriented experiments showcasing the performance of existing methods is presented. Specifically, three types of methods are employed: traditional approximate analytical methods, traditional machine learning-based methods, and deep learning-based methods. Following that, various traffic flows with different settings are collected, and intricate experimental analyses at both the feature and method levels under different traffic conditions are conducted. Finally, insights from the experimental analysis that may be beneficial for the future performance prediction of GPS are derived. [ABSTRACT FROM AUTHOR]

Time series data mining techniques have attracted extensive attention from researchers worldwide. Of these techniques, time series classification is an important part of time series mining. Among the many time series classification algorithms, methods based on the bag-of-patterns algorithm have attracted much attention from researchers because of their high accuracy and execution efficiency. However, when using these methods, only the frequency of different patterns is considered. Features such as the position of patterns in a sequence are not mined. Therefore, the aim of this paper is to determine how to solve the problem that the positional relationships among patterns are ignored when using the bag-of-patterns algorithm. To solve this issue, we introduce the graph embedding technique, and an attempt is made to capture the positional relationships among the patterns of time series from the graph perspective. To verify the performance of the method, we perform extensive experiments with the UCR time series archive, and the experimental results demonstrate that our proposed method generally improves the classification ability of models based on the bag-of-patterns algorithm. [ABSTRACT FROM AUTHOR]

Server logs are always essential in website maintenance. The services maintained may encounter various problems, and the logs generated by the server are the core data for locating the problem. However, existing log storage systems are susceptible to single points of failure, which can lead to potential log leakage. Meanwhile, external adversaries can tamper with logs to avoid tracking, and administrators can also tamper with logs to evade responsibility. In order to solve these tough shortcomings effectively, introducing blockchain networks seems to be an appropriate solution due to their immutability and decentralized nature. Traditional blockchains lack scalability in storing large files, but this can be tackled by storing log data in the interplanetary file system (IPFS). In this paper, we propose a log data storage scheme combined with blockchain networks and the IPFS technique. First, we utilize the IPFS to store extensive file log data and combine blockchain systems to realize decentralized log data storage in a secure manner. Second, to optimize the query efficiency of log data, we propose two improved Merkle tree methods, providing rapid queries of log data with a timestamp as query keywords (only one-third of the BBT in range query time). Finally, our comprehensive experimental results show that the two proposed methods have better log retrieval efficiency. [ABSTRACT FROM AUTHOR]

Emotion plays a critical role in making the computer more human-like. As the first and most essential step, emotion recognition emerges recently as a hot but relatively nascent topic, i.e., current research mainly focuses on single modality (e.g., facial expression) while human emotion expressions are multi-modal in nature. To this end, we propose an unobtrusive emotion recognition system leveraging two emotion-rich and tightly-coupled modalities, i.e., gesture and facial expression. The system design faces two major challenges, namely, how to capture the emotional expression in both modalities without disturbing the subject and how to leverage the relationship between modalities for recognizing the emotion. For the former, we explore WiFi and vision for unobtrusive and contactless gesture and facial expression sensing, respectively. For the latter, we propose a novel deep learning framework named Multi-Source Learning (MSL) to efficiently exploit both self-correlation in the modality and cross-correlation between modalities for fine-grained emotion recognition. To evaluate the proposed method, we prototype the system on low-cost commodity WiFi and vision devices, build a first-of-its-kind WiFi-Vision emotion dataset, and conduct extensive experiments. Empirical results not only verify the effectiveness of WiFE in emotion recognition, but also confirm the superiority of multi-modality over single-modality. [ABSTRACT FROM AUTHOR]

Mediated semi-quantum key distribution (M-SQKD) allows two "semi-quantum" or "classical" users to establish a secret key with the help of a third party (TP), in which TP has full quantum power and possibly adversarial. It is the basis of the multi-party semi-quantum cryptography protocol for securing the private data of classical users. In this work, we propose a M-SQKD protocol based on circular transport structure and prove the protocol is unconditional security. Even in the worst case (i.e., TP is untrusted), the protocol's noise tolerance is also close to the BB84 protocol. The results show that our protocol can reach the similar level of security as the full quantum protocol. Besides, due to the scalability of the circular transport structure, our protocol is easily extended to multi-party scenarios, which offers an approach to realizing multiple "classical" users' key distribution. Therefore, it can be used to protect the private data of multiple classical users and may have potential application scenarios in the future. [ABSTRACT FROM AUTHOR]

The understanding of long-range pixel–pixel dependencies plays a vital role in image segmentation. The use of a CNN plus an attention mechanism still has room for improvement, since existing transformer-based architectures require many thousands of annotated training samples to model long-range spatial dependencies. This paper presents a smooth attention branch (SAB), a novel architecture that simplifies the understanding of long-range pixel–pixel dependencies for biomedical image segmentation in small datasets. The SAB is essentially a modified attention operation that implements a subnetwork via reshaped feature maps instead of directly calculating a softmax value over the attention score for each input. The SAB fuses multilayer attentive feature maps to learn visual attention in multilevel features. We also introduce position blurring and inner cropping specifically for small-scale datasets to prevent overfitting. Furthermore, we redesign the skip pathway for the reduction of the semantic gap between every captured feature of the contracting and expansive path. We evaluate the architecture of U-Net with the SAB (SAB-Net) by comparing it with the original U-Net and widely used transformer-based models across multiple biomedical image segmentation tasks related to the Brain MRI, Heart MRI, Liver CT, Spleen CT, and Colonoscopy datasets. Our training set was made of random 100 images of the original training set, since our goal was to adopt attention mechanisms for biomedical image segmentation tasks with small-scale labeled data. An ablation study conducted on the brain MRI test set demonstrated that every proposed method achieved an improvement in biomedical image segmentation. Integrating the proposed methods helped the resulting models consistently achieve outstanding performance on the above five biomedical segmentation tasks. In particular, the proposed method with U-Net improved its segmentation performance over that of the original U-Net by 13.76% on the Brain MRI dataset. We proposed several novel methods to address the need for modeling long-range pixel–pixel dependencies in small-scale biomedical image segmentation. The experimental results illustrated that each method could improve the medical image segmentation accuracy to various degrees. Moreover, SAB-Net, which integrated all proposed methods, consistently achieved outstanding performance on the five biomedical segmentation tasks. [ABSTRACT FROM AUTHOR]

The global population is aging due to many factors, including longer life expectancy through better healthcare, changing diet, physical activity, etc. We are also witnessing various frequent epidemics as well as pandemics. The existing healthcare system has failed to deliver the care and support needed to our older adults (seniors) during these frequent outbreaks. Sophisticated sensor-based in-home care systems may offer an effective solution to this global crisis. The monitoring system is the key component of any in-home care system. The evidence indicates that they are more useful when implemented in a non-intrusive manner through different visual and audio sensors. Artificial Intelligence (AI) and Computer Vision (CV) techniques may be ideal for this purpose. Since the RGB imagery-based CV technique may compromise privacy, people often hesitate to utilize in-home care systems which use this technology. Depth, thermal, and audio-based CV techniques could be meaningful substitutes here. Due to the need to monitor larger areas, this review article presents a systematic discussion on the state-of-the-art using depth sensors as primary data-capturing techniques. We mainly focused on fall detection and other health-related physical patterns. As gait parameters may help to detect these activities, we also considered depth sensor-based gait parameters separately. The article provides discussions on the topic in relation to the terminology, reviews, a survey of popular datasets, and future scopes. [ABSTRACT FROM AUTHOR]

Despite wearable electronic devices having enormous potential applications, the recording quality has not sufficiently met the requirements for fetal disease detection. In this article, we aim to propose a novel framework for fetal electrocardiography (ECG) extraction, where the hybrid approach mainly considers the combination of a novel adaptive maternal QRS removal (AQRSR) algorithm and a tensor-based joint blind source separation (JBSS) approach. To adapt to the dynamic change of the ECG signal and suppress the powerful maternal heartbeat, AQRSR is explored with an adaptive template, which is generated by a model consisting of a generator, discriminator, and transformer. By replacing the oldest segmentation with the newest cycle, the approach is capable of quickly adapting to the estimated output to match the new input. In addition, due to the significant crossover between the target signal and the noise, even if the maternal ECG removal processing is implemented, the residual signal inevitably contains fetal ECG, part of maternal ECG, and noise. As a consequence of this, a novel JBSS that incorporates tensor decomposition is formulated to separate fetal heartbeat from long-term recordings contaminated with maternal movements noise and heartbeat interference. By dividing the collected data into multiple segments, this method can extract fetal ECG signals by exploiting the signals that are not only statistically independent within each segment but can be dependent on different segments. Experimental results show that even if the recorded signals are contaminated by the noise, the proposed framework has comparable better performance on fetal ECG extraction. [ABSTRACT FROM AUTHOR]

Vehicular Ad hoc Networks (VANETs), as the integration of mobile vehicle and intelligent technology, has raised plenty of attentions. In VANETs, mobile smart vehicles can timely get information from surrounding environment, which brings great convenience for society. However, this highlights the need to improve data quality while protecting privacy for vehicular data provider. Thus, in this article, we propose an AI-based Trust-aware and Privacy-preserving System (ATPS) to preserve privacy for vehicular data providers while improving quality of data collections in VANETs. Our proposed ATPS system mainly consists of two schemes: 1) a Partial Ordering based Trust Management (POTM) scheme and 2) a Trajectory Privacy Preserving (TPP) scheme jointly designed by Wasserstein Generative Adversarial Networks (WGAN) and differential privacy. The POTM scheme uses partial ordering relationship to accurately evaluate and manage trusts for vehicular data providers with the assistance of fully trusted drones, and selects the vehicles with top ranks to collect data. Then, TPP scheme skillfully incorporates WGAN and differential privacy to preserve trajectory privacy for vehicular data providers in VANETs, which also guarantees data availability by adding carefully designed noise to the original trajectory. Compared to existing scheme, extensive experiments conducted on the real-world datasets demonstrates efficiency of our ATPS in terms of improving the data quality by 45.76% to 52.57%, reducing the malicious vehicle participants by 15.48% to 16.95%, preserving privacy of vehicles, and guaranteeing data availability. [ABSTRACT FROM AUTHOR]

Mobile edge computing (MEC) has been recognized as a viable technology to satisfy low-delay computation requirements for resource-constrained Internet of things (IoT) devices. Nevertheless, the broadcast feature of wireless electromagnetic communications may lead to the security threats to IoT devices. In order to enhance the task offloading security, this paper proposes a reconfigurable intelligent surface (RIS)-assisted secure MEC network framework. Furthermore, we investigate the max-min computation efficiency problem under the secure computation rate requirements, by jointly optimizing the local computing frequencies and transmission power of IoT devices, time-slot assignment, and phase beamforming of the RIS. To solve the formulated non-convex problem, we further develop an iterative algorithm, in which the Dinkelbach-type method and block coordinate descent (BCD) technique are utilized to tackle the fractional objective function and coupled optimization variables, respectively. In particular, the successive convex approximation (SCA) and penalty function-based methods are exploited to solve the transmit power control and reflecting beamforming optimization subproblems, respectively, and the closed-form expression for local computing frequencies optimization subproblem is derived. Numerical results quantify the performance gain achieved by the proposed RIS-assisted secure MEC networks, when compared to existing benchmark methods. [ABSTRACT FROM AUTHOR]

Efficient transmission resource allocation in medium access control (MAC) is essential for ensuring high performance in vehicular ad hoc networks (VANETs). However, existing solutions cannot satisfy the requirements of high throughput and low delay in high-density environments. In this article, we propose a timeslot-reserved MAC protocol to meet such demands for VANETs. This protocol provides distributed resource allocation and has two striking features: the high efficiency of handshake negotiation and the different solutions to disparate delay requirements of packets. In this protocol, a novel neighbor information table adds a new field of the needed number of frames in the future multiple frames. This reflects the reservation idea and reduces the number of negotiations between neighbor vehicles. The service applicant without enough timeslot resources is excluded from resource allocation negotiation, to decrease the invalid resource allocation. The resource-insufficient service channel and the resource-insufficient timeslot segments are also excluded from packet transmission, to further reduce the invalid resource allocation. Delay-sensitive packets and delay-tolerant packets select different kinds of timeslot segments, to satisfy different delay demands of different packets. Simulation results demonstrate the advantages of the proposed protocol in terms of throughput, delay, and packet loss rate. [ABSTRACT FROM AUTHOR]

User authentication plays a critical role in access control of a man-machine system, where the knowledge factor, such as a personal identification number, constitutes the most widely used authentication element. However, knowledge factors are usually vulnerable to the spoofing attack. Recently, the inheritance factor, such as fingerprints, emerges as an efficient alternative resilient to malicious users, but it normally requires special equipment. To this end, in this article, we propose WiPass, a device-free authentication system only leveraging the pervasive Wi-Fi infrastructure to explore keystroke dynamics (manner and rhythm of keystrokes) captured by the channel state information to recognize legitimate users while rejecting spoofers. However, it remains an open challenge to characterize the behavioral features hidden in the human subtle motions, such as keystrokes. Therefore, we build a signal enhancement model using Ricean distribution to amplify user keystroke dynamics and a hybrid learning model for user authentication, which consists of two parts, i.e., convolutional neural network based feature extraction and support vector machine based classification. The former relies on visualizing the channel responses into time-series images to learn the behavioral features of keystrokes in energy and spectrum domains, whereas the latter exploits such behavioral features for user authentication. We prototype WiPass on the low-cost off-the-shelf Wi-Fi devices and verify its performance. Empirical results show that WiPass achieves on average 92.1% authentication accuracy, 5.9% false accept rate, and 6.3% false reject rate in three real environments. [ABSTRACT FROM AUTHOR]

Virtual reality (VR) technology is gaining popularity in a variety of fields, including education, games, movies, medicine, and engineering, to name a few. 360$^\circ$∘ VR video could provide an immersive experience and attract more researchers' and developers' attention. Some literature focused on the head movement when users watched 360$^\circ$∘ videos and released head tracking datasets. With the popularity of VR games, how the game contexts influence players' head movement and the effect of head movement on VR sickness is a topic worth studying. In this article, we collected a head movement dataset of 30 participants while playing five different VR games (Aircar, Beat Saber, Moss, Arizona Sunshine, and SUPERHOT), and the participants filled the Simulator Sickness Questionnaire (SSQ) after playing VR games. Finally, we analyzed the SSQ scores and the impact of VR games on VR sickness. To the best of our knowledge, this is the first available head movement trajectory dataset based on playing several types of VR games. [ABSTRACT FROM AUTHOR]

Vehicular Edge Computing (VEC) is a promising paradigm that leverages the vehicles to offload computation tasks to the nearby VEC server with the aim of supporting the low latency vehicular application scenarios. Incentivizing VEC servers to participate in computation offloading activities and make full use of computation resources is of great importance to the success of intelligent transportation services. In this paper, we formulate the competitive interactions between the VEC servers and vehicles as a two-stage Stackelberg game with the VEC servers as the leader players and the vehicles as the followers. After obtaining the full information of vehicles, the VEC server calculates the unit price of computation resource. Given the unit prices announced by VEC server, the vehicles determine the amount of computation resource to purchase from VEC server. In the scenario that vehicles do not want to share their computation demands, a deep reinforcement learning based resource management scheme is proposed to maximize the profits of vehicles and VEC server. The extensive experimental results have demonstrated the effectiveness of our proposed resource management scheme based on Stackelberg game and deep reinforcement learning. [ABSTRACT FROM AUTHOR]

As the inevitable part of intelligent service in the new era, the services for AI tasks themselves have received significant attention, which due to the urgency of energy and computing resources, is difficult to implement in a stable and widely distributed system and coordinately utilize remote edge devices and cloud. In this article, we introduce an AI-based holistic network optimization solution to schedule AI services. Our proposed deep Q-learning algorithm optimizes the overall throughput of AI co-inference tasks themselves by balancing the uneven computation resources and traffic conditions. We use a multi-hop DAG (Directed Acyclic Graph) to describe a deep neural network (DNN) based co-inference network structure and introduce the virtual queue to analyze the Lyapunov stability for the system. Then, a priority-based data forwarding strategy is proposed to maximize the bandwidth efficiency, and we develop a Real-time Deep Q-learning based Edge Forwarding Scheme Optimization Algorithm (RDFO) to maximize the overall task processing rate. Finally, we conduct the platform simulation for the distributed co-inference system. Through the comparison with other benchmarks, we testify to the optimality of our proposal. [ABSTRACT FROM AUTHOR]

Neural networks (NNs) have been deeply studied due to their wide applicability. Since time delays are unavoidable in reality, it is basic and crucial for all applications based on NNs to guarantee system stability under the influence of mixed time delays. To better exploit the variation information of time delay, we introduce the switching idea and approaches into mixed time-delay NNs to solve the stability problem. First, the considered mixed time-delay NNs are modeled as the switched NNs by dividing the two classes of time delays, discrete and distributed time delays, into some variable intervals and combining these intervals as new switching modes. With the help of mode-dependent average dwell-time switching, Lyapunov theory, and mathematical techniques, several exponential stability criteria on the modeled switched systems containing different modes are obtained. Moreover, via introducing the mathematical condition of the unstable subsystem in the switching system, a less conservativeness condition on the exponential stability of the modeled NNs is proposed. We perform three examples for testifying the validity of the proposed methods over existing ones. [ABSTRACT FROM AUTHOR]

Rail corrugation (RC) is one of the most important indicator to evaluate the quality of rail, which is used to describe the irregularity of rail surface, also known as rail’s vertical flatness. However, the definition of RC is still an empirical description for low-speed measurement devices. In this paper, the process of RC measurement is divided into two steps, sampling and extraction, which helps the users to understand more clearly. Then a new mathematic model is proposed to make the process of RC’s extraction be an executable operation for machine calculation. The model adopts a new concept of flexible virtual ruler to perform sliding filtering on an overall rail and extract the instantaneous RC with a successive approximation algorithm according to user’s requirements and national standards. The proposed FVR model can not only be fully compatible with the traditional extraction method of RC, but also provide a new idea for evaluating RC’s quality. Comparing with the current popular methods, the proposed model gives a meaningful strategy for rail maintenance with a complete mathematic description having more degrees of freedom. Experiment results demonstrate its validity and reliability for both indoor simulation and actual outdoor experiments. [ABSTRACT FROM AUTHOR]

One approach to achieving greener cellular networks is to power them with renewable energy. When there is insufficient renewable energy, base stations (BSs) with lower predicted traffic loads are dynamically switched to sleep mode as a means of reducing total BS energy consumption. Prediction is done using only the traffic logs of active BSs since traffic logs are not available for sleeping BSs. However, prediction accuracy is degraded if information for the entire network is not used. We propose a scheme built on software-defined-networking and edge computing technologies for maintaining the accuracy of traffic prediction by incorporating the most important traffic logs into the prediction even when using only traffic logs for active BSs. It works by estimating the contribution of the traffic logs for each BS to traffic prediction accuracy. Evaluation of the proposed scheme through extensive numerical experiments using a dataset of actual BS traffic logs demonstrates that the proposed scheme is superior to two benchmark schemes in terms of prediction accuracy and robustness against the reduction of active BSs for energy saving and different BS sets. [ABSTRACT FROM AUTHOR]

Computation offloading is a main issue in multi-access edge computing (MEC). In particular, in aerial MEC (AMEC) relying on UAVs/ drones, due to limited resources of single devices in providing services to end users, we always have to offload tasks to different devices within the network. However, in wireless communications, there are obstacles in transmission distance between devices sending out offloading requests and others with idle resources. Furthermore, UAVs in AMEC also have to face the challenges of limited energy supply from batteries. As a result, in this article, we design a Monte Carlo tree search enhanced hybrid strategy in achieving time- and energy-aware computation offloading for AMEC. The simulation results show that the proposed solutions can greatly reduce time cost and energy consumption while using UAVs as flying offloaders. [ABSTRACT FROM AUTHOR]

The combination of wireless energy transfer (WET) and mobile edge computing (MEC) has been proposed to satisfy the energy supply and computation requirements of resource-constrained Internet of Things (IoT) devices. However, the energy transfer efficiency and task offloading rate cannot be guaranteed when wireless links between the hybrid access point (HAP) and IoT devices are hostile. To address this problem, this paper aims at utilizing the intelligent reflecting surfaces (IRS) technique to improve the efficiency of WET and task offloading. In particular, we investigate the total computation bits maximization problem for IRS-enhanced wireless powered MEC networks, by jointly optimizing the downlink/uplink phase beamforming of IRS, transmission power and time slot assignment used for WET and task offloading, and local computing frequencies of IoT devices. Furthermore, an iterative algorithm is presented to solve the non-convex non-linear optimization problem, while the optimal transmission power and time allocation, uplink phase beamforming matrixes and local computing frequencies are derived in closed-form expressions. Finally, extensive simulation results validate that our proposed IRS-enhanced wireless powered MEC strategy can achieve higher total computation rate as compared to existing baseline schemes. [ABSTRACT FROM AUTHOR]

Recently, artificial intelligence (AI) is undergoing a sustained success renaissance as it can substantially improve networks' cognitive performance and intelligence, thereby contributing to fully unleashing the potential of big data. Pushing the AI frontiers to the network edge in this context and trends has given rise to an emerging interdiscipline, namely, edge intelligence (EI). Indeed, EI can sink the cloud's processing capabilities to the edge side, and provide real-time response while enabling more intelligent services with high performance. However, the successful realization of EI is still in its infancy. Thus, this article aims to provide a comprehensive study of this young field from a broader perspective. We first discuss the prior knowledge based on which we take a holistic overview of EI, including its key concepts, advantages, and development trend. Then we highlight the collaboration modes in EI, and discuss two typical case categories. Subsequently, the entire processes of model training and inference in EI are elaborated. Finally, we discuss a typical application scenario and its specific embodiment of EI and strive to shed light on some potential challenges, which may facilitate the transformation of EI from theory to practice. [ABSTRACT FROM AUTHOR]

Through offloading tasks to surrounding edge nodes, the edge-assistant vehicular network (EAVN) could provide faster and more efficient services. To promote the deployment of EAVN, we need a reasonable resource allocation scheme. Nowadays, there are few auction mechanisms in EAVN scenario. Moreover, the auction mechanism in other edge computing scenario has following limitations. Firstly, current works only focus on the static/offline auction and do not consider that users will join and leave the auction system at any time in EAVN. Secondly, current works that determine the winning buyers/sellers only based on the price do not consider the unique attributes of EAVN, such as poor communication quality and various task demands. We design an online auction scheme for EAVN, which could satisfy the dynamics of users in the auction system. Moreover, the proposed auction mechanism also considers the non-price attributes, such as location, reputation, and computing ability, when constructing the matching between buyers and sellers. To verify our work, we simulate EAVN using a vehicular network simulator. Experimental results show that the proposed mechanism could meet the properties of computational efficiency, individual rationality, budget balance, and truthfulness. [ABSTRACT FROM AUTHOR]

Media access control (MAC) plays a pivotal role in ensuring proper operation in wireless body area networks (WBANs). However, current solutions still cannot satisfy multiple performance requirements of such networks. In this article, we propose a priority ladders-based resource scheduling (PLRS) scheme to meet these demands. First, we design a multiple-level priority division approach, which is based on health parameters and data types of patients. Unlike only two priorities of nodes in other schemes, there are four priorities in our solution. This is more practical and more fair for different nodes. Moreover, we devise a ladders-based time-slot allocation method, in which time slots of nodes are increased round by round based on the priority ladders of nodes. This reduces the difference of the number of time slots between disparate nodes with close priorities, and accordingly enhances the fairness of nodes. In addition, we design an interval-based task ordering approach, in which nodes use the allocated time slots in dispersed multiple intervals rather than a continuous time span in conventional solutions. This guarantees high throughput of high-priority nodes and short waiting time of emergency data. Extensive simulations exhibit the advantages of PLRS in terms of energy overhead, delay, and utility. [ABSTRACT FROM AUTHOR]

Nowadays, wilderness monitoring provides massive data output for supporting agricultural production, environmental protection, and disaster monitoring. However, smart upgrading alone for these wireless nodes cannot meet the softwarized network needs today, relating to the explosion of multi-dimensional data and multi-species equipment. In this article, we conduct a comprehensive solution for the UAV based data collection strategy in an “air-to-ground” intelligent softwarized collection system. The innovation in this article is that after using the IoT nodes to complete the data collection process through the proposed bandwidth-weighted traffic pushing optimization (BWPTO) algorithm, the system infers the future changes according to the current network state using a deep Q-learning (DQL) network. Then, by developing the proposed AIIPO (Air-to-Ground Intelligent Information Pushing Optimization) algorithm, the entire network can “forward-looking” the uploaded information to potentially idle nodes in the future, thus achieve the optimized system performance. Through the final mathematical experiments, we prove the optimality of our proposed routing algorithm and forwarding strategy, which are more applicable in the dynamic “air-to-ground” distributed data collection system than other benchmark solutions. [ABSTRACT FROM AUTHOR]

Interest flooding attacks (IFAs) are widely regarded as being among the most harmful security risks in named data networking (NDN). Through an IFA, the attacker injects numerous Interest packets into a network to drain network resources such as bandwidth, caching capacity, and computational capacity, which can seriously affect the normal data content requests of legitimate consumers and degrade the network quality of service (QoS). To design a high-efficiency IFA mitigation scheme, it is critical to detect attacks accurately and rapidly. Therefore, there is high interest in developing an optimized attack detection scheme. In this study, the concept of an isolation forest (iForest) is introduced to develop an IFA detection mechanism in which the iForest construction process isolates abnormal and legitimate prefixes. This approach enables malicious prefixes to be identified among abnormal prefixes to mitigate IFAs by restricting the forwarding of malicious Interest packets. The results of extensive simulations show that the proposed iForest-based IFA detection mechanism (IFDM) outperforms other related schemes in terms of attack detection accuracy and speed and thus can offer effective support for preserving NDN QoS. [ABSTRACT FROM AUTHOR]

When natural disasters such as earthquakes happen, network architectures are highly vulnerable. Once the network connections cannot be guaranteed, even rescue operations are difficult to carry out. in this article, to help the people in affected areas retrieve network services, we come up with a solution based on fog computing to quickly deploy emergent distributed service. in case the original network infrastructures are damaged to varying degrees, we make use of the available equipment including routers and mobile devices as fog nodes to realize emergency networking and communication. in the simulation evaluation, we compare our method with existed one. The results show that our method can reduce time costs on fast deployment and improve work efficiency. [ABSTRACT FROM AUTHOR]

Powered by the technologies that have originated from manufacturing, the fourth revolution of healthcare technologies is happening (Healthcare 4.0). As an example of such revolution, new generation homecare robotic systems (HRS) based on the cyber-physical systems (CPS) with higher speed and more intelligent execution are emerging. In this article, the new visions and features of the CPS-based HRS are proposed. The latest progress in related enabling technologies is reviewed, including artificial intelligence, sensing fundamentals, materials and machines, cloud computing and communication, as well as motion capture and mapping. Finally, the future perspectives of the CPS-based HRS and the technical challenges faced in each technical area are discussed. [ABSTRACT FROM AUTHOR]

Modern cloud computing platforms based on virtual machine monitors (VMMs) host a variety of complex businesses which present many network security vulnerabilities. In order to protect network security for these businesses in cloud computing, nowadays, a number of middleboxes are deployed at front-end of cloud computing or parts of middleboxes are deployed in cloud computing. However, the former is leading to high cost and management complexity, and also lacking of network security protection between virtual machines while the latter does not effectively prevent network attacks from external traffic. To address the above-mentioned challenges, we introduce a novel customized network security for cloud service (CNS), which not only prevents attacks from external and internal traffic to ensure network security of services in cloud computing, but also affords customized network security service for cloud users. CNS is implemented by modifying the Xen hypervisor and proved by various experiments which showing the proposed solution can be directly applied to the extensive practical promotion in cloud computing. [ABSTRACT FROM AUTHOR]

Cloud computing offers various services based on outsourced data by utilizing its huge volume of resources and great computation capability. However, it also makes users lose full control over their data. To avoid the leakage of user data privacy, encrypted data are preferred to be uploaded and stored in the cloud, which unfortunately complicates data analysis and access control. In particular, few existing works consider the fine-grained access control over the computational results from ciphertexts. Though our previous work proposed a framework to support several basic computations (such as addition, multiplication and comparison) with flexible access control, privacy-preserving division calculations over encrypted data, as a crucial operation in many statistical processes and machine learning algorithms, is neglected. In this paper, we propose four privacy-preserving division computation schemes with flexible access control to fill this gap, which can adapt to various application scenarios. Furthermore, we extend a division scheme over encrypted integers to support privacy-preserving division over multiple data types including fixed-point numbers and fractional numbers. Finally, we give their security proof and show their efficiency and superiority through comprehensive simulations and comparisons with existing work. [ABSTRACT FROM AUTHOR]

Individually reinforcing the robustness of a single deep learning model only gives limited security guarantees especially when facing adversarial examples. In this article, we propose DeSVig, a decentralized swift vigilance framework to identify adversarial attacks in an industrial artificial intelligence systems (IAISs), which enables IAISs to correct the mistake in a few seconds. The DeSVig is highly decentralized, which improves the effectiveness of recognizing abnormal inputs. We try to overcome the challenges on ultralow latency caused by dynamics in industries using peculiarly designated mobile edge computing and generative adversarial networks. The most important advantage of our work is that it can significantly reduce the failure risks of being deceived by adversarial examples, which is critical for safety-prioritized and delay-sensitive environments. In our experiments, adversarial examples of industrial electronic components are generated by several classical attacking models. Experimental results demonstrate that the DeSVig is more robust, efficient, and scalable than some state-of-art defenses. [ABSTRACT FROM AUTHOR]

With the exponential expansion of the number of Internet of Things (IoT) devices, many state-of-the-art communication technologies are being developed to use the lowerpower but extensively deployed devices. Due to the limits of pure channel characteristics, most protocols cannot allow an IoT network to be simultaneously large-scale and energy-efficient, especially in hybrid architectures. However, different from the original intention to pursue faster and broader connectivity, the daily operation of IoT devices only requires stable and low-cost links. Thus, our design goal is to develop a comprehensive solution for intelligent green IoT networking to satisfy the modern requirements through a data-driven mechanism, so that the IoT networks use computational intelligence to realize self-regulation of composition, size minimization, and throughput optimization. To the best of our knowledge, this study is the first to use the green protocols of LoRa and ZigBee to establish an ad hoc network and solve the problem of energy efficiency. First, we propose a unique initialization mechanism that automatically schedules node clustering and throughput optimization. Then, each device executes a procedure to manage its own energy consumption to optimize switching in and out of sleep mode, which relies on AI-controlled service usage habit prediction to learn the future usage trend. Finally, our new theory is corroborated through real-world deployment and numerical comparisons. We believe that our new type of network organization and control system could improve the performance of all green-oriented IoT services and even change human lifestyle habits. [ABSTRACT FROM AUTHOR]

Smart grid has emerged as a successful application of cyber-physical systems in the energy sector. Among numerous key technologies of the smart grid, vehicle-to-grid (V2G) provides a promising solution to reduce the level of demand–supply mismatch by leveraging the bidirectional energy-trading capabilities of electric vehicles. In this paper, we propose a secure and efficient V2G energy trading framework by exploring blockchain, contract theory, and edge computing. First, we develop a consortium blockchain-based secure energy trading mechanism for V2G. Then, we consider the information asymmetry scenario, and propose an efficient incentive mechanism based on contract theory. The social welfare optimization problem falls into the category of difference of convex programming and is solved by using the iterative convex–concave procedure algorithm. Next, edge computing has been incorporated to improve the successful probability of block creation. The computational resource allocation problem is modeled as a two-stage: 1) Stackelberg leader–follower game and 2) the optimal strategies are obtained by using the backward induction approach. Finally, the performance of the proposed framework is validated via numerical results and theoretical analysis. [ABSTRACT FROM AUTHOR]

Currently, IoT has become an important carrier of blockchains, which not only makes blockchain more ubiquitous but also improves the security of IoT. Consensus is the core component of blockchains with various forms, which raises the following challenges. Dynamic management and configuration of the consensuses in a blockchain are required because IoT applications have high dynamics. Moreover, an IoT node is usually reutilized by various applications in different blockchains, which means the IoT node should be switched frequently to cross consensuses in different blockchains. To address this, a software-defined blockchain architecture is proposed to realized the dynamic configurations for blockchains. Then a consensus function virtualization approach with application-aware work flow is proposed, which can abstract and manage various consensus resources. Next, a transfer-learning-based intelligent scheme is designed to implement the application- layer packet analysis and perform the efficient management of virtualized consensus resources. Experiment results indicate the feasibility of the proposed scheme. This work is significant in enhancing the flexibility and extendibility of blockchains in IoT. [ABSTRACT FROM AUTHOR]

Graph-based approaches for fake account detection is one of the important means to fight against fake accounts' attacks on social networks. With the growth of the scale of social networks, more and more researchers begin to use the graph computing framework to boost their detection algorithms. We make detailed analyses of social networks' graph data and state-of-the-art graph computing frameworks, and find that some techniques of the current graph computing systems are overgeneralized and suboptimal, which means they only focus on how to design a graph processing framework on general graphs but miss the optimization of social networks graphs. So, in this paper we propose QuickSquad, a graph computing system on a single server which is specific to the optimization of social networks graph structures. QuickSquad uses the method of "divide and rule" instead of overgeneralization. First, we divide the graph structure data into the heavy set and the light set according to the out-degree of vertices. Then, we 1) store them with different formats, 2) process them with edge-based updating and vertex-based updating appropriately in a two-phase processing model, 3) apply two selective scheduler strategies of different level, i.e. vertex-level and file-level, and 4) provide four cache priorities when the memory is not enough to cache all data. Finally, we implement two detection methods, dSybilRank and dCOLOR, on our system, and the experiments demonstrate that our system can increase the performance up to 5.91X (from 1.14X) compared with the performance of the current graph computing systems, like GridGraph. [ABSTRACT FROM AUTHOR]

Internet of Things (IoT) becomes a very important area for providing various services on connected smart devices. For the isolation of different services in IoT, software-defined networking (SDN)-based virtual networks will be a scalable and flexible solution. However, in a large-scale IoT, as smart devices will move long distance between different positions, virtual network management becomes very difficult in providing network services. In this paper, we propose the LS-SDV, an efficient virtual network management framework in large-scale software-defined IoT (SDIoT). In this framework, we design a two-layer distributed control plane to manage devices and virtual networks in a large-scale environment. To the best of our knowledge, the LS-SDV is the first work to apply distributed control plane for virtual network management in softwarized networks. Moreover, based on the novel structure of the LS-SDV, we also provide a solution for network flow scheduling through network analysis. We evaluate the performance of our framework and virtual network management by extensive simulation and experiment in an open SDN framework. [ABSTRACT FROM AUTHOR]

Information-centric networking (ICN) is an important trend that will impact the future of the Internet. ICN caters to large content consumption patterns while achieving high performance. New features in the information-centric future Internet, such as caching, name-based routing, and content-based security, bring novel challenges to a decentralized environment. On one hand, the processing capabilities on the edge in an information-centric future Internet need to implement smart analysis for large quantities of content. On the other hand, the computational and storage resources need to be configured and controlled on demand and based on cognition of the content from users. To address these challenges, this article proposes a fog-computing- enabled cognitive network functions virtualization approach for an information-centric future Internet. We first propose an on-demand caching function virtualization scheme and design a communication scheme between the fog nodes and the future Internet nodes for the forwarding process. Then, to attain smart control for related operations (i.e., routing, cache policy, and security), we propose a control function virtualization approach. Finally, a cognitive resource configuration mechanism is proposed. The simulation results show the advantages and efficiency of the proposed approach. [ABSTRACT FROM AUTHOR]

Modern data centers usually use virtual machine technology to host various big data applications in a single physical machine, not only enhancing the server utilization, but also providing them with the hardware-level isolation. However, in a typical virtualized environment an extra software layer called virtual machine monitor (VMM) is often interposed between the hardware resource and guest operating system (virtual machine, VM), shielding the specific user-process semantic inside a running VM. As a result, it obstructs the disk I/O scheduler of VMM to acquire the accurate information of a user-process (often a big data application), and thus proposes a challenge on the I/O request scheduling as well as the disk resource management at the granularity of VM user-process. Eventually, the disk I/O performance of a virtualized system is sub-optimal. This paper introduces an improved disk I/O scheduling framework for the virtualized system. It aims at bridging the semantic gap between physical disk I/O scheduler and VM user-process, providing a fair sharing of disk I/O resource among concurrent VMs. At the same time, it improves the overall disk I/O performance through a novel method for creating the image file of VM. Besides, an extra scheduling algorithm is proposed to further refine the storage performance. Finally, we implement these improvements on Xen hypervisor and conduct extensive experiments to verify our framework. The experimental result shows that our work improve the performance of read-intensive, write-intensive and mixed workloads up to 9, 10.7 and 20% respectively. [ABSTRACT FROM AUTHOR]

Traditional Network Functions Virtualization (NFV) implementations are somehow too heavy and do not have enough functionality to conduct complex tasks. In this work, we propose a lightweight NFV framework named DeepNFV, which is based on the Docker container running on the network edge, and integrates state-of-the-art deep learning models with NFV containers to address some complicated problems, such as traffic classification, link analysis, and so on. We compare the DeepNFV framework with several existing works, and detail its structures and functions. The most significant advantage of DeepNFV is its lightweight design, resulting from the virtualization and low-cost nature of the container technology. Also, we design this framework to be compatible with edge devices, in order to decrease the computational overhead of the central servers. Another merit is its strong analysis ability brought by deep learning models, which make it suitable for many more scenarios than traditional NFV approaches. In addition, we also describe some typical application scenarios, regarding how the NFV container works and how to utilize its learning ability. Simulations demonstrate its high efficiency, as well as the outstanding recognition performance in a typical use case. [ABSTRACT FROM AUTHOR]

Network function virtualization (NFV) is an emerging scheme to provide virtualized network function services for next-generation networks. However, finding an efficient way to distribute different resources to customers is difficult. In this paper, we develop a new double-auction approach named DARA that is used for both service function chain routing and NFV price adjustment to maximize the profits of all participants. To the best of our knowledge, this is the first work to adopt a double-auction strategy in this area. The objective of the proposed approach is to maximize the profits of three types of participants: 1) NFV broker; 2) customers; and 3) service providers. Moreover, we prove that the approach is a weakly dominant strategy in a given NFV market by finding the Bayesian Nash equilibrium in the double-auction game. Finally, according to the results of the performance evaluation, our approach outperforms the single-auction mechanism with higher profits for the three types of participants in the given NFV market. [ABSTRACT FROM AUTHOR]

As one of the promising information-centric networking (ICN) architectures, NDN has attracted tremendous research attention and effort in the past decade. In particular, security and privacy remain significant concerns and challenges due to the fact that most of the traditional cryptographic primitives are no longer suitable for NDN architecture. For example, the traditional cryptographic primitives aim to secure point-to-point communications, always requiring explicit descriptions of to where or whom the data packets are intended, while network addressing or locating in NDN becomes implicit. To deal with such issues, the recently developed cryptographic primitives such as ABE have been applied to NDN. Also, to efficiently solve the trust-roots problem and seamlessly deploy cryptographic infrastructures, the concept of SDN has been introduced to NDN as well. This tutorial is devoted to exploring the interesting integration between NDN, ABE, and SDN. [ABSTRACT FROM AUTHOR]

Human activity recognition (HAR) is widely applied to many industrial applications. In the context of Industry 4.0, driven by the same demand of machines’ self-organizing ability, HAR can also be adopted in elderly healthcare. However, HAR should be adaptive to the application scenarios in elderly healthcare. In this paper, we propose a nonintrusive activity recognition method that can be applied to long-term and unobtrusive monitoring for elderlies. The method is robust to obstruction and nontarget object interference. Skeleton sequence is estimated from RGB images. Based on two activity continuity metrics, an interframe matching algorithm is proposed to filter nontarget objects. In order to make full use of spatial-temporal information, we propose a novel activity encoding method based on the interframe joints distances. A convolutional neural network is used to learn the distinguishing features automatically. A specific data augmentation method is designed to avoid the overfitting problem on small-scale datasets. The experiments are performed on two public activity datasets and a newly released noisy activity dataset (NAD). The NAD contains obstruction, nontarget object interference. The experimental results show that the proposed method achieves the state-of-the-art performance while only using one ordinary camera. The proposed method is robust to a realistic environment. [ABSTRACT FROM AUTHOR]

Modern data center consists of thousands of servers, racks and switches. Complicated structure means it requires well-designed algorithms to utilize resources of data centers efficiently. Current virtual machine scheduling algorithms mainly focus on the initial allocation of virtual machines based on the CPU, memory and network bandwidth requirements. However, when tasks finished or lease expired, related virtual machines would be deleted from the system which would generate resource fragments. Such fragments lead to unbalanced resource utilization and decline of communication performance. This paper investigates the network influence on typical applications in data centers and proposed a self-adaptive network-aware virtual machine clustering and consolidation algorithm to maintain an optimal system-wide status. Our consolidation algorithm periodically checks whether consolidation is necessary and then clusters and consolidates virtual machines to lower communication cost with an online heuristic. We used two benchmarks in a real environment to examine network influence on different tasks. To evaluate the advantages of the proposed algorithm, we also built a cloud computing testbed. Real workload trace-driven simulations and testbed-based experiments showed that, our algorithm greatly shortened the average finish time of map-reduce tasks and reduced time delay of web applications. Simulation results showed that our algorithm considerably reduced the amount of high-delay jobs, lowered the average traffic passed through aggregate switches and improved the communication ability among virtual machines. [ABSTRACT FROM AUTHOR]

With the rapid development of Internet of things devices and network infrastructure, there have been a lot of sensors adopted in the industrial productions, resulting in a large size of data. One of the most popular examples is the manufacture inspection, which is to detect the defects of the products. In order to implement a robust inspection system with higher accuracy, we propose a deep learning based classification model in this paper, which can find the possible defective products. As there may be many assembly lines in one factory, one huge problem in this scenario is how to process such big data in real time. Therefore, we design our system with the concept of fog computing. By offloading the computation burden from the central server to the fog nodes, the system obtains the ability to deal with extremely large data. There are two obvious advantages in our system. The first one is that we adapt the convolutional neural network model to the fog computing environment, which significantly improves its computing efficiency. The other one is that we work out an inspection model, which can simultaneously indicate the defect type and its degree. The experiments well prove that the proposed method is robust and efficient. [ABSTRACT FROM AUTHOR]