Knowledge graph (KG) embedding aims to study the embedding representation to retain the inherent structure of KGs. Graph neural networks (GNNs), as an effective graph representation technique, have ...shown impressive performance in learning graph embedding. However, KGs have an intrinsic property of heterogeneity, which contains various types of entities and relations. How to address complex graph data and aggregate multiple types of semantic information simultaneously is a critical issue. In this article, a novel heterogeneous GNNs framework based on attention mechanism is proposed. Specifically, the neighbor features of an entity are first aggregated under each relation-path. Then the importance of different relation-paths is learned through the relation features. Finally, each relation-path-based features with the learned weight values are aggregated to generate the embedding representation. Thus, the proposed method not only aggregates entity features from different semantic aspects but also allocates appropriate weights to them. This method can capture various types of semantic information and selectively aggregate informative features. The experiment results on three real-world KGs demonstrate superior performance when compared with several state-of-the-art methods.
Reconfigurable wireless network can flexibly provide efficient spectrum access service and keep stable operation in highly dynamic environment. In this paper, a primary-prioritized recurrent deep ...reinforcement learning algorithm for dynamic spectrum access based on cognitive radio (CR) technology is proposed. The spectrum Markov state is modeled to capture the evolution behavior to achieve the priority queuing of the primary users and the secondary users. According to the spectrum access strategies of the secondary users under different optimal criteria, we can obtain the best tradeoff benefits of spectrum access fairness and throughput. Furthermore, we proposed a learning-based algorithm for dynamic spectrum access, which allows the secondary users to modify their parameters to select the optimal access policy to maximize network throughput utilization. The Dueling Deep <inline-formula><tex-math notation="LaTeX">Q</tex-math></inline-formula>-Network (Dueling DQN) with prioritized experience replay combined with recurrent neural network is used to improve the convergence speed. Extensive experimental results demonstrate that the proposed RDRL scheme outperforms the existing Dueling DQN and DQN schemes in terms of convergence speed and channel throughput.
Providing the ubiquitous network accessibility is the key goal for 5G and Beyond 5G (B5G) networks. As the number of devices and data increase, ensuring the Quality of Service(QoS) by the existing ...network with a singular focus could be challenging. Meanwhile, it always is of the utmost importance to perform the computation-intensive or delay-sensitive tasks welland provide long-term services for the B5G networks. To tackle these challenges, we present an air-ground integrated network in B5G wireless communications, where an UAV is deployed as an aerial radio access platform to formulate system strategy intelligently, as well as provide task offloading and energy harvesting opportunities for terrestrial devices. To get more insight of it, we propose an intelligent charging-offloading scheme and formulate the joint multi-taskcharging-offloading scheduling as an optimization problem aiming to minimizing the system servicelatency of all devices by jointly optimizing the task offloading decisions, connection scheduling, charging and computation resources allocation of UAV. However, the formulated optimization problem is a Mixed-Integer Nonlinear Programming (MINLP) problem which is challenging to solve in general. Therefore, we decompose it into multiple convex sub-problems based on Block-Coordinate Descent (BCD) method and develop an improved greedy algorithm to obtain a feasible optimal solution. To validate the proposed algorithm, it is comprehensively compared with several existing schemes. Performance evaluation demonstrates that our scheme outperforms the benchmarks in terms of the system service latency of all UDs. Moreover, we represent the system working process and the paradigm of industrial applications.
UAV-assisted wireless communications facilitate the applications of Internet of Things (IoT), which employ billions of devices to sense and collect data with an on-demand style. However, there are ...numerous malicious Mobile Data Collectors (MDCs) mixing into the network, stealing or tampering with data, which greatly damages IoT applications. So, it is urgent to build a ubiquitous trust communication system. In this paper, a UAV-assisted Ubiquitous Trust Evaluation (UUTE) framework is proposed, which combines the UAV-assisted global trust evaluation and the historical interaction based local trust evaluation. We first propose a global trust evaluation model for data collection platforms. It can accurately eliminate malicious MDCs and create a clean data collection environment, by dispatching UAVs to collect baseline data to validate the data submitted by MDCs. After that, a local trust evaluation model is proposed to help select credible MDCs for collaborative data collection. By letting UAVs distribute the data verification hash codes to MDCs, the MDCs can verify whether the exchanged data from the interacted MDCs is reliable. Extensive experiments conduct on a real-life dataset demonstrate that our UUTE system outperforms the existing trust evaluation systems in terms of accuracy and cost.
The development of Industrial Internet of Things (IIoT) and Industry 4.0 has completely changed the traditional manufacturing industry. Intelligent IIoT technology usually involves a large number of ...intensive computing tasks. Resource-constrained IIoT devices often cannot meet the real-time requirements of these tasks. As a promising paradigm, the mobile-edge computing (MEC) system migrates the computation intensive tasks from resource-constrained IIoT devices to nearby MEC servers, thereby obtaining lower delay and energy consumption. However, considering the varying channel conditions as well as the distinct delay requirements for various computing tasks, it is challenging to coordinate the computing task offloading among multiple users. In this article, we propose an autonomous partial offloading system for delay-sensitive computation tasks in multiuser IIoT MEC systems. Our goal is to provide offloading services with minimum delay for better Quality of Service (QoS). Enlighten by the recent advancement of reinforcement learning (RL), we propose two RL-based offloading strategies to automatically optimize the delay performance. Specifically, we first implement the <inline-formula> <tex-math notation="LaTeX">Q </tex-math></inline-formula>-learning algorithm to provide a discrete partial offloading decision. Then, to further optimize the system performance with more flexible task offloading, the offloading decisions are given as continuous based on deep deterministic policy gradient (DDPG). The simulation results show that the <inline-formula> <tex-math notation="LaTeX">Q </tex-math></inline-formula>-learning scheme reduces the delay by 23%, and the DDPG scheme reduces the delay by 30%.
Recommendation accuracy is a fundamental problem in the quality of the recommendation system. In this article, we propose an efficient deep matrix factorization (EDMF) with review feature learning ...for the industrial recommender system. Two characteristics in user's review are revealed. First, interactivity between the user and the item, which can also be considered as the former's scoring behavior on the latter, is exploited in a review. Second, the review is only a partial description of the user's preferences for the item, which is revealed as the sparsity property. Specifically, in the first characteristic, EDMF extracts the interactive features of onefold review by convolutional neural networks with word-attention mechanism. Subsequently, <inline-formula><tex-math notation="LaTeX">{L}_{0}</tex-math></inline-formula> norm is leveraged to constrain the review considering that the review information is a sparse feature, which is the second characteristic. Furthermore, the loss function is constructed by maximum a posteriori estimation theory, where the interactivity and sparsity property are converted as two prior probability functions. Finally, the alternative minimization algorithm is introduced to optimize the loss functions. Experimental results on several datasets demonstrate that the proposed methods, which show good industrial conversion application prospects, outperform the state-of-the-art methods in terms of effectiveness and efficiency.
The Internet of things (IoT) is composed of billions of sensing devices that are subject to threats stemming from increasing reliance on communications technologies. A Trust-Based Secure Routing ...(TBSR) scheme using the traceback approach is proposed to improve the security of data routing and maximize the use of available energy in Energy-Harvesting Wireless Sensor Networks (EHWSNs). The main contributions of a TBSR are (a) the source nodes send data and notification to sinks through disjoint paths, separately; in such a mechanism, the data and notification can be verified independently to ensure their security. (b) Furthermore, the data and notification adopt a dynamic probability of marking and logging approach during the routing. Therefore, when attacked, the network will adopt the traceback approach to locate and clear malicious nodes to ensure security. The probability of marking is determined based on the level of battery remaining; when nodes harvest more energy, the probability of marking is higher, which can improve network security. Because if the probability of marking is higher, the number of marked nodes on the data packet routing path will be more, and the sink will be more likely to trace back the data packet routing path and find malicious nodes according to this notification. When data packets are routed again, they tend to bypass these malicious nodes, which make the success rate of routing higher and lead to improved network security. When the battery level is low, the probability of marking will be decreased, which is able to save energy. For logging, when the battery level is high, the network adopts a larger probability of marking and smaller probability of logging to transmit notification to the sink, which can reserve enough storage space to meet the storage demand for the period of the battery on low level; when the battery level is low, increasing the probability of logging can reduce energy consumption. After the level of battery remaining is high enough, nodes then send the notification which was logged before to the sink. Compared with past solutions, our results indicate that the performance of the TBSR scheme has been improved comprehensively; it can effectively increase the quantity of notification received by the sink by 20%, increase energy efficiency by 11%, reduce the maximum storage capacity needed by nodes by 33.3% and improve the success rate of routing by approximately 16.30%.
The development of a 5G-enabled Internet of Things has led to a dramatic increase in network traffic load, which has presented tremendous challenges to network management. In this paper, a ...service-oriented network architecture is proposed to support the effective management of 5G-enabled IoT systems. This architecture effectively reduces the traffic load and simplifies network management by introducing a service aggregation and caching (SAaC) scheme. Specifically, SAaC first breaks through the data-centric network architecture by converting data into services. Then, SAaC significantly reduces traffic load and energy consumption by service aggregation. Finally, SAaC introduces service caching, and each content router caches new services locally after aggregating received services so that user requests are handled at the network layer. Experimental results demonstrate that compared with traditional solutions, the SAaC scheme improves the request response time by 20.52%–56.09%, reduces the traffic load by 10.85%–37.67%, and reduces energy consumption by more than 50%.
Internet of Things (IoT) realizes the interconnection of heterogeneous devices by the technology of wireless and mobile communication. The data of target regions are collected by widely distributed ...sensing devices and transmitted to the processing center for aggregation and analysis as the basis of IoT. The quality of IoT services usually depends on the accuracy and integrity of data. However, due to the adverse environment or device defects, the collected data will be anomalous. Therefore, the effective method of anomaly detection is the crucial issue for guaranteeing service quality. Deep learning is one of the most concerned technology in recent years which realizes automatic feature extraction from raw data. In this article, the integrated model of the convolutional neural network (CNN) and recurrent autoencoder is proposed for anomaly detection. Simple combination of CNN and autoencoder cannot improve classification performance, especially, for time series. Therefore, we utilize the two-stage sliding window in data preprocessing to learn better representations. Based on the characteristics of the Yahoo Webscope S5 dataset, raw time series with anomalous points are extended to fixed-length sequences with normal or anomaly label via the first-stage sliding window. Then, each sequence is transformed into continuous time-dependent subsequences by another smaller sliding window. The preprocessing of the two-stage sliding window can be considered as low-level temporal feature extraction, and we empirically prove that the preprocessing of the two-stage sliding window will be useful for high-level feature extraction in the integrated model. After data preprocessing, spatial and temporal features are extracted in CNN and recurrent autoencoder for the classification in fully connected networks. Empiric results show that the proposed model has better performances on multiple classification metrics and achieves preferable effect on anomaly detection.
Billions of sensors and devices are connecting to the Internet of Thing (IoT) and generating massive data which are benefit for smart network systems. However, low-cost, secure, and efficient data ...collection from billions of IoT devices in smart city is a huge challenge. Recruiting mobile vehicles (MVs) has been proved to be an effective data collection scheme. However, the previous approaches rarely considered the security. In this paper, a novel Baseline Data based Verifiable Trust Evaluation (BD-VTE) scheme is proposed to guarantee security at a low cost. BD-VTE scheme includes Verifiable Trust Evaluation (VTE) mechanism, Effectiveness-based Incentive (EI) mechanism, and Secondary Path Planning (SPP) strategy, which are respectively used for reliable trust evaluation, reasonable reward, and efficient path adjustment. Among them, an active trust verification mechanism is innovatively proposed in the VTE mechanism, which evaluates the trust of MVs by sending UAVs to perceive IoT devices data as baseline data. This is a fundamental change to the previous passive and unverifiable trust mechanism. The simulation results show that BD-VTE scheme reduces the cost by at least 25.12% ∼ 38.03%, improves the collection rate by 0.91% ∼ 9.65% and increases the accuracy by 10.28% on average compared with the previous strategies.