Search

As today's Internet of Things (IoT) applications are becoming more complicated and intelligent, IoT devices alone can no longer well support the ever-increasing demand for powerful computation and high energy efficiency. Mobile Edge Computing (MEC) and 5G technology emerge as promising solutions, which enable IoT tasks to be offloaded to edge servers for effective processing. Though desirable, there however exists a mismatch between the massive IoT task workloads and limited wireless bandwidth, making it challenging to achieve an optimal offloading strategy at the mobile edge, e.g., the base station (BS) server. In this paper, we aim to migrate the most suitable offloading tasks to fully obtain the benefits of the MEC task offloading. We first formulate the task offloading model as an optimization problem, and theoretically prove the NP-hardness in achieving the optimal solution. Thus, a Genetic algorithm, named M-COGA, is proposed to solve the task offloading selection in both single and multiple BS scenarios. The algorithm focuses on offloading as many tasks as possible with the maximum cost offloading. The proposed cost function takes into account both the computation overhead and energy consumption. Besides, for the multi-BS coverage scenario, we also consider the approach flexibility as well as link load balance. And an enhanced dynamic task offloading scenario is further discussed. We verify the efficiency of our algorithm under the condition of both uniform and non-uniform distribution of covered nodes. Numerical experiments demonstrate that our dynamic allocating scheme can effectively work in MEC offloading. Besides, it largely outperforms the single BS scenarios and reduces the cost of edge devices. [ABSTRACT FROM AUTHOR]

Reviews of product or stores exist extensively in online e-commerce platform which is important for customers to make decisions. For economic reasons some dishonest people are employed to write fake reviews which is also called "opinion spamming" to promote or demote target products and services. Previous researches have made use of text similarity, linguistics, rating patterns, graph relationship and other behaviors for spammer detection. They mainly utilized product review list while it is difficult to find fake reviews by glancing over product reviews in time-descending order. Meanwhile there exists lots of useful information in the list of reviews of reviewers and relationships between reviewers when reviewers commonly reviewed the same stores. We propose the concept of review group and to the best of our knowledge, it's the first time the review group concept is proposed and used. Review grouping algorithm is designed to effectively split reviews of reviewer into groups which participate in building novel grouping models to identify both positive and negative deceptive reviews. Several new features which are language independent based on group model are constructed. Additionally, we explore the collusion relationship between reviewers to build reviewer group collusion model. Evaluations show that the review group method and reviewer group collusion models can effectively improve the precision by 4%–7% compared to the baselines in fake reviews classification task especially when reviews are posted by professional review spammers. [ABSTRACT FROM AUTHOR]

The widespread adoption of video surveillance systems in public security and network security domains has underscored the importance of video anomaly detection as a pivotal research area. To enhance the precision and robustness of anomaly detection, this manuscript introduces an innovative method for video anomaly detection. The approach begins with the application of multi-scale feature extraction technology to capture visual information across varying scales in video data. Leveraging the Spatial Pyramid Convolution (SPC) module as the cornerstone for multi-scale feature learning, the study addresses the impact of scale variations, thereby augmenting the model’s detection capabilities across different scales. Furthermore, a Weakly Supervised Data Augmentation Network (WSDAN) module is incorporated to facilitate attention-guided data augmentation, enhancing the richness of input images. These augmented images undergo training with the U-Net network to elevate detection accuracy. Additionally, the integration of the improved Convolutional Block Attention Module (CBAM) into the base U-Net architecture enables end-to-end training. CBAM dynamically adjusts feature map weights, allowing the model to concentrate on anomaly relevant regions in the video while suppressing interference from non-anomalous areas. To assess anomalies, the paper employs the Peak Signal-to-Noise Ratio (PSNR) between predicted and original frames, normalizing PSNR values for anomaly identification. The proposed method is then evaluated using publicly available datasets CUHK Avenue and UCSD Ped2, with results visually presented. Experimental findings showcase Area Under the Receiver-Operating Characteristic Curve (AUC) values of 86.2% and 97.9% for these datasets, surpassing comparative methods and confirming the effectiveness and superiority of the proposed approach. [ABSTRACT FROM AUTHOR]

Four panels with a semicircular arch structure comprised of glass-fiber-reinforced polymer (GFRP) laminates are subjected to low-velocity impact, compression-after-impact (CAI), and quasi-static compression tests. Mechanical response and the failure mechanism of the complex structure are investigated with the assistance of a three-dimensional digital image correlation (3D-DIC) system and numerical simulation based on Hashin’s failure criteria in Abaqus software. The results show that the addition of a semicircular arch can absorb more incident energy during the impact process. The failure modes of the panels are depended on the semicircular arch dimension, and excellent anti-impact performance is found in the semicircular arch with a diameter close to the impactor. Meanwhile, the CAI strength is increased with the semicircular arch size. [ABSTRACT FROM AUTHOR]