Current Issue : October-December Volume : 2023 Issue Number : 4 Articles : 5 Articles
Mobile edge computing (MEC) has produced incredible outcomes in the context of computationally intensive mobile applications by offloading computation to a neighboring server to limit the energy usage of user equipment (UE). However, choosing a pool of application components to offload in addition to the volume of data transfer along with the latency in communication is an intricate issue. In this article, we introduce a novel energy-efficient offloading scheme based on deep neural networks. The proposed scheme trains an intelligent decision-making model that picks a robust pool of application components. The selection is based on factors such as the remaining UE battery power, network conditions, the volume of data transfer, required energy by the application components, postponements in communication, and computational load. We have designed the cost function taking all the mentioned factors, get the cost for all conceivable combinations of component offloading decisions, pick the robust decisions over an extensive dataset, and train a deep neural network as a substitute for the exhaustive computations associated. Model outcomes illustrate that our proposed scheme is proficient in the context of accuracy, root mean square error (RMSE), mean absolute error (MAE), and energy usage of UE....
Aiming at the problems of high delay and vulnerable to network attack in the traditional microgrid centralized architecture, a collaborative microgrid security defense method in the edge-computing environment is proposed. First, we build the edgecomputing framework for microgrid, deploy the edge-computing server near the equipment terminal to improve the data processing efficiency, and deploy the blockchain in the edge server to ensure the reliability of the system. Then, the fully homomorphic encryption algorithm is used to design the smart contract, and the secure sharing of information is ensured through identity authentication, data encryption call, and so on. Finally, the credibility model is integrated into the election algorithm and is used to build a trusted edge cooperation mechanism to further improve the ability of the system to defend against network attacks. Based on the microgrid model, the experimental demonstration of the proposed method is carried out. The results show that when subjected to a network attack, the current fluctuation range is small and the defense success rate exceeds 95%, which is better than other methods and can better meet the requirements of practical application....
Deep changes are occurring in the components and forms of education as a result of the ongoing integration and development of emerging technologies like cloud computing, mobile computing, and artificial intelligence with teaching and learning, and the digital transformation of education is consistently being pushed to new heights. Simultaneously, China’s higher education has concurrently reached the stage of popularization. The digitalization of higher education is related to the development quality and value proposition of higher education and determines whether it can adapt to the needs of quality diversification, lifelong learning, training personalization, and governance modernization in the popularization stage. As a result, the current and future phases of China’s higher education reform call for accelerating the pace of higher education’s digital transformation and guiding the high-quality growth of higher education with digital innovation. The application potential of intelligent learning systems in higher education is becoming more and more clear in this context. In view of this, this work draws from previous research and experiences to build and implement an embedded voice teaching system based on cloud computing and a deep learning model to meet the development needs of the current digital transformation of higher education. On the one hand, the new system can well compensate for the flaws and shortcomings of the current teaching means in universities and realize the accompanying ubiquitous learning by relying on the powerful storage and computing capacity of the cloud computing platform. On the other hand, this study designs a set of voice recognition methods integrating HMM+ LSTM to enhance the embedded voice system’s recognition performance, ultimately allowing for the voice recognition feature to be implemented in the pedagogical system. When it comes to processing audio signals, the hybrid model makes use of both the HMM’s robust time processing capability and the deep neural network’s robust characterization capability and generalization performance. As a result, the voice recognition rate, anti-interference performance, and noise robustness can all be significantly improved. Finally, the embedded voice system is put through its paces in an experimental setting to gauge its performance and functionality. The results of the tests demonstrate that the created hybrid model has high recognition accuracy and good noise immunity, which will be utilized as a foundation for the design and development of the final system. Meanwhile, the new system’s functional modules have achieved the expected results with good stability and reliability. Trial results gathered through interviews and questionnaires demonstrate that the new system significantly enhances the intelligence and adaptability of college teaching methods and is conducive to promoting the improvement of college students’ cultural literacy and innovation ability....
Edge computing can reduce the transmission pressure of wireless networks in earthquakes by pushing computing functionalities to network edges and avoiding the data transmission to cloud servers. However, this also leads to the scattered storage of data content in each edge server, increasing the difficulty of content search. This paper investigates the seismic data query problem supported by edge computing. We first design a lookup mechanism based on bloom filter, which can quickly determine if there is the information that we need on a particular edge server. Then, the MEC-based data query problem is formulated as an optimization problem whose goal is to minimize the long-term average task delay with the constraints of computing capacity of edge servers. To reduce the complexity of problem, we further transform it as a Markov Decision Process by defining state space, action space and reward function. A novel DQN-based seismic data query algorithm is proposed to solve problem effectively. Extensive simulation-based testing shows that the proposed algorithm performances better when compared with two state-of-the-art solutions....
In a smart parking system, the license plate recognition service controls the car’s entry and exit and plays the core role in the parking lot system. When the Internet is interrupted, the parking lot’s business will also be interrupted. Hence, we proposed an Edge-Cloud-Dew architecture for the mobile industry in order to tackle this critical problem. The architecture has an innovative design, including LAN-level deployment, Platform-as-a-Dew Service (PaaDS), the dew version of license plate recognition, and the dew type of machine learning model training. Based on these designs, the architecture presents many benefits, such as: (1) reduced maintenance and deployment issues and increased dew service reliability and sustainability; (2) effective release of the network constraint on cloud computing and increase in the horizontal and vertical scalability of the system; (3) enhancement of dew computing to resolve the heavy computing process problem; and (4) proposal of a dew type of machine learning training mechanism without requiring periodic retraining, but with acceptable accuracy. Finally, business owners can reduce their burdens when introducing machine learning technology. Our research goal is to make parking systems smarter in edge computing through the integration of cloud and dew architecture technology....
Loading....