Concurrency and Computation-Practice & Experience最新文献

This paper introduces an enhanced BERT-DPCNN model for the task of Chinese news text classification. The model addresses the common challenge of balancing accuracy and computational efficiency in existing models, especially when dealing with large-scale, high-dimensional text data. To tackle this issue, the paper proposes an improved BERT-DPCNN model that integrates BERT's pre-trained language model with DPCNN's efficient convolutional structure to capture deep semantic information and key features from the text. Additionally, the paper incorporates the zebra optimization algorithm (ZOA) to dynamically optimize the model's hyperparameters, overcoming the limitations of manual tuning in traditional models. By automatically optimizing hyperparameters such as batch size, learning rate, and the number of filters through ZOA, the model's classification performance is significantly enhanced. Experimental results demonstrate that the improved ZOA-BERT-DPCNN model outperforms traditional methods on the THUCNEWS Chinese news dataset, not only verifying its effectiveness in news text classification tasks but also showcasing its potential to enhance classification performance.

With the increasingly widespread application of multiprocessor systems, some processors in multiprocessor systems are inevitably prone to malfunctions. The reliability and effectiveness of the system are key issues. As a standard for measuring system fault tolerance, connectivity, and edge connectivity have many drawbacks. Therefore, Haray proposed conditional connectivity by restricting the connected components in disconnected subgraphs $��G�-�F�$ to satisfy certain properties, where $��G�$ and $��F�$ represent the interconnection network and its set of faulty vertices, respectively. Restricted connectivity is a special type of conditional connectivity. Exchanged crossed cube, as a deformation of hypercube, has more favorable properties, such as smaller diameter, smaller link size, and lower cost. We prove that the 2-restricted connectivity of the exchanged crossed cubes $��\text{ECQ}�(�s�,�t�)�$ is $��4�s�-�4�$ for $��2�\le �s�\le �t�$.

Internet of Drones (IoD) is one of the most beneficial and has many versatile applications like Surveillance and Security, Delivery and Logistics, Environmental Monitoring, Agriculture, and so forth. The IoD network is crucial for collecting sensitive data like geo-coordinates, vehicle traffic data, and property details while surveying the various deployment locations in smart cities. The communication between users and drones can be compromised over insecure wireless channels by multiple attacks such as Man-in-the-middle-attack, Denial of Service, and so forth. Many schemes have already been propounded in the field of IoD. Still, many of them cannot address the resource constraints problem of drones, and existing protocols have higher computation and communication costs. Therefore, this paper has proposed a robust and efficient Hyper-Elliptic Curve-based authentication scheme (REHAS), which provides a session key for secure communication. Artificial Identities are generated using a hash function and random numbers. Fuzzy Extractor is used for user biometric authentication, which makes the smart device secure when lost. HECC is used with a smaller bit size of 80 bits rather than ECC of 160 bits. The security of the REHAS has been ensured using Scyther simulation. Furthermore, the resilience, safety, and robustness of REHAS are ensured by Informal security analysis. Lastly, a comparative study of the REHAS has been performed with other related Authentication and key agreement (AKA) protocols regarding communication cost, Computation cost, and security features, demonstrating that REHAS incurred less computation cost (6.7171 ms), communication overhead (1696 bits), and energy consumption (22.5 mJ) than other existing AKA schemes.

5G, 6G, and beyond networks promise to support vertical industrial services with strict QoS parameters, but the hardware-based "one-size-fits-all" model of legacy networks lacks the flexibility needed for diverse services. The foundation of 5G networks lies in softwarization, with network slicing, Software Defined Networking (SDN), and Network Function Virtualisation (NFV) serving as its core components. The network-slicing-based shared network environment necessitates an intelligent and flexible resource management approach. In this case, traditional approaches are no longer suitable for dealing with a dynamic network environment. With recent advancements, AI-based approaches have the potential to manage resources autonomously. This paradigm shift underscores the need for deep and extensive investigation. However, existing literature on this subject is fragmented and lacks a cohesive overview of network slicing. To address these gaps, our review paper aims to provide a comprehensive scope of network slicing in a unified manner. In this sequence at first, this paper presented a conceptual overview of network slicing and enabling technologies, including SDN, NFV, and edge computing. Secondly, this paper identifies the relevant phases of resource management and presents AI-based resource management for network traffic classification, admission, allocation, and scheduling. Finally, it also discusses the deployment of network slicing-enabled key use cases and their practical deployment, the research gap, and open research challenges. To the best of our knowledge, this is the first attempt to critically analyze and present a consolidated review of the state of the art in network slicing resource management modules and network slicing-enabled key industrial use cases. This paper aims to guide researchers in developing innovative solutions and assist network players in the practical deployment of network slices for industrial applications.

Underwater Wireless Sensor Networks (UWSNs) play a pivotal role in various applications, ranging from environmental monitoring to disaster prevention, necessitating robust and efficient communication protocols tailored to the challenging underwater environment. This paper introduces the Quantum-Inspired Void-Based Source Distributed Opportunistic Routing Protocol (QIVSORP) to address the inherent limitations of classical routing protocols in UWSNs. Motivated by the unique challenges posed by underwater conditions, QIVSORP leverages principles from quantum mechanics to enhance routing efficiency. The protocol employs a source-distributed approach, utilizing quantum entanglement, superposition, and opportunistic routing strategies to enable adaptive and reliable data transmission in underwater scenarios. QIVSORP incorporates void-based forwarding, adaptive decision-making, and multipath routing to dynamically adjust routing decisions based on real-time network conditions. The protocol's source-informed decisions and opportunistic forwarding contribute to the adaptability and reliability of communication in dynamic underwater environments. QIVSORP achieves outstanding performance metrics: maintaining a Packet Delivery Ratio (PDR) of 98.9% with 50 nodes and 80% with 400 nodes, reducing end-to-end delays to 12 ms at 50 nodes, 15 ms at 100 nodes, and 52 ms at 600 nodes, and demonstrating energy efficiency ranging from 0.2 to 200 J per delivered packet across varying node densities. These results highlight the QIVSORP's capability to optimize communication in dynamic underwater environments effectively.

Elasticity is an essential treatment in Cloud environment employed in academic and industrial contexts. The main purpose of elasticity is to reduce the deployment cost while optimizing computing resources. Multiple studies were conducted to tackle classic applications using monolithic architecture deployed with virtual machines (VMs). However, with the spread of microservice pattern, recent studies have been investigating this new trend using containers. This paper classifies and discusses existing approaches dealing with cloud elasticity. It provides a novel taxonomy for elastic approaches while focusing on microservices-based solutions. We additionally specify the strength and the shortcomings of each class of works. As a conclusion, we report the challenges for microservices-based applications elasticity and provide requirements for future investigations.

With the development of new energy generation technology and power electronic technology, power electronic equipment occupies an increasing proportion of the power system, and its control flexibility makes the power system complicated, bringing problems such as low inertia, low damping, high harmonics, low reliability, and weak anti-interference ability. Especially the influence of converter on inertia of power system, even causing excessive frequency fluctuations, leading to the collapse of the power system. In view of the inertia problem of converters, this article starts with the principle of converters and analyzes the inertia transmission characteristics of different types of converters in power systems by three types of control strategies. Based on the influence of the inertia parameters of converters on system frequency and power, a converter inertia target function is established, and a neural network adaptive adjustment strategy for converter inertia coefficient is proposed to achieve self-adaptive optimization of converter output power and system frequency. The corresponding converter model is established, and the simulation circuit and control model are built by Simulink to verify the inertia transfer characteristics. The simulation results show the correctness of the relevant theories and provide theoretical support for the inertia design of high-proportion power electronic systems.

Voting has always been a crucial topic of public attention for democratic reasons. The ease of use and low cost are the result of e-voting being frequently used for such important decision outcomes. However, the tremendous authority and intervening data in current e-voting systems make it risky and difficult to achieve correct equity and clarity in e-voting. So, by combining e-voting with blockchain technology, these issues can be resolved while providing reorganization and intervention-resistant characteristics. A voter's improper manipulation, frequent voting, or non-party voting, may also undermine fairness. A verifier is therefore required to check the e-voting mechanism in order to ensure its effectiveness and control the process equality and fairness. In this paper, a Blockchain-based e-Voting Mechanism (BVM) is developed for providing the end to end security and fairness for transparent voting. This mechanism also provides a zero-knowledge proof (ZP) based verifier to inspect the voting procedure against voter's mis-operations and uses a novel Improved Master-key Administration (IMA) based public key cryptography to attack prevention. The utilization of blockchain technology ensures transparency, anonymity, confidentiality, authentication, tamper resistance, and a high level of data integrity, making it a promising choice for modernizing and enhancing the electoral process. Also, the performance of BVM has been compared with similar voting mechanisms and analyzed based on time complexity, security analysis, performance factors like delay and throughput, and anti-attack examination.

Segmenting breast tumors from dynamic contrast-enhanced magnetic resonance images is a critical step in the early detection and diagnosis of breast cancer. However, this task becomes significantly more challenging due to the diverse shapes and sizes of tumors, which make it difficult to establish a unified perception field for modeling them. Moreover, tumor regions are often subtle or imperceptible during early detection, exacerbating the issue of extreme class imbalance. This imbalance can lead to biased training and challenge accurately segmenting tumor regions from the predominant normal tissues. To address these issues, we propose a hierarchical region contrastive learning approach for breast tumor segmentation. Our approach introduces a novel hierarchical region contrastive learning loss function that addresses the class imbalance problem. This loss function encourages the model to create a clear separation between feature embeddings by maximizing the inter-class margin and minimizing the intra-class distance across different levels of the feature space. In addition, we design a novel Attention-based 3D Multi-scale Feature Fusion Residual Module to explore more granular multi-scale representations to improve the feature learning ability of tumors. Extensive experiments on two breast DCE-MRI datasets demonstrate that the proposed algorithm is more competitive against several state-of-the-art approaches under different segmentation metrics.