Cluster Computing最新文献

Intrusion detection datasets are highly likely to contain numerous redundant, irrelevant, and noisy features that slow the performance of the machine learning techniques and classifiers that may be applied to them. The feature selection approach is used for reducing the number of features in intrusion detection datasets and eliminating those that are not important. One of the most powerful structured population approaches is the Cellular Automata approach, which is used to enhance the diversity and convergence of population-based optimization algorithms. In this work, the Cellular Automata approach, Mixed opposition-based learning, and the K-Nearest Neighbor classifier are incorporated with the Prairie dog optimization algorithm (PDO) in a new intrusion detection framework called Binary Opposition Cellular Prairie dog optimization algorithm (BOC-PDO). The proposed framework contains four key features. First, the Cellular Automata model is utilized to enhance the population of feasible solutions in the PDO. Second, four S-shaped and four V-shaped Binary Transfer Functions are used to convert the continuous solutions in BOC-PDO to binary ones. Third, the Mixed opposition-based learning approach is used at the end of the optimization loop of BOC-PDO to improve capacity for exploration. Fourth, the K-Nearest Neighbor classifier is used as the main learning model in BOC-PDO. Eleven famous intrusion detection datasets were employed in the evaluation of the effectiveness of BOC-PDO compared to eight popular binary optimization algorithms and four machine learning approaches. According to the overall simulation results, BOC-PDO scored the highest accuracy, best objective value, and fewest selected features for each of the eleven intrusion detection datasets. Besides, the reliability and consistency of the simulation results of BOC-PDO compared to the other tested algorithms were established using Friedman and Wilcoxon statistical tests.

Recently, the spotlight has been cast on smart healthcare by many researchers to provide better facilities to patients. Improved services, such as reducing health hazards, monitoring patient health, tracking disease trends, and enhancing service quality, can be offered by smart healthcare. Despite its numerous potential benefits, smart healthcare is associated with some security challenges. These challenges can be mitigated by utilizing blockchain technology, which is characterized by decentralization, cryptography, consensus mechanisms, transparency and accountability, smart contracts, ownership of data, immutability, and distributed ledger. Therefore, the latest blockchain technology is focused in this article to address the security challenges of smart healthcare. In this article, attention is given to smart healthcare, smart cities for smart healthcare, smart and secure healthcare, and cutting-edge technologies for smart cities and smart healthcare.Please provide author biography and photo.

Electroencephalograms (EEG)-based technology for recognizing emotions has attracted a lot of interest lately. However, there is still work to be done on the efficient fusion of different temporal and spatial features of EEG signals to improve performance in emotion recognition. Therefore, this study suggests a new deep learning architecture that combines a time–frequency convolutional neural network (TFCNN), a bidirectional gated recurrent unit (BiGRU), and a self-attention mechanism (SAM) to categorize emotions based on EEG signals and automatically extract features. The first step is to use the continuous wavelet transform (CWT), which responds more readily to temporal frequency variations within EEG recordings, as a layer inside the convolutional layers, to create 2D scalogram images from EEG signals for time series and spatial representation learning. Second, to encode more discriminative features representing emotions, two-dimensional (2D)-CNN, BiGRU, and SAM are trained on these scalograms simultaneously to capture the appropriate information from spatial, local, temporal, and global aspects. Finally, EEG signals are categorized into several emotional states. This network can learn the temporal dependencies of EEG emotion signals with BiGRU, extract local spatial features with TFCNN, and improve recognition accuracy with SAM, which is applied to explore global signal correlations by reassigning weights to emotion features. Using the SEED and GAMEEMO datasets, the suggested strategy was evaluated on three different classification tasks: one with two target classes (positive and negative), one with three target classes (positive, neutral, and negative), and one with four target classes (boring, calm, horror, and funny). Based on the comprehensive results of the experiments, the suggested approach achieved a 93.1%, 96.2%, and 92.9% emotion detection accuracy in two, three, and four classes, respectively, which are 0.281%, 1.98%, and 2.57% higher than the existing approaches working on the same datasets for different subjects, respectively. The open source codes are available at https://www.mathworks.com/matlabcentral/fileexchange/165126-tfcnn-bigru

Global optimization and biomedical image segmentation are crucial in diverse scientific and medical fields. The Honey Badger Algorithm (HBA) is a newly developed metaheuristic that draws inspiration from the foraging behavior of honey badgers. Similar to other metaheuristic algorithms, HBA encounters difficulties associated with exploitation, being trapped in local optima, and the pace at which it converges. This study aims to improve the performance of the original HBA by implementing the Enhanced Solution Quality (ESQ) method. This strategy helps to prevent becoming stuck in local optima and speeds up the convergence process. We conducted an assessment of the enhanced algorithm, mHBA, by utilizing a comprehensive collection of benchmark functions from IEEE CEC’2020. In this evaluation, we compared mHBA with well-established metaheuristic algorithms. mHBA demonstrates exceptional performance, as shown by both qualitative and quantitative assessments. Our study not only focuses on global optimization but also investigates the field of biomedical image segmentation, which is a crucial process in numerous applications involving digital image analysis and comprehension. We specifically focus on the problem of multi-level thresholding (MT) for medical image segmentation, which is a difficult process that becomes more challenging as the number of thresholds needed increases. In order to tackle this issue, we suggest a revised edition of the standard HBA, known as mHBA, which utilizes the ESQ approach. We utilized this methodology for the segmentation of Magnetic Resonance Images (MRI). The evaluation of mHBA utilizes existing metrics to gauge the quality and performance of its segmentation. This evaluation showcases the resilience of mHBA in comparison to many established optimization algorithms, emphasizing the effectiveness of the suggested technique.

Software-defined networking (SDN) has emerged as a transformative technology that separates the control plane from the data plane, providing advantages such as flexibility, centralized control, and programmability. This innovation proves particularly beneficial for Internet of Vehicles (IoV) networks, which amalgamate the Internet of Things (IoT) and Vehicular Ad Hoc Network (VANET) to implement Intelligent Transportation Systems (ITS). IoV provides a safe and secured vehicular environment by supporting V2V, V2I, V2S, and V2P. By employing an SDN controller, IoV networks can leverage centralized control and enhanced manageability, leading to the emergence of Software-Defined Internet of Vehicles (SD-IoV) as a promising solution for future communications. However, the SD-IoV networks introduces a potential vulnerability in the form of a single point of failure, particularly susceptible to Distributed Denial of Service (DDoS) attacks. This is because of the centralized nature of SDN and the dynamic nature of IoV. In this context, the SDN controller becomes a prime target for attackers who flood it with massive packet-in messages. To address this security concern, we propose an efficient and lightweight attack detection and mitigation scheme within the SDN controller. The scheme includes a detection module that utilizes entropy and flow rate to identify patterns indicative of attack traffic behavior. Additionally, a mitigation module is designed to minimize the effect of attack traffic on the normal operation, this is performed through analysis of payload lengths.The mitigation flow rule is set for specific traffic type if its payload is less than the threshold value to decrease the false positive rate. An adaptive threshold computation for all parameter values enhances the scheme’s effectiveness. We conducted simulations using SUMO, Mininet-WiFi, and Scapy. We evaluated the system performance by using Mininet-wifi SDN simulation tool and Ryu controller for control plane. The system detects DDoS attack traffic within a single window by checking both entropy and flow rate simultaneously. The simulation results demonstrate the efficacy of our proposed scheme in terms of detection time, accuracy, mitigation efficiency, controller load, and link bandwidth consumption, showcasing its superiority compared to existing works in the field.

With the rapid development of the logistics industry and the continuous growth of e-commerce, effectively monitoring logistics warehouses has become increasingly important to ensure the security of goods and oversee activities within storage facilities. Although current surveillance systems provide a certain level of security for logistics warehouses, they still face issues such as data tampering, storage, and access management. These challenges can compromise the integrity of surveillance video data, making the system vulnerable to unauthorized access. To address these challenges, this paper proposes the implementation of blockchain-based security management and access control of video data in logistics warehouses. Specifically, the solution employs the Hyperledger Fabric consortium blockchain to execute smart contracts and store the hash values of video data, thereby detecting any tampering and enhancing the security and integrity of the data. Additionally, hybrid encryption technology is utilized to ensure the confidentiality of video data during transmission and storage. Furthermore, the solution leverages the InterPlanetary File System (IPFS) for distributed video storage. This not only increases the redundancy and accessibility of data storage but also reduces the risk of single-point failures. A Role-Based Access Control (RBAC) mechanism is also introduced to strictly manage access permissions to video data, ensuring that only authorized users can access the data, thereby effectively preventing unauthorized access and data breaches. Through a comprehensive analysis of computational and communication costs and the evaluation of blockchain performance at 100 transactions per second for different transaction volumes using Hyperledger Caliper, the results demonstrate the effectiveness and efficiency of the proposed method. Compared to current research, this solution exhibits higher security, providing a new approach for the secure management and access control of video data in logistics warehouses.

Virtual machine placement (VMP) directly impacts the energy consumption, resource utilization, and service quality of cloud data centers (CDCs), and it has become an active research topic in cloud computing. Inspired by the ant colony system (ACS) which has been proven effective metaheuristic approach for solving NP-hard problems, this paper proposes an improved ACS-based energy efficiency strategy (EEACS) for VMP problems. Our approach considers each virtual machine (VM) as an energy-consuming block, taking into account its individual energy requirements. EEACS ranks the physical machines (PMs) in a CDC in descending order based on their energy efficiency and optimizes both server selection and pheromone updating rules within the ACS. By guiding artificial ants towards promising solutions that balance energy consumption and resource utilization, EEACS ensures that VMs are placed efficiently based on pheromone and heuristic information. Extensive simulations in both homogeneous and heterogeneous computing environments demonstrate the effectiveness of our proposed strategy. The experimental results show that the EEACS enhances the resource utilization and achieves a notable reduction in energy consumption in comparison to conventional heuristic and evolutionary-based algorithms.

Massive Machine Type Communication (mMTC) uses sub-6 GHz bands frequency bands for their applications in the 5G network. The exponential growth of mMTC wireless networks has made these bands overcrowded. Due to the increase in wireless traffic, spectrum scarcity is a significant constraint at sub-6 GHz bands for the 5G and beyond networks. Cognitive radio technology uses spectrum sensing (SS) techniques to access the spectrum opportunistically to resolve this issue where signal processing techniques (SPTs) are considered to design SS. However, the adaptiveness of SPTs is not feasible in the real-time environment due to the random spectrum access behaviour of the primary user and the fading environment. To minimize this issue, machine learning (ML) models are adapted. Different ML models are examined, and their performance is analyzed to find a better accuracy model in spectrum hole identification at 5G sub-6 GHz bands. The dataset of large-scale frequency samples is built from a universal software radio peripheral (USRP—2953R) at sub-6 GHz bands over (eta - mu) fading environmental conditions. A highly imbalanced dataset issue is reduced and compared with varying resampling techniques, and random oversampling is best to resolve the anomalies in the dataset. Random forest, naive Bayes, logistic regression, K-nearest neighbor, and decision trees are primary classifiers to train and detect the spectrum holes at 5G sub-6 GHz bands. Random forest outperforms the remaining ML models in spectrum hole identification in terms of probability of detection and accuracy.

Accurate channel estimation poses a significant challenge in the reconfigurable intelligent surface (RIS)-aided millimeter-wave (mmWave) wireless communication system. The fully passive nature of the RIS primarily relies on cascaded channel estimation, given its limitation in transmitting and receiving signals. Although the advantageous of this approach, the increase in the number of RIS elements leads to an exponential growth in the channel coefficient, resulting in costly pilot overhead. To address this challenge, the paper proposes a two-phase framework for separate channel estimation. The framework involves incorporating a few active elements within the passive RIS, enabling the reception and processing of pilot signals at the RIS. Through leveraging the difference in coherence time of the channel, the estimation of the time-varying channel among user equipment (UE) and RIS, as well as the estimation of the pseudo-static channel among RIS and base station (BS), can be performed separately. The two-phase separate channel estimation framework operates as follows: In the first phase, the BS-RIS channel is estimated at the RIS through the utilization of the few active elements. An iterative weighting methodology is employed to formulate the mathematical optimization problem for estimating the BS-RIS signal model. Subsequently, a proposed algorithm grounded on gradient descent (GD) is introduced to efficiently address and solve the optimization problem. In the second phase, the estimation of the UE-RIS channel is achieved by transforming the signal model of the received channel into an analogous tensor model known as Parallel Factor (PARAFAC). This transformation is followed by the application of the least squares (LS) algorithm within this tensor-based representation at BS. The effectiveness of the proposed framework is demonstrated through simulation findings, considering minimum pilot overhead, average spectral efficiency, and normalized mean square error (NMSE). A comparative analysis is performed with three other state-of-the-art existing schemes.

The relevance of images in people's daily lives is growing, and content-based image retrieval (CBIR) has received a lot of attention in research. Images are much better at communicating information than text documents. This paper deals with security and efficient retrieval of images based on the texture features extracted by the dominant local patterns of an image in cloud. Here, we proposed a method that supports secure and efficient image retrieval over cloud. The images are encrypted with the watermark before deploying the image database to the cloud, this process prevents from the outflow of sensitive information to the cloud. A reduced dimension feature vector database has been created for all the images using relative directional edge patterns (RDEP), facilitating efficient storage and retrieval. The significance of the specified local pattern for effectively extracting texture information has been demonstrated. A notable level of accuracy has been established when compared to existing algorithms in terms of precision and recall. Additionally, a watermark-based system is proposed to prevent unauthorized query users from illicitly copying and distributing the acquired images to others. An inimitable watermark is entrenched into the image by the encryption module before storing into the cloud. Hence, when an image copy is discovered, the watermark extraction can be used to track down the illegal query image user who circulated the image. The proposed method's significance is assessed by comparing it to other existing feature extractors incorporating watermark encryption. Additionally, the effectiveness of the method is demonstrated across various numbers of watermark bits. Trials and security analyses affirm that the suggested approach is both robust and efficient.