Recent Advances in Computer Science and Communications - Volume 18, Issue 5, 2025

Emotions play a significant role in human relationships in everyday challenges. Human-machine interactions and the complexities of its research include humanoid robots which rely significantly on emotional queries. The present study recommends a novel methodology for human emotion classification using electroencephalogram (EEG) signals. Machine learning (ML) techniques based on electroencephalogram have demonstrated impressive results in distinguishing emotions.

The present study investigates five ensemble learning-based machine learning (EML) and five conventional machine learning (CML) algorithms for achieving the recognition of human emotions from EEG signals. The main objective of the current research study is to implement the optimal approach for classifying emotions and subsequently will achieve it by using physiological characteristics to categorise the seven main emotional states (stress, happy, sadness, anger, fear, disgust, and surprise). This research follows the study investigation of ensemble learning-based machine learning (EML) and conventional machine learning (CML) algorithms for achieving the recognition of human emotions from EEG signals. Initially, the EEG data in this study are categorized into theta, alpha, beta, and gamma bands using a DWT.

With this idea, the research elaborates on further decomposition based on band-separated EEG signals known as intrinsic mode functions (IMFs) along with empirical mode decomposition is subsequently implemented. Then, 13 statistical characteristics are extracted from the IMFs using five multiclass EML algorithms: adaptive boosting, rotation random forest, bagging, random forest, and extreme gradient boost. The following procedure is carried out to construct an ML-based system. In the final phase, 10-fold cross-validation is used to assess the efficacy of these five EML algorithms. Accuracy, F1-score, and area-under-the-curve (AUC) are performance assessment metrics are employed to compare the final results of these algorithms with the five CML techniques.

The proposed method for EEG-based emotion analysis is significantly more effective and achieves greater outcomes than all of the techniques that were compared, based on the data from the experiments.

Recorded brain activity analysis is a major tool used in modern medicine and sleep-based research. However, the volume and complexity of the data make manual investigation difficult. Furthermore, high inter-subject variability presents a challenge to deep learning techniques, especially in the detection of sleep-related EEG patterns and artifacts.

Sleep Pattern Identifier (SPI), a hybrid classifier, was used in this study to address the aforementioned issues. SPI tests the efficacy of artifact detection while examining strategies and tactics used in real-world sleep research. A focus was on inter-subject variability, especially in data gathered from participants suffering from sleep disorders. For comparison and visual inspection of the data, formal statistical measures like accuracy, model loss, precision, recall, and ROC were employed.

With an accuracy of 94.85%, the suggested model outperformed current techniques and demonstrated higher accuracy. Further, the Sleep-EDF dataset was used in this investigation.

The conclusion highlights the effectiveness of the SPI hybrid classifier in identifying EEG patterns, especially in situations where sleep research is conducted in the real world. Research methodologies based on sleep have made significant progress in handling artifacts and adapting to inter-subject variability.

Virtual machines are used to reduce cloud platform application performance, management costs, and access irregularities. Virtual machines are frequently vulnerable to delays, overburdening workloads, and other obstacles while consolidating and migrating servers. To significantly disperse loads among virtual machines, dynamic consolidation techniques are implemented to control energy dissipation, monitor overloading, and address underloading problems. The process of consolidation involves more calculations and resources in order to transfer services between virtual machines, provided that Service Level Agreements are observed.

The suggested approach promotes the use of cutting-edge architecture to combine virtual machines, and, therefore, strike a balance between performance and energy requirements. The main design considerations for the suggested Dynamic Weightage algorithm, which includes the clustering approach in relation to reinforcement learning approaches, are overall resource needs and Performance to Power Ratio (PPR). A cluster of ideal virtual machines is created, and resources are distributed according to performance and energy requirements. Virtual machine resource requests are converted into a matching relationship factor, which represents the individual hosts while taking PPR into account. The overall workload associated with virtual machine consolidation is also provided by these estimations. It is noted that there is little energy trade-off and that performance is maintained at a nominal level across the cluster. The architecture is put into practice throughout offline platforms, which are dispersed ecosystems that allow for increased system performance and scaling.

The CloudSim simulator is used to validate the system using datasets that are obtained from PlanetLab. According to the data, energy saving has produced yields of up to 47% and promising quality of service attributes.

The validation of the system is performed using the CloudSim simulator with datasets from PlanetLab. The results indicate significant energy conservation, up to 47%, along with promising quality of service parameters. The proposed architecture is compared with other state-of-the-art algorithms for distributed architectures and heterogeneous environments, showcasing its efficiency. The conclusion emphasizes the prioritization of VM consolidation and energy efficiency in the proposed architecture, which has been tested on a Proliant G7-based data center using a variety of hosts. Notably, the CloudSim Toolkit is highlighted as outperforming OpenStack-based techniques in simulation results.

Cyber-physical systems (CPSs) integrate computing, control, and communication technologies, bridging cyberspace and the physical world to enhance critical infrastructure and safety-critical systems. Existing surveys often address CPS security from a single perspective, necessitating a more comprehensive approach.

This paper presents a thorough analysis of CPS security from three perspectives: the physical domain, the cyber domain, and the cyber-physical domain. We examine attacks on physical components like sensors, cyber-attacks targeting CPSs, and integrated cyber-physical attacks. For each domain, we analyse corresponding detection and defence mechanisms.

Our study reveals that CPSs face significant security threats across all domains. In the physical domain, attacks on sensors can disrupt system operations, but various defences are available. In the cyber domain, CPSs are vulnerable to malware, hacking, and denial-of-service attacks, with several detection and defence strategies in place. The cyber-physical domain highlights complex threats where cyber-attacks cause physical damage, requiring comprehensive security approaches.

By examining CPS security from multiple perspectives, this study provides a holistic understanding of current threats and defence mechanisms. It identifies future research directions to enhance CPS security, aiming to better protect critical infrastructure against evolving cyber threats.

Wireless networks are essential communication technologies that prevent cable installation prices and burdens. Because of this technology's pervasive usage, wireless network safety is a significant problem. Owing to distributed and open wireless medium aspects, attackers might use different jamming methods to exploit physical and MAC layer protocol vulnerabilities. In addition, jamming attacks require to be accurately grouped so that suitable countermeasures can be considered. Given the potential severity of such attacks, precisely identifying and classifying them is critical for implementing effective responses. The motivation for this paper is the need to improve the detection and categorization of jamming signals using modern machine learning algorithms, consequently enhancing wireless network security and reliability.

In this paper, we compare some machine learning models' efficiency for diagnosing jamming signals.

Such algorithms refer to support vector machine (SVM) and k-nearest neighbors (KNN). We checked the signal features that recognize jamming signals. After the jamming attack model, the developed grey wolf optimizer version known as IGWO (improved grey wolf optimizer) has been discussed for feature extraction of software usability. Four separate metrics were employed as features to detect jamming attacks in order to evaluate the machine learning models. This novel feature extraction method is crucial for improving the accuracy of jamming detection.

The measurements of these parameters were gathered through a simulation of a real setting. And generated a large dataset using these parameters.

The simulation results illustrate that the KNN algorithm based on jamming detection could diagnose jammers having a minimal likelihood of false alarms and a high level of accuracy.

This study analyzes the transmission of the current signal of an electronic analgesic apparatus in arm muscles and provides a theoretical foundation for electrical stimulation analgesia.

By combining human anatomy and tissue structure, a numerical simulation-based finite element model of the electronic analgesic apparatus is established using a frustum, cylinder, and ellipsoid as geometric entities in COMSOL Multiphysics 5.5. In the frequency domain environment, the transmission mechanism of the signal in the arm is analyzed by inputting current signals of 100 kHz, 1 MHz, and 10 MHz with an amplitude of ±20 mA.

With a continuous increase in carrier frequency, the effect of skin tissue becomes increasingly clear, and the signal becomes increasingly concentrated in the part of the skin that contacts the electrode. Moreover, diffusion inside the volume conductor loses consistency. At 100 kHz, as the communication distance from the electrode center gradually increases, the signal spreads more evenly within the arm.

In the process of implementing a muscle soreness treatment using the electronic analgesic apparatus, a higher communication frequency makes it more difficult for the signal to enter the interior of the body and degrade the consistency of the signal. Therefore, the signal electrode should be placed as close as possible to the analgesic target area during the implementation of electric current analgesia.