IET Communications最新文献

Machine learning faces challenges in detecting outliers, especially in high-dimensional datasets. Effective data quality is crucial for better results, and many algorithms identify outliers by analysing outlying aspects of data objects and objects within the dataset. The proposed Advanced Distance-Based Unsupervised Local Outlier Detection (DU-LOD) method improves this process by continuously evaluating and identifying outliers using unsupervised learning and distance-based calculations. DU-LOD identifies outliers by comparing differences between data objects and their neighbours, making it the first method to combine unsupervised local outlier detection with nearest cluster point identification. Experimental analysis through accuracy performance of 96.12%, detection rate performance of 41.89%, precision of 56.12%, and recall of 1.79% proves that our model performs best over the various parameters compared with other existing algorithms. Therefore, measures such as area under the ROC curve (AUC), precision and recall are more appropriate in such a scenario.

This work presents a comprehensive study on the performance of an asymmetric dual-hop radio frequency (RF)/free-space optical (FSO) transmission system that utilizes a fixed gain amplify-and-forward protocol. The RF path is characterized by the Beaulieu–Xie model, and the FSO link is affected by Málaga ($�M$) turbulence in the presence of pointing errors. In this study, we introduce novel and precise analytical expressions for the probability density function, the cumulative distribution function (CDF), and the moment generating function (MGF) of the overall signal-to-noise ratio (SNR). Using the derived expressions, we proceed to obtain accurate infinite series representations for the outage probability (OP), the average bit-error rate (BER), and the ergodic capacity (EC). In addition, we conduct asymptotic analysis for the CDF, the MGF, the OP, the average BER, and the EC in the high SNR regime. Numerical results are compared with the Monte Carlo simulations to validate the accuracy of these derived expressions.

Nowadays, mobile robot navigation is a crucial topic in the Industrial Internet of Things, and a number of sensors are arranged around robots to avoid obstacles on navigation paths. Obtaining optimal sensor readings that can be used to optimize the path planning of mobile robots is essential. Thus, an effective feature selection technique is explored in this study to select the optimal subset of sensor readings for mobile robot navigation. Feature selection can reasonably remove irrelevant and redundant features, reduce the dimensionality of data, and improve the learning accuracy and comprehensibility. A novel feature selection method based on 3D mutual information is studied. First, the proposed method defines symmetrical uncertainty with 3D mutual information to measure the correlation between candidate features and select features. Afterward, a merit function of feature sets is defined based on the symmetrical uncertainty to search for the optimal feature subset. Lastly, the proposed feature selection method is applied to a wall-following robot navigation data set. Results show that the proposed method can obtain few sensor readings but with enhanced prediction accuracy of the robot's movements.

Over the past few years, there has been significant research on the Internet of Things (IOT), with a major challenge being network security and penetration. Security solutions require careful planning and vigilance to safeguard system security and privacy. This paper proposes a new hybrid intrusion detection system (IDS) based on machine learning and metaheuristic algorithms, which has 3 stages: (1) Pre-processing, (2) feature selection, and (3) attack detection. In the pre-processing stage including, cleaning, visualization, feature engineering and vectorization. In the feature selection stage, a combined grasshopper optimization algorithm and sine–cosine algorithm is used; the modified grasshopper algorithm improves the performance of the grasshopper algorithm with a centralized population initialization in terms of search capability, convergence speed, and capacity to deviate from the local optimum. In the attack detection stage, a random neural network is used, and the modified grasshopper algorithm adjusts the structure and parameters of the random neural network. The proposed method is evaluated using the DS2OS datasets, CIC-IOT2023 and CIC-IDS2018. The results have shown that the proposed approach in these experiments, through a multiple learning model, resulted in an improvement in accuracy to 99.56%.

Relay-assisted symbiotic radio (SR) has been recently proposed to overcome the blocking of the direct link from the primary transmitter (PT) or backscatter node (BN) to the destination node (DN). However, the energy efficiency (EE), which is an important performance metric for SR networks, has been largely ignored in existing studies of the relay-assisted SR. To fill the gap, this work maximizes the EE of a relay-assisted parasitic SR network, which comprises a PT, a BN, a relay node (RN), and a DN. More specifically, we formulate a mixed-integer programming optimization problem that maximizes the system EE by jointly optimizing the transmit power of the PT, the power reflection coefficient of the BN, the transmit power and the power allocation ratio at the RN as well as the successive interference cancellation (SIC) decoding order at the DN. We decompose the formulated non-convex problem into two subproblems corresponding to the two different SIC decoding orders, respectively. For each subproblem, we convert its objective function from a fractional form into a subtractive form by using a Dinkelbach-based method, and then utilize the block coordinate descent (BCD) method to further decouple it into two subsubproblems that are proved to be convex. Based on the obtained solutions, we devise an iterative algorithm to solve each subproblem by solving its two subsubproblems alternately. The optimal solution to the subproblem with a higher system EE returns a near-optimal solution to the original problem. Simulation results demonstrate the rapid convergence of the proposed algorithms and validate the significant advantages of our proposed algorithms over the baseline schemes.

Multiplayer virtual reality (VR) games represent the increasing future direction of VR game development. However, currently it still falls short in terms of the fluidity of social interaction, response speed, and sense of immersion, which affects players' engagement and satisfaction. This paper proposes an optimisation design for multiplayer VR game systems that integrates BP neural networks based on an intelligent computing framework. By applying the fast convergence of traditional BP algorithms and enhancing it with genetic algorithms to improve global search capabilities and avoid local optima, ensuring more accurate and efficient neural network training for enhanced VR gaming experiences. This paper compares the performance of traditional neural networks and evolutionary neural networks through extensive simulation and testifies to the engagement experience of players through quantitative analysis. The results show that evolutionary neural networks outperform traditional neural networks in system performance, such as severe latency and technical lagging. The paper also finds that technological preparedness significantly affects behaviour engagement through embodied social presence, emotional and cognitive engagement. Based on these findings, this paper suggests strategies to optimise the user experience of multiplayer VR games by improving game technology quality, enriching content, maintaining continuous communication with players, and establishing reasonable incentive mechanisms.

Battery-limited devices and time-sensitive applications are considered as key players in forthcoming wireless sensor network. So, the main goal of the network is two-fold; charge battery-limited devices, and provide status updates to users where information-freshness matters. In this paper, a multi-antenna base station (BS) in assistance of simultaneously-transmitting-and-reflecting reconfigurable intelligent surface (STAR-RIS) transmits power to energy harvesting (EH) devices while controlling status update performance at information users by analysing age of information (AoI) metric. Therefore, we derive a scheduling policy at BS, and analyse joint transmit beamforming and amplitude-phase optimization at BS and STAR-RIS, respectively, to reduce average sum-AoI for the time-sensitive information users while satisfying minimum required energy at EH users. Moreover, two different energy-splitting and mode-switching policies at STAR-RIS are studied. Then, by use of an alternating optimization algorithm, the optimization problem is studied and non-convexity of the problem is tackled by using the successive convex approximation technique. Through numerical results, AoI-metric and EH requirements of the network are analysed versus different parameters such as number of antennas at BS, size of STAR-RIS, and transmitted power to highlight how we can improve two-fold performance of the system by utilizing STAR-RIS compared to the conventional RIS structure.

The emergence of terahertz (THz) communication in ultra-massive multiple-input multiple-output (UM-MIMO) systems presents new challenges for accurate and efficient channel estimation, particularly under hybrid-field propagation conditions. Conventional estimation techniques struggle to meet the demands of such high-dimensional systems, especially in the presence of limited radio frequency (RF) chains and mixed near- and far-field effects. To address these limitations, this paper proposes a deep learning-based framework that combines a fully connected neural network (FCNN) for linear channel estimation with a convolutional neural network (CNN) for non-linear refinement. The architecture is designed to adapt to diverse propagation environments while maintaining computational efficiency. Simulation studies based on realistic THz scenarios demonstrate that the proposed approach significantly improves estimation accuracy, achieving up to 90% reduction in normalized mean squared error (NMSE) compared to traditional and advanced estimation techniques. The robustness of the model under varying signal-to-noise ratios and noise power levels underscores its potential for deployment in future 6G THz communication networks.

Malignant lung nodules can significantly affect patients' normal lives and, in severe cases, threaten their survival. Owing to the heterogeneity of computed tomography scans and the varying sizes of nodules, physicians often face challenges in diagnosing this condition. Therefore, a novel adaptive multi-channel fusion network (AMCF-Net) is proposed for computer-aided diagnosis of lung nodules. First, a Multi-Channel Fusion Model module is designed, which divides the channels into two parts in specific proportions, effectively extracting multi-scale channel information while reducing network parameters. After the feature maps output at each layer of the AMCF-Net, a novel adaptive depth-wise separable convolution with a squeeze-and-excitation module is designed to adaptively integrate the feature maps of various stages of the AMCF-Net, ensuring that the key lesions of lung nodules are not lost during classification. Finally, a hybrid loss scheme based on an adaptive mixing ratio is proposed to solve the problem of an imbalanced number of positive and negative nodule samples in the dataset. The model achieved the following test results: an accuracy of 90.22%, a specificity of 98.19%, an F1-score of 86.57%, a sensitivity of 86.49%, and a G-mean of 87.72%. Compared with other advanced networks, AMCF-net delivers high-precision lung nodule classification with minimal inference cost. Related codes have been released at: https://github.com/GuYuIMUST/AMCF-net.

In this paper, we investigate the computation offloading and resource allocation strategy of the coexistence and synergy between fog computing and cloud computing in heterogeneous networks. Consider that the reported schemes have prohibitive complexity when achieving the optimal computation offloading strategy in cloud-fog cooperative heterogeneous networks, an improved genetic algorithm (IGA) is proposed in this paper, which can maintain a low computation complexity while obtaining the optimal solution. In the IGA algorithm, we propose to use a penalty function to express the constraint conditions of the optimisation problem and use a non-uniform mutation operator to accelerate the convergence speed. Besides, an improved method of parameter self-adaptation and a perturbation method of mutation probability based on population fitness standard deviation are proposed to optimise the genetic algorithm. The numerical results show that the proposed genetic algorithm can obtain a lower average cost of the system while keeping a smaller computational cost.