High Detection Accuracy Research Articles

Theoretical study of a triple-band tunable terahertz metamaterial absorber with polarization insensitive and high refractive index sensing based on bulk Dirac semimetal

In this paper, a triple-band metamaterial absorber in the terahertz frequencies is proposed, and its refractive index sensing characteristics are analyzed, where the bulk Dirac semimetal (BDS) periodic array is on top of a photonic crystal slab backed with a metal ground plane. The simulation results show that the absorber achieves three perfect absorption peaks in the range of 3.4–5.2 THz, whose absorption rates are over 96%, and a maximum quality factor (Q) of 74.1. The designed absorber exhibits excellent polarization insensitivity and dynamic tunability; further, the tuning of the Fermi energy level of BDS enables the dynamic adjustment of absorption frequencies and absorption rates of these peaks. By analyzing the distributions of the electromagnetic field and different structural parameters, it is revealed that the absorber mainly dissipates the electromagnetic wave through coupled resonance and localized surface plasmon resonance (LSPR) effects to achieve perfect absorption. Further, the metamaterial absorber shows the capacity to detect analytes with varying refractive indices, and the absorber has a maximum sensitivity S of 405 GHz/RIU with high detection accuracy. This work provides novel design options for triple-band terahertz metamaterial absorbers and their potential applications in refractive index sensing.

An Adaptive and Automatic Power Supply Distribution System with Active Landmarks for Autonomous Mobile Robots.

With the development of automation and intelligent technologies, the demand for autonomous mobile robots in the industry has surged to alleviate labor-intensive tasks and mitigate labor shortages. However, conventional industrial mobile robots' route-tracking algorithms typically rely on passive markers, leading to issues such as inflexibility in changing routes and high deployment costs. To address these challenges, this study proposes a novel approach utilizing active landmarks-battery-powered luminous landmarks that enable robots to recognize and adapt to flexible navigation requirements. However, the reliance on batteries necessitates frequent recharging, prompting the development of an automatic power supply system. This system integrates omnidirectional contact electrodes on mobile robots, allowing to recharge active landmarks without precise positional alignment. Despite these advancements, challenges such as the large size of electrodes and non-adaptive battery charging across landmarks persist, affecting system efficiency. To mitigate these issues, this research focuses on miniaturizing active landmarks and optimizing power distribution among landmarks. The experimental results of this study demonstrated the effectiveness of our automatic power supply method and the high accuracy of landmark detection. Our power distribution calculation method can adaptively manage energy distribution, improving the system's persistence by nearly three times. This study aims to enhance the practicality and efficiency of mobile robot remote control systems utilizing active landmarks by simplifying installation processes and extending operational durations with adaptive and automatic power supply distribution.

Comparative Study of Lightweight Target Detection Methods for Unmanned Aerial Vehicle-Based Road Distress Survey.

Unmanned aerial vehicles (UAVs) are effective tools for identifying road anomalies with limited detection coverage due to the discrete spatial distribution of roads. Despite computational, storage, and transmission challenges, existing detection algorithms can be improved to support this task with robustness and efficiency. In this study, the K-means clustering algorithm was used to calculate the best prior anchor boxes; Faster R-CNN (region-based convolutional neural network), YOLOX-s (You Only Look Once version X-small), YOLOv5-s, YOLOv7-tiny, YOLO-MobileNet, and YOLO-RDD models were built based on image data collected by UAVs. YOLO-MobileNet has the most lightweight model but performed worst in accuracy, but greatly reduces detection accuracy. YOLO-RDD (road distress detection) performed best with a mean average precision (mAP) of 0.701 above the Intersection over Union (IoU) value of 0.5 and achieved relatively high accuracy in detecting all four types of distress. The YOLO-RDD model most successfully detected potholes with an AP of 0.790. Significant or severe distresses were better identified, and minor cracks were relatively poorly identified. The YOLO-RDD model achieved an 85% computational reduction compared to YOLOv7-tiny while maintaining high detection accuracy.

An FPGA-Based YOLOv5 Accelerator for Real-Time Industrial Vision Applications.

The You Only Look Once (YOLO) object detection network has garnered widespread adoption in various industries, owing to its superior inference speed and robust detection capabilities. This model has proven invaluable in automating production processes such as material processing, machining, and quality inspection. However, as market competition intensifies, there is a constant demand for higher detection speed and accuracy. Current FPGA accelerators based on 8-bit quantization have struggled to meet these increasingly stringent performance requirements. In response, we present a novel 4-bit quantization-based neural network accelerator for the YOLOv5 model, designed to enhance real-time processing capabilities while maintaining high detection accuracy. To achieve effective model compression, we introduce an optimized quantization scheme that reduces the bit-width of the entire YOLO network-including the first layer-to 4 bits, with only a 1.5% degradation in mean Average Precision (mAP). For the hardware implementation, we propose a unified Digital Signal Processor (DSP) packing scheme, coupled with a novel parity adder tree architecture that accommodates the proposed quantization strategies. This approach efficiently reduces on-chip DSP utilization by 50%, offering a significant improvement in performance and resource efficiency. Experimental results show that the industrial object detection system based on the proposed FPGA accelerator achieves a throughput of 808.6 GOPS and an efficiency of 0.49 GOPS/DSP for YOLOv5s on the ZCU102 board, which is 29% higher than a commercial FPGA accelerator design (Xilinx's Vitis AI).

Forensic analysis of bad USB attacks: A methodology for detecting and mitigating malicious USB device activities

BadUSB is one of the most dangerous cybersecurity threats, given that it uses the firmware of USB devices to perform various undetectable actions with numerous tools. This research aims to evaluate the efficiency of different forensic approaches, such as signature-based detection, behavioral analysis, and the machine learning (ML) approach, in detecting and analyzing BadUSB attacks. Experiments were conducted with preconfigured USB peripherals to perform keystroke injection, data exfiltration, malware delivery, and network traffic manipulation. The analysis shows that the behavioral analysis and the ML-based methods show high detection accuracy and low false positives. Machine learning detection is the most efficient method. Behavioral analysis had higher accuracy in detecting abnormal device behavior but had a longer detection time than the ML methods. This research beneficently addresses the issues and challenges in the field of digital forensics and calls for further improvement in the detection methods. It proposes ways to implement these methods within the existing cybersecurity models. Future studies should focus on the best approaches to fine-tune these techniques, diversify datasets for machine learning detection methods, and advance methodologies in forensics to accommodate new generations of technologies like the Internet of Things and cloud systems.

SBD-Net: Incorporating Multi-Level Features for an Efficient Detection Network of Student Behavior in Smart Classrooms

Detecting student behavior in smart classrooms is a critical area of research in educational technology that significantly enhances teaching quality and student engagement. This paper introduces an innovative approach using advanced computer vision and artificial intelligence technologies to monitor and analyze student behavior in real time. Such monitoring assists educators in adjusting their teaching strategies effectively, thereby optimizing classroom instruction. However, the application of this technology faces substantial challenges, including the variability in student sizes, the diversity of behaviors, and occlusions among students in complex classroom settings. Additionally, the uneven distribution of student behaviors presents a significant hurdle. To overcome these challenges, we propose Student Behavior Detection Network (SBD-Net), a lightweight target detection model enhanced by the Focal Modulation module for robust multi-level feature fusion, which augments feature extraction capabilities. Furthermore, the model incorporates the ESLoss function to address the imbalance in behavior sample detection effectively. The innovation continues with the Dyhead detection head, which integrates three-dimensional attention mechanisms, enhancing behavioral representation without escalating computational demands. This balance achieves both a high detection accuracy and manageable computational complexity. Empirical results from our bespoke student behavior dataset, Student Classroom Behavior (SCBehavior), demonstrate that SBD-Net achieves a mean Average Precision (mAP) of 0.824 with a low computational complexity of just 9.8 G. These figures represent a 4.3% improvement in accuracy and a 3.8% increase in recall compared to the baseline model. These advancements underscore the capability of SBD-Net to handle the skewed distribution of student behaviors and to perform high-precision detection in dynamically challenging classroom environments.

Model optimization and acceleration method based on meta-learning and model pruning for laser vision weld tracking system

Purpose This paper aims to propose a lightweight, high-accuracy object detection model designed to enhance seam tracking quality under strong arcs and splashes condition. Simultaneously, the model aims to reduce computational costs. Design/methodology/approach The lightweight model is constructed based on Single Shot Multibox Detector (SSD). First, a neural architecture search method based on meta-learning and genetic algorithm is introduced to optimize pruning strategy, reducing human intervention and improving efficiency. Additionally, the Alternating Direction Method of Multipliers (ADMM) is used to perform structural pruning on SSD, effectively compressing the model with minimal loss of accuracy. Findings Compared to state-of-the-art models, this method better balances feature extraction accuracy and inference speed. Furthermore, seam tracking experiments on this welding robot experimental platform demonstrate that the proposed method exhibits excellent accuracy and robustness in practical applications. Originality/value This paper presents an innovative approach that combines ADMM structural pruning and meta-learning-based neural architecture search to significantly enhance the efficiency and performance of the SSD network. This method reduces computational cost while ensuring high detection accuracy, providing a reliable solution for welding robot laser vision systems in practical applications.

YOLOv8-Based Drone Detection: Performance Analysis and Optimization

The extensive utilization of drones has led to numerous scenarios that encompass both advantageous and perilous outcomes. By using deep learning techniques, this study aimed to reduce the dangerous effects of drone use through early detection of drones. The purpose of this study is the evaluation of deep learning approaches such as pre-trained YOLOv8 drone detection for security issues. This study focuses on the YOLOv8 model to achieve optimal performance in object detection tasks using a publicly available dataset collected by Mehdi Özel for a UAV competition that is sourced from GitHub. These images are labeled using Roboflow, and the model is trained on Google Colab. YOLOv8, known for its advanced architecture, was selected due to its suitability for real-time detection applications and its ability to process complex visual data. Hyperparameter tuning and data augmentation techniques were applied to maximize the performance of the model. Basic hyperparameters such as learning rate, batch size, and optimization settings were optimized through iterative experiments to provide the best performance. In addition to hyperparameter tuning, various data augmentation strategies were used to increase the robustness and generalization ability of the model. Techniques such as rotation, scaling, flipping, and color adjustments were applied to the dataset to simulate different conditions and variations. Among the augmentation techniques applied to the specific dataset in this study, rotation was found to deliver the highest performance. Blurring and cropping methods were observed to follow closely behind. The combination of optimized hyperparameters and strategic data augmentation allowed YOLOv8 to achieve high detection accuracy and reliable performance on the publicly available dataset. This method demonstrates the effectiveness of YOLOv8 in real-world scenarios, while also highlighting the importance of hyperparameter tuning and data augmentation in increasing model capabilities. To enhance model performance, dataset augmentation techniques including rotation and blurring are implemented. Following these steps, a significant precision value of 0.946, a notable recall value of 0.9605, and a considerable precision–recall curve value of 0.978 are achieved, surpassing many popular models such as Mask CNN, CNN, and YOLOv5.

Enhancing Detection of Pedestrians in Low-Light Conditions by Accentuating Gaussian–Sobel Edge Features from Depth Maps

Owing to the low detection accuracy of camera-based object detection models, various fusion techniques with Light Detection and Ranging (LiDAR) have been attempted. This has resulted in improved detection of objects that are difficult to detect due to partial occlusion by obstacles or unclear silhouettes. However, the detection performance remains limited in low-light environments where small pedestrians are located far from the sensor or pedestrians have difficult-to-estimate shapes. This study proposes an object detection model that employs a Gaussian–Sobel filter. This filter combines Gaussian blurring, which suppresses the effects of noise, and a Sobel mask, which accentuates object features, to effectively utilize depth maps generated by LiDAR for object detection. The model performs independent pedestrian detection using the real-time object detection model You Only Look Once v4, based on RGB images obtained using a camera and depth maps preprocessed by the Gaussian–Sobel filter, and estimates the optimal pedestrian location using non-maximum suppression. This enables accurate pedestrian detection while maintaining a high detection accuracy even in low-light or external-noise environments, where object features and contours are not well defined. The test evaluation results demonstrated that the proposed method achieved at least 1–7% higher average precision than the state-of-the-art models under various environments.

Diagnose Alzheimer’s disease and mild cognitive impairment using deep CascadeNet and handcrafted features from EEG signals

Alzheimer’s disease (AD) is the most prevalent clinically diagnosed neurodegenerative disorder. Early detection of mild cognitive impairment (MCI) is crucial for implementing effective interventions and potentially preventing further cognitive decline. Due to its efficiency, the electroencephalogram (EEG) is a promising tool for AD diagnosis. This paper proposes a computer-aided diagnostic model for identifying AD using EEG data analysis. The proposed approach comprises two key steps: signal processing and classification. Initially, the EEG signal is decomposed into sub-bands using the discrete wavelet transform (DWT), followed by windowing for data augmentation. Subsequently, an improved CascadeNet model is employed for feature extraction and classification from the windowed EEG data. CascadeNet’s architecture was specifically chosen for its ability to achieve high accuracy in AD and MCI detection, even with relatively small EEG datasets. Moreover, the paper explores the potential of deep Cascade learning for AD prediction. The effectiveness of the proposed strategy is evaluated using metrics such as F-measures, specificity, sensitivity, and overall detection accuracy. The developed model attains impressive accuracies of 98.84 % and 97.78 % on the Figshare and Brazilian datasets, respectively, significantly outperforming existing methods. To validate the model’s efficacy, it was applied to the Figshare and Brazilian datasets encompassing control (CO), MCI, and AD classes. The results confirm the proposed method as a valuable tool for identifying potential biomarkers aiding in the clinical diagnosis of AD. Notably, unlike previous methods limited to two-class identification, this technique effectively distinguishes between MCI and AD.

Deep Complex Gated Recurrent Networks-Based IoT Network Intrusion Detection Systems.

The explosive growth of the Internet of Things (IoT) has highlighted the urgent need for strong network security measures. The distinctive difficulties presented by Internet of Things (IoT) environments, such as the wide variety of devices, the intricacy of network traffic, and the requirement for real-time detection capabilities, are difficult for conventional intrusion detection systems (IDS) to adjust to. To address these issues, we propose DCGR_IoT, an innovative intrusion detection system (IDS) based on deep neural learning that is intended to protect bidirectional communication networks in the IoT environment. DCGR_IoT employs advanced techniques to enhance anomaly detection capabilities. Convolutional neural networks (CNN) are used for spatial feature extraction and superfluous data are filtered to improve computing efficiency. Furthermore, complex gated recurrent networks (CGRNs) are used for the temporal feature extraction module, which is utilized by DCGR_IoT. Furthermore, DCGR_IoT harnesses complex gated recurrent networks (CGRNs) to construct multidimensional feature subsets, enabling a more detailed spatial representation of network traffic and facilitating the extraction of critical features that are essential for intrusion detection. The effectiveness of the DCGR_IoT was proven through extensive evaluations of the UNSW-NB15, KDDCup99, and IoT-23 datasets, which resulted in a high detection accuracy of 99.2%. These results demonstrate the DCG potential of DCGR-IoT as an effective solution for defending IoT networks against sophisticated cyber-attacks.

Discharge Experiment and Structure Optimisation Simulation of Impulse Sound Source

The wave frequency and energy of traditional piezoelectric emission sources used in acoustic logging are limited, which results in an inadequate detection resolution for measuring small-scale geological formations. Additionally, the propagation of these waves in formations is prone to loss and noise interference, restricting detection to only a few tens of meters around the well. This paper investigates an impulse sound source, a new emission source that can effectively enhance the frequency range and wave energy of traditional sources by generating excitation waves through high-voltage discharges in a fluid-penetrated electrode structure. Firstly, a high-voltage circuit experimental system for the impulse sound source was constructed, and the discharge and response characteristics were experimentally analyzed. Then, four types of needle series electrode structure models were developed to investigate and compare the effects of different electrode structures on the impulse sound source, with the needle-ring electrode demonstrating superior performance. Finally, the needle-ring electrode structure was optimized to develop a ball-tipped needle-ring electrode, which is more suitable for acoustic logging. The results show that the electrode structure directly influences the discharge characteristics of the impulse sound source. After comparison and optimization, the final ball-tipped needle-ring electrode exhibited a broader frequency range—from zero to several hundred thousand Hz—while maintaining a high acoustic amplitude. It has the capability to detect geological areas beyond 100 m and is effective for evaluating micro-fractures and small fracture blocks near wells that require high detection accuracy. This is of significant importance in oil, gas, new energy, and other drilling fields.

High performance RGB-Thermal Video Object Detection via hybrid fusion with progressive interaction and temporal-modal difference

RGB-Thermal Video Object Detection (RGBT VOD) is to localize and classify the predefined objects in visible and thermal spectrum videos. The key issue in RGBT VOD lies in integrating multi-modal information effectively to improve detection performance. Current multi-modal fusion methods predominantly employ middle fusion strategies, but the inherent modal difference directly influences the effect of multi-modal fusion. Although the early fusion strategy reduces the modality gap in the middle stage of the network, achieving in-depth feature interaction between different modalities remains challenging. In this work, we propose a novel hybrid fusion network called PTMNet, which effectively combines the early fusion strategy with the progressive interaction and the middle fusion strategy with the temporal-modal difference, for high performance RGBT VOD. In particular, we take each modality as a master modality to achieve an early fusion with other modalities as auxiliary information by progressive interaction. Such a design not only alleviates the modality gap but facilitates middle fusion. The temporal-modal difference models temporal information through spatial offsets and utilizes feature erasure between modalities to motivate the network to focus on shared objects in both modalities. The hybrid fusion can achieve high detection accuracy only using three input frames, which makes our PTMNet achieve a high inference speed. Experimental results show that our approach achieves state-of-the-art performance on the VT-VOD50 dataset and also operates at over 70 FPS. The code will be freely released at https://github.com/tzz-ahu for academic purposes.

Iris Detection Using Deep Learning and OpenCV

The proposed system's efficacy is evaluated through extensive simulations and real- world testing, demonstrating high accuracy in accident detection and timely activation of the emergency response. This integration of deep learning with OpenCV and an automated alert system aims to reduce response times and enhance the chances of survival and support in the critical moments following a car accident. Keywords--- Deep Learning, OpenCV

Deep Learning-Based Image Classification and Segmentation on Digital Histopathology for Oral Squamous Cell Carcinoma: A Systematic Review and Meta-Analysis.

Artificial intelligence (AI)-based tools have shown promise in histopathology image analysis in improving the accuracy of oral squamous cell carcinoma (OSCC) detection with intent to reduce human error. This systematic review and meta-analysis evaluated deep learning (DL) models for OSCC detection on histopathology images by assessing common diagnostic performance evaluation metrics for AI-based medical image analysis studies. Diagnostic accuracy studies that used DL models for the analysis of histopathological images of OSCC compared to the reference standard were analyzed. Six databases (PubMed, Google Scholar, Scopus, Embase, ArXiv, and IEEE) were screened for publications without any time limitation. The QUADAS-2 tool was utilized to assess quality. The meta-analyses included only studies that reported true positives (TP), true negatives (TN), false positives (FP), and false negatives (FN) in their test sets. Of 1267 screened studies, 17 studies met the final inclusion criteria. DL methods such as image classification (n = 11) and segmentation (n = 3) were used, and some studies used combined methods (n = 3). On QUADAS-2 assessment, only three studies had a low risk of bias across all applicability domains. For segmentation studies, 0.97 was reported for accuracy, 0.97 for sensitivity, 0.98 for specificity, and 0.92 for Dice. For classification studies, accuracy was reported as 0.99, sensitivity 0.99, specificity 1.0, Dice 0.95, F1 score 0.98, and AUC 0.99. Meta-analysis showed pooled estimates of 0.98 sensitivity and 0.93 specificity. Application of AI-based classification and segmentation methods on image analysis represents a fundamental shift in digital pathology. DL approaches demonstrated significantly high accuracy for OSCC detection on histopathology, comparable to that of human experts in some studies. Although AI-based models cannot replace a well-trained pathologist, they can assist through improving the objectivity and repeatability of the diagnosis while reducing variability and human error as a consequence of pathologist burnout.

PHIGrader: Evaluating the effectiveness of Manifest file components in Android malware detection using Multi Criteria Decision Making techniques

The popularity of the Android operating system has itself become a reason for privacy concerns. To deal with such malware threats, researchers have proposed various detection approaches using static and dynamic features. Static analysis approaches are the most convenient for practical detection. However, several patterns of feature usage were found to be similar in the normal and malware datasets. Such high similarity in both datasets’ feature patterns motivates us to rank and select only the distinguishing set of features. Hence, in this study, we present a novel Android malware detection system, termed as PHIGrader for ranking and evaluating the efficiency of the three most commonly used static features, namely permissions, intents, and hardware components, when used for Android malware detection. To meet our goals, we individually rank the three feature types using frequency-based Multi-Criteria Decision Making (MCDM) techniques, namely TOPSIS and EDAS. Then, the system applies a novel detection algorithm to the rankings involving machine learning and deep learning classifiers to present the best set of features and feature type with higher detection accuracy as an output. The experimental results highlight that our proposed approach can effectively detect Android malware with 99.10% detection accuracy, achieved with the top 46 intents when ranked using TOPSIS, which is better than permissions, hardware components, or even the case where other popular MCDM techniques are used. Furthermore, our experiments demonstrate that the proposed system with frequency-based MCDM rankings is better than other statistical tests such as mutual information, Pearson correlation coefficient, and t-test. In addition, our proposed model outperforms various popularly used feature ranking methods such as Chi-square, Principal Component Analysis (PCA), Entropy-based Category Coverage Difference (ECCD), and other state-of-the-art Android malware detection techniques in terms of detection accuracy.

MpScope: Enabling multi-pipeline monitoring inside a switch

The core of programmable switches is the flexible data plane, composed of multiple programmable pipelines in existing programmable switches. These pipelines are isolated from each other and cannot share state and data. However, most of network monitoring systems ignore this condition and implicitly assume that the switch has only a single pipeline. This results in an inaccurate measurement and high communication overhead with the practical switch. To tackle this problem, we propose MpScope, a general multi-pipeline monitoring framework, which centers around the control plane, supporting accurate and efficient network monitoring. Specifically, MpScope combines the switch’s data plane and control plane to achieve comprehensive network monitoring of the whole switch scope. The control plane aggregates the statistical results from multiple pipelines and tunes the monitoring module residing in the different pipelines in the data plane dynamically. The data plane is responsible for real-time traffic measurement and statistic reports. Its behaviors can be adjusted periodically with the instructions from the control plane. Two typical monitoring applications, i.e., heavy hitter detection and distinct counting, are developed with MpScope to validate the effectiveness of multi-pipeline monitoring. Experiments show that MpScope significantly reduces communication overhead compared to the static threshold scheme while maintaining high detection accuracy over time.

Application of Generative Adversarial Networks in Semi-Annotated Defect Synthesis and Detection Under Limited Resources

Defect detection is a crucial technology that is extensively employed in the manufacturing industry to monitor and ensure the quality of output. Deep learning models have shown remarkable potential for defect detection. However, the success of these models heavily relies on voluminous training data. Collecting substantial amounts of defect data is challenging in practical settings, and the tedious process of pixel-level defect annotation further complicates the task. Among the common defects encountered in manufacturing, scratches are particularly significant. To address these challenges, this study proposes a two-phase generative adversarial network (GAN) approach for synthesizing defect images and generating semi-automatic pixel-wise labels for anomaly detection. The first phase primarily focuses on synthesizing images, while the second phase involves the pixel-wise labeling of the images. The synthesized paired images generated by the GANs serve as input to the semantic network. Notably, the proposed methodology requires only a few real defect samples for training and a small amount of annotated data, making it practical and computationally efficient for implementation in the manufacturing industry. Experimental results indicate the effectiveness of the proposed deep-learning solution in defect detection, specifically in identifying scratches on various textured and patterned surfaces. A notably high detection accuracy is achieved, validating the potential of the approach in real-world manufacturing scenarios.

基于高光谱和人工智能模型的白茶鲜叶农药残留检测

The detection of pesticide residues in white tea fresh leaves is an important step to ensure the quality safety of white tea finished products. Traditional detection methods are costly and inefficient to realize the demand for fast, low-cost, and accurate detection of pesticide residues in white tea fresh leaves. In this study, five types of white tea fresh leaf pesticide residue sample data were obtained using hyperspectral imaging technology for the high-frequency detected pesticides Glyphosate and Bifenthrin, and the SVM and 1D-CNN models were established to detect the samples after noise reduction processing and feature band screening methods. The study shows that the 1D-CNN model has better feature extraction ability, in which the SG-CARS-1D-CNN model has the highest detection accuracy, which is 94.62%, 95.12%, 94.35%, 94.95%, and 95.27% for the five type of species samples, respectively. This study provides pesticide residue detection for white tea fresh leaves based on the combination of hyperspectral data and an artificial intelligence model, which provides an intelligent, nondestructive, efficient, and high-precision pesticide residue detection model for white tea fresh leaves.

Distributed and Collaborative Lightweight Edge Federated Learning for IoT Zombie Devices Detection

The fast development of artificial intelligence and Internet of Things (IoT) technologies has enabled various applications of smart cities, e.g., smart monitoring and surveillance. However, vulnerabilities of IoT devices bring new threats to the security of smart cities. To identify ubiquitous IoT botnet attacks, a distributed and collaborative lightweight edge federated learning model for IoT zombie devices detection is proposed, named FIOT. To reduce computational complexity and enhance the adaptability to new attack environment at the network edge, FIOT is designed in a lightweight manner based on feature dimensionality reduction and transfer learning. Three IoT botnet datasets are used to validate the effectiveness of the proposed FIOT. Experimental results show that FIOT has an accuracy loss of less than 3% in terms of F1 value compared to the centralized learning, but the training time of FIOT is only 14.3% of that of centralized learning. While ensuring high detection accuracy, the number of parameters of FIOT is compressed to 37.58% of the comparison method.