IEEE Journal of Selected Areas in Sensors最新文献

{"title":"Generative AI-Native Edge Sensors: Collaborative Data Reconstruction via Lightweight Federated Learning","authors":"Jiaming Pei;Minxi Feng;Pounambal Muthukumar;Parimala Venkata Krishna;Minghui Dai","doi":"10.1109/JSAS.2026.3680380","DOIUrl":"https://doi.org/10.1109/JSAS.2026.3680380","url":null,"abstract":"The reliability of time-series data streams in distributed Internet of Things (IoT) sensor networks is frequently compromised by sensor faults and unstable communication, leading to critical data gaps. Traditional centralized approaches for data reconstruction introduce high latency and data privacy risks, failing to meet the requirements of resource-constrained edge environments. To address this, we propose a novel <italic>lightweight federated generative learning</i> framework, termed <italic>FedGen-Sensor</i>, which enables collaborative, privacy-preserving, and resilient data reconstruction directly on heterogeneous edge sensor nodes. Our framework employs a lightweight variational autoencoder optimized for temporal data and integrates a deep <italic>hardware–software co-design</i> approach. Specifically, we utilize an aggressive compression strategy combining <italic>INT8 posttraining quantization</i> and <italic>sparse update transmission</i> to drastically reduce the model’s footprint. Validated on a physical testbed comprising ESP32 microcontrollers and Raspberry Pi servers, FedGen-Sensor demonstrates superior performance: it achieves data reconstruction fidelity comparable to the standard FP32 FedAvg baseline (<inline-formula><tex-math>$text{0.086}$</tex-math></inline-formula> versus 0.082), while simultaneously reducing communication overhead by <inline-formula><tex-math>$text{5.5}times$</tex-math></inline-formula> and cutting the average edge inference latency by <inline-formula><tex-math>$text{2.5}times$</tex-math></inline-formula>. This work validates the feasibility of embedding sophisticated generative intelligence directly onto ultra-low-power edge sensors, setting a new paradigm for autonomous and resilient IoT data services.","PeriodicalId":100622,"journal":{"name":"IEEE Journal of Selected Areas in Sensors","volume":"3 ","pages":"180-190"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11473293","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147796101","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Room-Temperature Low-Bias LPG Sensing via Barrier Modulation in WO3/Polyaniline Heterojunction Diodes","authors":"Seungmin Oh;Chihoon Kim","doi":"10.1109/JSAS.2026.3681240","DOIUrl":"https://doi.org/10.1109/JSAS.2026.3681240","url":null,"abstract":"We demonstrate a room-temperature liquefied petroleum gas (LPG) sensor based on a tungsten oxide /polyaniline heterojunction diode fabricated on a tungsten substrate. Tungsten oxide nanostructures were grown by hydrothermal synthesis, and polyaniline was deposited by cell-voltage-controlled electrodeposition, where the deposition time (10–25 s) tuned junction coverage. The optimized device (20 s) exhibited pronounced suppression of forward current upon LPG exposure at 300 K. To ensure interpretable quantification under variable ambient backgrounds, an air–N₂–LPG sequence was employed and a purge-referenced normalized response, <italic>S</i>(%), was defined using the N₂-purged baseline at a fixed operating voltage (<italic>V</i>* = 0.60 V). A baseline shift of 27.87% during air-to-N₂ switching confirms strong sensitivity to background constituents, motivating the purge-referenced protocol to decouple LPG-induced modulation from ambient-dependent baseline variations. At <italic>V</i>* = 0.60 V, the response increased monotonically over 100–900 ppm LPG and reached a representative value of 71.61% at 500 ppm. Within 300–700 ppm, a quasi-linear operating window was identified with a segment sensitivity of 0.0481%·ppm−1 (<inline-formula><tex-math>${bm{R}}_{mathrm{adj}}^2$</tex-math></inline-formula> = 0.966). The measured response/recovery times (<italic>t</i>₉₀ ≈ 20 s; Tr₉₀ ≈ 7 s) represent system-level upper bounds under the present gas delivery configuration. Repeatability and device-to-device reproducibility of <italic>S</i>(%) were 5.33% (<italic>n</i> = 5) and 9.70% (<italic>n</i> = 3) relative standard deviation (RSD), respectively. These results support low-bias diode readout as a practical route to room-temperature LPG sensing and provide a purge-referenced framework for interpreting diode-type gas signals under variable backgrounds.","PeriodicalId":100622,"journal":{"name":"IEEE Journal of Selected Areas in Sensors","volume":"3 ","pages":"199-206"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11480967","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147828883","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Triple-Notch-Band Reconfigurable Ultrawideband Antenna Based on Liquid Metal for Sensor Network Applications","authors":"Xiangrong Peng;Shuang Ma;Zhandong Li;Ruifang Wang;Junjun Ding","doi":"10.1109/JSAS.2026.3688196","DOIUrl":"https://doi.org/10.1109/JSAS.2026.3688196","url":null,"abstract":"This article presents a tri-notched reconfigurable ultrawideband (UWB) antenna utilizing liquid metal (LM) microfluidic technology. The design integrates a polydimethylsiloxane layer with embedded microchannels onto a half-circular monopole radiator. Three etched slots generate band-notches for WiMAX, WLAN, and X-band systems. A coplanar waveguide feed with truncated ground planes ensures UWB operation. By selectively injecting LM into the microchannels, four operational modes are enabled: conventional UWB, single-, dual-, and triple-notched states. The fabricated prototype exhibits an impedance bandwidth from 2.77 to 15 GHz, with triple-notched bands at 3.33–3.67, 5.03–5.89, and 6.97–7.81 GHz. Unlike active-component-based approaches, the proposed antenna features a passive reconfiguration strategy where addressable LM microchannels are mechanically controlled, thereby simplifying the system architecture and enhancing reliability. The antenna, with its compact size and effective interference suppression, is a promising solution for spectrum-agile data transmission in dense wireless sensor networks and Internet of Things systems.","PeriodicalId":100622,"journal":{"name":"IEEE Journal of Selected Areas in Sensors","volume":"3 ","pages":"207-217"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11501728","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148171582","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Digital Twin-Assisted Large AI Task-Aware Edge Offloading and Resource Allocation for Low-Altitude Wireless Sensor Networks","authors":"Bintao Hu;Hengyan Liu;Haotong Cao;Shijing Yuan;Xiaoli Chu;Shugong Xu","doi":"10.1109/JSAS.2026.3679846","DOIUrl":"https://doi.org/10.1109/JSAS.2026.3679846","url":null,"abstract":"Emerging vehicle-to-everything (V2X) applications, especially for smart transportation, require high-accuracy sensing and low-latency communications and computation; however, existing V2X architectures that deal with sensing, communication, and computation separately are ill-equipped to meet these coupled requirements. In this article, we propose a framework that combines integrated sensing and communications and digital twins with low-altitude edge intelligence for V2X, and formulate an optimization problem to minimize the total service delay (i.e., the transmission delay plus the task computation delay) of all vehicular users by jointly optimizing task offloading decisions, communication and computation resource allocation, and the association between unmanned aerial vehicles (UAVs) and vehicular users. To solve this problem under fast varying mobility and high-dimensional coupled constraints, we propose a task-aware multiagent resource allocation optimization algorithm, which enables scalable cooperative decision-making, including UAV-vehicular user association, and communication and computation resource allocation. Simulation results show substantial reductions in total service delay over the traditional deep reinforcement learning benchmark, especially in dense, dynamic low-altitude edge intelligent V2X scenarios.","PeriodicalId":100622,"journal":{"name":"IEEE Journal of Selected Areas in Sensors","volume":"3 ","pages":"159-170"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11466417","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147737086","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"FBG-Based Wearable Sensors for Healthcare 4.0: A System-Level Review With AI, Generative AI, and XAI Integration","authors":"Pradeep Kumar;Ruchi Antil;Eugene Su;Chao-Ching Ho","doi":"10.1109/JSAS.2026.3702624","DOIUrl":"https://doi.org/10.1109/JSAS.2026.3702624","url":null,"abstract":"Wearable sensor (WS) technology has advanced rapidly in recent years, driven by the increasing demand for continuous physiological monitoring. Among various sensing approaches, fiber Bragg grating (FBG) sensors have emerged as strong candidates for wearable systems due to their lightweight nature, electromagnetic immunity, multiplexing capability, and high sensitivity to strain, temperature, and motion. This review presents recent developments in FBG-based WSs within the framework of Healthcare 4.0, where sensing, computation, and communication are integrated into a unified cyber-physical architecture. A structured literature review methodology is adopted to ensure comprehensive coverage and reproducibility. The fundamentals of FBG sensing and their implementation in wearable platforms are first outlined. Representative applications—including magnetic resonance imaging-compatible cardiorespiratory monitoring, carotid pulse assessment, skin temperature mapping, gait analysis, and joint posture detection—are examined to highlight the performance advantages of FBG systems. A comparative analysis is further provided to distinguish FBG wearables from conventional electronic sensors in terms of sensitivity, robustness, and compatibility with clinical environments. Key challenges, such as interrogator size, power consumption, and placement-dependent sensitivity variations, are critically discussed. Emerging solutions are explored through the integration of artificial intelligence, explainable artificial intelligence (XAI) for model interpretability, generative artificial intelligence for synthetic data augmentation and improved generalization, and singular value decomposition-based signal decoupling and dimensionality reduction for efficient edge processing. In addition, advancements in compact photonic interrogators and fog–cloud data processing frameworks are highlighted. Therefore, FBG-based wearable technologies demonstrate strong potential for enabling reliable, high-precision, and intelligent health monitoring in next-generation Healthcare 4.0 systems.","PeriodicalId":100622,"journal":{"name":"IEEE Journal of Selected Areas in Sensors","volume":"3 ","pages":"230-252"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11561813","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148479447","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Low-Latency Human-in-the-Loop Robotic Teleoperation With Control–Media Separation Using MQTT and WebRTC","authors":"Shih-Chieh Sun;Yun-Cheng Tsai;Han-Wen Kuo;Shen-Chi Chen","doi":"10.1109/JSAS.2026.3729555","DOIUrl":"https://doi.org/10.1109/JSAS.2026.3729555","url":null,"abstract":"Low-latency teleoperation is a core requirement for Artificial Intelligence of Things (AIoT) robotics, yet achieving responsive closed-loop manipulation while streaming live video and running on-the-fly vision inference remains challenging under geographically distributed networks. This article presents a modular AIoT teleoperation framework for human-in-the-loop robotic manipulation that separates the control and media pipelines using message queuing telemetry transport (MQTT) for real-time command delivery and web real-time communication (WebRTC) (via LiveKit) for low-latency video streaming. Lightweight visual perception is enabled through a You Only Look Once version 11-nano (YOLOv11-nano) proof-of-concept assistive visual overlay that provides real-time object-level annotations to a cross-platform Flutter-based mobile interface during teleoperation. On the actuation side, control commands are transmitted to an ESP8266-based communication node and executed by an Arduino Mega2560 controller to coordinate multiaxis robotic arm motion. Backend services are deployed on cloud infrastructure to support scalable access across regions. The perception model is trained on a 32 GB workstation and achieves millisecond-level inference performance (<italic>1.5 ms inference-only</i>, <italic><inline-formula><tex-math>$approx$</tex-math></inline-formula>11.5 ms end-to-end per frame</i>). The trained model is further evaluated on lightweight 8 GB RAM nodes under VPN-enabled cross-border conditions, including routes to Hong Kong, Japan, and Belgium. Experimental results demonstrate actuator response latencies as low as <italic>0.2 s in local networks</i> and below <italic>0.7 s in cross-border scenarios</i>, confirming that the proposed framework maintains real-time responsiveness even on resource-constrained hardware. Beyond baseline system benchmarking, we enhance journal-grade reproducibility by reporting multirun training stability under controlled illumination variability using precision–recall curves and normalized confusion matrices. Furthermore, the framework supports <italic>nonintrusive extensibility</i>, exemplified by an optional infrared gesture-control module that can be integrated without altering the perception–control decoupling or the underlying actuation interface. Overall, the work provides a standards-aligned reference design for human-in-the-loop AIoT teleoperation, while demonstrating how lightweight object-level perception can be integrated as an assistive visual overlay.","PeriodicalId":100622,"journal":{"name":"IEEE Journal of Selected Areas in Sensors","volume":"3 ","pages":"425-438"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11676824","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149133806","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Fast and Robust Stationary Crowd Counting With Commodity WiFi","authors":"Mert Torun;Alireza Parsay;Yasamin Mostofi","doi":"10.1109/JSAS.2025.3648956","DOIUrl":"https://doi.org/10.1109/JSAS.2025.3648956","url":null,"abstract":"This article introduces a novel method for estimating the size of seated crowds with commodity WiFi signals, by leveraging natural body fidgeting behaviors as a passive sensing cue. Departing from prior binary fidget representations, our approach leverages the bandwidth of the received signal as a finer-grained and robust indicator of crowd counts. More specifically, we propose a mathematical model that relates the probability distribution function (PDF) of the signal bandwidth to the crowd size, using a principled derivation based on the PDF of an individual's fidget-induced bandwidth. To characterize the individual fidgeting PDF, we use publicly available online videos, each of a seated individual, from which we extract body motion profiles using vision techniques, followed by a speed-to-bandwidth conversion inspired by Carson's Rule from analog FM radio design. Finally, to enhance robustness in real-world deployments where unrelated motions may occur nearby, we further introduce an anomaly detection module that filters out nonfidget movements. We validate our system through 42 experiments across two indoor environments with crowd sizes up to and including 13 people, achieving a mean absolute error of 1.04 and a normalized mean square error of 0.15, with an average convergence time of 51 s, significantly reducing the convergence time as compared to the state of the art. Additional simulation results demonstrate scalability to larger crowd sizes. Overall, our results showcase that the proposed pipeline enables fast, robust, and highly accurate counting of seated crowds.","PeriodicalId":100622,"journal":{"name":"IEEE Journal of Selected Areas in Sensors","volume":"3 ","pages":"57-68"},"PeriodicalIF":0.0,"publicationDate":"2025-12-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11316410","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146082099","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"MedLiteNet: Lightweight Hybrid Medical Image Segmentation Model for the Medical Internet of Things (MIoT)","authors":"Pengyang Yu;Haoquan Wang;Gerard Marks;Tahar Kechadi;Laurence T. Yang;Sahraoui Dhelim;Nyothiri Aung","doi":"10.1109/JSAS.2025.3648685","DOIUrl":"https://doi.org/10.1109/JSAS.2025.3648685","url":null,"abstract":"Accurate skin-lesion segmentation remains a key technical challenge for computer-aided diagnosis of skin cancer. Convolutional neural networks, while effective, are constrained by limited receptive fields and thus struggle to model long-range dependencies. Vision Transformers capture global context, yet their quadratic complexity and large parameter budgets hinder use on the small-sample medical datasets common in dermatology. We introduce the MedLiteNet, a lightweight CNN–Transformer hybrid tailored for dermoscopic segmentation that achieves high precision through hierarchical feature extraction and multiscale context aggregation. The encoder stacks depthwise mobile inverted bottleneck blocks to curb computation, inserts a bottleneck-level cross-scale token-mixing unit to exchange information between resolutions, and embeds a boundary-aware self-attention module to sharpen lesion contours. On the ISIC 2018 benchmark, a single MedLiteNet model attains a Dice score of <inline-formula><tex-math>$0.897 pm 0.010$</tex-math></inline-formula> and an IoU of <inline-formula><tex-math>$0.821 pm 0.015$</tex-math></inline-formula> with fewer than <inline-formula><tex-math>$3.3,mathrm{M}$</tex-math></inline-formula> parameters. A performance-weighted ensemble of three complementary variants raises accuracy to <inline-formula><tex-math>$0.904 pm 0.012$</tex-math></inline-formula> Dice and <inline-formula><tex-math>$0.830 pm 0.018$</tex-math></inline-formula> IoU while keeping the total parameter count below <inline-formula><tex-math>$10,mathrm{M}$</tex-math></inline-formula>—over 90% smaller than Vision-Transformer backbones. Qualitative results confirm superiority on irregular borders, low-contrast regions and multiscale lesions, indicating MedLiteNet’s suitability for real-time, resource-aware computer-aided dermatology.","PeriodicalId":100622,"journal":{"name":"IEEE Journal of Selected Areas in Sensors","volume":"3 ","pages":"78-89"},"PeriodicalIF":0.0,"publicationDate":"2025-12-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11316249","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146175808","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Stress Classification From ECG Signals Using Vision Transformer","authors":"Zeeshan Ahmad;Naimul Khan","doi":"10.1109/JSAS.2025.3648684","DOIUrl":"https://doi.org/10.1109/JSAS.2025.3648684","url":null,"abstract":"Vision transformers have shown tremendous success in numerous computer vision applications; however, they have not been exploited for stress assessment using physiological signals such as electrocardiogram (ECG). In order to get the maximum benefit from the vision transformer for multilevel stress assessment, in this article, we transform the raw ECG data into 2-D spectrograms using short-time Fourier transform. These spectrograms are divided into patches for feeding to the transformer encoder. We also perform experiments with 1-D convolutional neural network (CNN) and ResNet-18 (CNN model). We perform leave-one-subject-out cross validation (LOSOCV) experiments on wearable stress and affect detection (WESAD) and Ryerson Multimedia Lab (RML) dataset. One of the biggest challenges of LOSOCV-based experiments is to tackle the problem of intersubject variability. In this research, we address the issue of intersubject variability and show our success using 2-D spectrograms and the attention mechanism of transformer. Experiments show that vision transformer handles the effect of intersubject variability much better than CNN-based models and beats all previous state-of-the-art methods by a considerable margin. Moreover, our method is end-to-end, does not require handcrafted features, and can learn robust representations. The proposed method achieved 71.01% and 76.7% accuracies with RML dataset and WESAD dataset, respectively, for three class classification and 88.3% for binary classification on WESAD.","PeriodicalId":100622,"journal":{"name":"IEEE Journal of Selected Areas in Sensors","volume":"3 ","pages":"47-56"},"PeriodicalIF":0.0,"publicationDate":"2025-12-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11316267","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146026552","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Acoustic-Based Fault Detection for Robotic Arms","authors":"Cihun-Siyong Gong;Chih-Hui Simon Su;Kuo-Wei Chao;Cheng-Yen Lin;Yu-Hua Chen","doi":"10.1109/JSAS.2025.3638975","DOIUrl":"https://doi.org/10.1109/JSAS.2025.3638975","url":null,"abstract":"With the rise of Industry 4.0 and smart manufacturing, the integration of the Industrial Internet of Things and autonomous systems has become crucial for improving productivity and operational efficiency. For robotic arms, the core of automation, fault diagnosis is essential for ensuring production line reliability and sustainable operations. Traditional diagnostic methods, particularly those based on visual sensing, face challenges such as high deployment costs, insufficient contactlessness, and enormous data volumes. Acoustic-based fault detection has become a promising technique for monitoring the condition and ensuring the reliability of robotic arms. It is especially important in industrial automation, where noninvasive and real-time diagnostics are essential. This article systematically reviews recent state-of-the-art studies on acoustic signal acquisition, preprocessing, feature extraction, and classification for robotic arm fault diagnosis. Combining advanced signal processing techniques, such as time–frequency analysis, wavelet transforms, and empirical mode decomposition, with machine learning and deep learning models, has significantly improved fault detection in terms of accuracy and robustness. This review provides a detailed discussion of current research achievements and future directions, highlighting hybrid models. By integrating acoustic-sensing-based fault diagnosis with large-scale artificial intelligence models deployed in edge computing environments, users or companies can not only leverage neural network architectures to identify abnormal sound patterns amid complex background noise accurately but also directly address key challenges such as resource constraints, real-time performance, and energy efficiency at the edge.","PeriodicalId":100622,"journal":{"name":"IEEE Journal of Selected Areas in Sensors","volume":"3 ","pages":"1-22"},"PeriodicalIF":0.0,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11271558","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145778207","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}