Edge Devices Research Articles

CNN-LSTM for MFCC-based Speech Recognition on Smart Mirrors for Edge Computing Command

Smart mirrors are conventional mirrors that are augmented with embedded system capabilities to provide comfort and sophistication for users, including introducing the speech command function. However, existing research still applies the Google Speech API, which utilizes the cloud and provides sub-optimal processing time. Our research aim is to design speech recognition using Mel-frequency cepstral coefficients (MFCC) and convolutional neural network–long short-term memory (CNN-LSTM) to be applied to smart mirror edge devices for optimum processing time. Our first step was to download a synthetic speech recognition dataset consisting of waveform audio files (WAVs) from Kaggle, which included the utterances “left,” “right,” “yes,” “no,” “on,” and “off. ” We then designed speech recognition by involving Fourier transformation and low-pass filtering. We benchmark MFCC with linear predictive coding (LPC) because both are feature extraction methods on speech datasets. Then, we benchmarked CNN-LSTM with LSTM, simple recurrent neural network (RNN), and gated recurrent unit (GRU). Finally, we designed a smart mirror system complete with GUI and functions. The test results show that CNN-LSTM performs better than the three other methods with accuracy, precision, recall, and an f1-score of 0.92. The speech command with the best precision is "no," with a value of 0.940. Meanwhile, the command with the best recall is "off," with a value of 0.963. On the other hand, the speech command with the worst precision and recall is "other," with a value of 0.839. The contribution of this research is a smart mirror whose speech commands are carried out on the edge device with CNN-LSTM.

BCEdge: SLO-Aware DNN Inference Services With Adaptive Batch-Concurrent Scheduling on Edge Devices

BCEdge: SLO-Aware DNN Inference Services With Adaptive Batch-Concurrent Scheduling on Edge Devices

Efficient fruit disease diagnosis on resource-constrained agriculture devices

Climate change in recent years has caught the attention of researchers worldwide, with drastic effects on many economic sectors, especially agriculture and livestock. Due to global warming and declining water levels, food security is becoming an increasingly urgent concern. The situation is more challenging in underdeveloped nations such as Pakistan, where little attention has been paid to improving the processes to produce effective seeds and herbicides. The problem has been exacerbated due to financial constraints and a lack of technology to monitor crops. It has been observed that even a noncritical disease can ruin the crop if timely inspections aren’t performed. Although many solutions have been proposed recently, most are based on heavy mobile applications and cloud-based solutions, which results in considerable computational cost, power consumption, and latency. This work proposes a Deep Neural Networks (DNNs)-based light-weight solution for diagnosing and classifying apple crop diseases for farmers working in rural places (where the internet is not accessible). With the increasing need for efficient and robust Machine Learning (ML) applications on edge devices, such as mobile phones, Raspberry Pi, and Jetson Nano-based devices, there has been a demand for more efficient Deep Neural Networks (DNN) models. To meet this demand, various DNN-based models have been experimented with, including Basic CNN Architecture, AlexNet, and EfficientNet Lite to measure performance After a thorough examination and evaluation of each model’s performance, efficiency, and resource usage, the most optimal model was picked and developed applications for classifying apple crop disease. Using a transfer learning strategy on a specially developed EfficientNet DNN architecture, achieving 85% test accuracy.

A review on microstrip patch antenna for wireless communication systems at 3.5 GHz

This article presents a review of several microstrip patch antennas for 3.5 GHz wireless applications. Different substrate materials, FR-4 (loss), FR-4 Epoxy, Rogers RT/droid 5880, TLC-30, and Rogers RT/droid 5880 LZ, are used. In recent years, wireless antenna applications have increased, including biomedical appliances, internet of things (IoT) terminals, edge devices, radars, mobile phones, and many more. In this work, several articles were reviewed and investigated, and several microstrip patch antennas with a resonance frequency of 3.5 GHz were designed using different substrate materials and shapes. This article also discussed the geometric shapes of antennas, antenna properties, sizes of substrate materials, loss tangent, thickness, return loss, bandwidth, voltage standing wave ratio (VSWR), gain, efficiency, and directivity. Several software is used for design and simulation, including computer simulation technology (CST), high-frequency simulation frequency (HFSS), and advanced design system (ADS), FEKO, and MATLAB. The main goal of this paper is to talk about different wireless application papers that work in the S-band at a frequency of 3.5 GHz and have been published in various international journals and conferences.

Edge Device Fault Probability Based Intelligent Calculations for Fault Probability of Smart Systems

Edge Device Fault Probability Based Intelligent Calculations for Fault Probability of Smart Systems

Collaborative learning-based inter-dependent task dispatching and co-location in an integrated edge computing system

Recently, several edge deployment types, such as on-premise edge clusters, Unmanned Aerial Vehicles (UAV)-attached edge devices, telecommunication base stations installed with edge clusters, etc., are being deployed to enable faster response time for latency-sensitive tasks. One fundamental problem is where and how to offload and schedule multi-dependent tasks so as to minimize their collective execution time and to achieve high resource utilization. Existing approaches randomly dispatch tasks naively to available edge nodes without considering the resource demands of tasks, inter-dependencies of tasks and edge resource availability. These approaches can result in the longer waiting time for tasks due to insufficient resource availability or dependency support, as well as provider lock-in. Therefore, we present EdgeColla, which is based on the integration of edge resources running across multi-edge deployments. EdgeColla leverages learning techniques to intelligently dispatch multi-dependent tasks, and a variant bin-packing optimization method to co-locate these tasks firmly on available nodes to optimally utilize them. Extensive experiments on real-world datasets from Alibaba on task dependencies show that our approach can achieve optimal performance than the baseline schemes.

Clarity in Wireless Quantum Communication: Exploring Explainable AI for Signal Modulation Classification

The advent of artificial intelligence (AI) and edge computing technologies has ushered in a new era for automatic modulation classification (AMC), with a particular focus on deep learning (DL)-based approaches. These AI-based AMC systems, deployed at edge devices, have shown remarkable advancements in recognizing and classifying a wide range of wireless signals. However, a significant challenge in this domain lies in the lack of adequate explanations for the models employed, which, in turn, hampers model accuracy and training efficiency. Consequently, the practical applications and further enhancements of these models are constrained. To address this limitation, researchers have started to explore interpretable methods for the technical analysis of DL-based AMC systems. In this paper, we delve into the subject, leveraging recent developments in interpretable methods, to provide a comprehensive review of applicable techniques and to outline the existing research challenges in the field of interpretable AMC. Moreover, we introduce a novel interpretable AI-based AMC framework designed to augment the interpretability of AMC outcomes. This framework achieves this by elucidating the contribution of wireless signal features to the training process of DL networks. In addition, we explore the potential impact of quantum computing on AMC systems. Quantum computing, with its ability to process complex computations at unprecedented speeds, offers a promising avenue for enhancing the efficiency and accuracy of AMC. By integrating quantum algorithms with AI-based AMC frameworks, it is possible to significantly improve the performance of wireless signal classification, especially in environments with high signal variability and noise. Our experimental results demonstrate that the proposed approach possesses the capacity to unravel the classification mechanisms hidden within opaque auto-modulated classification networks. Furthermore, it exhibits the potential to facilitate the training of networks on edge devices with reduced energy consumption and enhanced classification accuracy, thereby addressing critical challenges in the field of wireless signal classification. Additionally, the integration of quantum computing techniques holds the promise of further advancing the capabilities of AMC systems, paving the way for more robust and efficient wireless communication technologies.

Deep Learning Approach to Detect Fake Video on Raspberry PI

Abstract: This project proposes a deep learning-based approach for real-time detection of fake videos on a resource-constrained device, specifically the Raspberry Pi. The solution combines the power of computer vision and recurrent neural networks to discern manipulated content from authentic videos effectively. The methodology involves using a pre-trained ResNeXt model for feature extraction, capturing spatial information from each video frame. These features are then fed into a Long Short-Term Memory (LSTM) network, allowing the model to understand and exploit temporal dependencies within the sequence of frames. The LSTM network learns patterns and nuances indicative of authentic or manipulated video content. The training process involves a carefully curated dataset containing both real and fake videos. The model is fine-tuned to optimize its performance, and metrics such as accuracy, precision, recall, and F1 score are employed for evaluation. To accommodate the constraints of the Raspberry Pi, the model is further optimized through techniques such as quantization, ensuring a balance between model size and inference accuracy. The final model is deployed on the Raspberry Pi, with a user-friendly interface capturing video frames in real-time. The system provides instantaneous feedback, indicating whether the observed video content is genuine or manipulated. This project contributes to the growing field of deepfake detection while addressing the challenges of implementing sophisticated models on edge devices. The combination of ResNeXt and LSTM offers a robust solution for discerning manipulated videos, making it suitable for real-world applications where computational resources are limited.

The impact of semiconductor technology on edge computing

Semiconductor technology has profoundly shaped the landscape of edge computing by enhancing energy efficiency, processing power, and device miniaturization. Continued advancements in semiconductor tech promise further integration of AI, improved edge-to-cloud synchronization, and heightened security, fueling growth and innovation in edge computing. This synergy is evident as semiconductor tech enables versatile edge device deployment, while edge computing applications like real-time analytics and machine learning drive innovation, resulting in purpose-built chips for edge computing applications. Semiconductor tech and edge computing are symbiotic, with one driving the others evolution, forming an inseparable partnership in the realm of computing. This paper describes the role of semiconductor technology in driving edge computing technology in terms of energy, performance, and miniaturization, and incorporates data that predicts future directions for semiconductor-based technologies in edge computing.

Flame and smoke detection using Kafka on edge devices

This paper presents object detection methods to accurately identify the sources of flame and smoke in vast circumstances. Aerial drones collected the data, analyzed the recognition outputs in real time on an edge device, and then transferred them to the back-end for data processing and warnings using Kafka. To detect flame and smoke occurrences, the models were compared using various convolutional neural networks (CNN). Several factors considered include streaming speed, accuracy, portability, efficiency, and power consumption on edge devices. This work conducted training comparisons of YOLOV4, YOLOV5, YOLOV7, YOLOV8, and Faster RCNN. The inference performance was then evaluated on an edge computing device. The findings showed an accuracy of 0.91 and 0.87, while maintaining a processing speed of roughly 1 frame per second on the Nvidia Jetson NX without acceleration.

Cybersecurity Implications of Edge Computing in Data Engineering

Edge computing, as an extension of cloud computing, brings computation and data storage closer to the data source. This shift offers significant advantages in terms of latency reduction, bandwidth optimization, and real-time processing capabilities. By minimizing the distance that data needs to travel, edge computing enhances the performance of applications that require rapid data processing and immediate response times. This is particularly beneficial for Internet of Things (IoT) devices, autonomous vehicles, smart grids, and other applications that demand low-latency interactions. However, the decentralization inherent in edge computing introduces a unique set of cybersecurity challenges that are distinct from those faced in traditional centralized cloud environments. The distributed architecture of edge computing creates numerous points of vulnerability, each of which can be exploited by cyber attackers. Additionally, edge devices often operate with limited computational resources and power, which complicates the implementation of robust security measures. This paper explores the cybersecurity implications of edge computing in the context of data engineering. It begins with a comprehensive review of existing literature to establish the current understanding of edge computing security issues. The review identifies the primary vulnerabilities associated with edge computing, including those related to distributed architecture, data transmission, and resource constraints. Following the literature review, the paper delves into specific security threats that edge computing environments face. These include man-in-the-middle (MitM) attacks, distributed denial-of-service (DDoS) attacks, malware and ransomware, and insider threats. Each threat is analyzed to understand its potential impact on edge computing systems and the data they process.

Development and evaluation of a machine vision and deep learning-based smart sprayer system for site-specific weed management in row crops: An edge computing approach

Traditional weed management often involves blanket herbicide spraying, resulting in substantial herbicide wastage, environmental concerns, and herbicide resistant issues. Smart spraying systems utilizing robotics and sensors technologies can minimize herbicide usage and provide a sustainable solution for site-specific weed management. A machine vision-based spraying system was designed and developed for weed identification and precise spray application onto the target weeds. The sprayer platform utilizes a deep learning YOLOv4 model to accurately recognize multiple weed species, facilitating targeted spray application. The platform is equipped with an FLIR RGB camera for real-time image acquisition and Nvidia Jetson AGX Orin edge device for deploying weed detection deep-learning model. The GPIO pins of Nvidia Jetson were utilized to activate relay, providing precise on/off control over the TeeJet solenoid valves for spot spraying. Both indoor and field experiments were conducted to evaluate and compare the performance of vision-based sprayer system for weed identification and precise spraying onto the target weeds. In the indoor experiment, the sprayer system showed the average effective spraying rate of 93.33 %, with the precision of 100 % and recall of 92.8 %. Conversely, the field experiment resulted in a slightly lower average effective spraying rate of 90.6 %, while maintaining a precision of 95.5 % and a recall of 89.47 %. The reduced accuracy of the spraying system in field experiment was due to varying outdoor conditions such as lighting, shadows, and wind velocity. Overall, the result of this study demonstrates the spraying system's potential for targeted herbicide application onto the grid cells containing weeds, effectively reducing herbicide usage and overall weed management costs.

Streamlining plant disease diagnosis with convolutional neural networks and edge devices

Streamlining plant disease diagnosis with convolutional neural networks and edge devices

Multi source data security protection of smart grid based on edge computing

In order to cope with the continuous development of information technology, the data volume of edge devices in the power supply network is increasing rapidly, and the higher requirements for real-time data processing and transmission bandwidth are put forward, the author proposes the research on security protection of multi-source data in smart grid based on edge computing. Established a distribution network safety risk map, calculated and analyzed safety risks through potential functions, and obtained node potential values and key node data; The intelligent grid multi-source heterogeneous data monitoring model based on edge computing can realize the timely perception and real-time response of distribution network faults, shorten the outage time, and improve the power supply reliability and user satisfaction of the distribution network. The results indicate that: Taking a distribution network line in a certain area as an example, the highest potential value is node 9, followed by nodes 10 and 3. This indicates that in order to ensure the reliable operation of the distribution network and avoid power outages, it is necessary to improve the repair capacity, the proportion of old equipment, and the power supply radius exceeding the standard. ConclusionThe effectiveness of potential function in risk assessment of edge computing is verified, which can provide theoretical guidance for distribution network fault diagnosis.

The effect of task processing management on energy consumption at the edge of Internet of things network with using reinforcement learning method

In this paper, the task processing approach at the edge of the Internet of Things (IoT) network in improving energy consumption was investigated. First, the subject of maximizing the use of energy is defined as the main problem and then based on the state of assignment of task processing from end devices at the edge of the network to smart gateways and local servers at the edge using possible low bandwidth and low battery consumption, the problem is divided into two subject which are obtained independently. After modeling by recombining these two sub-problems, the subject of maximizing the utility gain and battery life of the end devices is solved despite the limitations of processing time and energy resources.Considering that the environment of the problem and how the terminal devices are connected with the edge devices are unknown, a repetitive reinforcement learning algorithm has been used to create an optimal solution to maximize the operational benefit of the energy consumed in edge processing. The results achieved from the simulation show the increase in network load and processing overhead with the increase in the number of active devices. The suggested method, while increasing the edge processing speed, decreases the delay and maximizes the performance gain of the IOT edge system compared to the central cloud system. It should be noted that when the number of active end devices dramatically increases, the energy consumption and bandwidth at the network edge improves, and energy consumption at the network edge decreases lonely by 12.5%.

YOLO-FMDI: A Lightweight YOLOv8 Focusing on a Multi-Scale Feature Diffusion Interaction Neck for Tomato Pest and Disease Detection

At the present stage, the field of detecting vegetable pests and diseases is in dire need of the integration of computer vision technologies. However, the deployment of efficient and lightweight object-detection models on edge devices in vegetable cultivation environments is a key issue. To address the limitations of current target-detection models, we propose a novel lightweight object-detection model based on YOLOv8n while maintaining high accuracy. In this paper, (1) we propose a new neck structure, Focus Multi-scale Feature Diffusion Interaction (FMDI), and inject it into the YOLOv8n architecture, which performs multi-scale fusion across hierarchical features and improves the accuracy of pest target detection. (2) We propose a new efficient Multi-core Focused Network (MFN) for extracting features of different scales and capturing local contextual information, which optimizes the processing power of feature information. (3) We incorporate the novel and efficient Universal Inverted Bottleneck (UIB) block to replace the original bottleneck block, which effectively simplifies the structure of the block and achieves the lightweight model. Finally, the performance of YOLO-FMDI is evaluated through a large number of ablation and comparison experiments. Notably, compared with the original YOLOv8n, our model reduces the parameters, GFLOPs, and model size by 18.2%, 6.1%, and 15.9%, respectively, improving the mean average precision (mAP50) by 1.2%. These findings emphasize the excellent performance of our proposed model for tomato pest and disease detection, which provides a lightweight and high-precision solution for vegetable cultivation applications.

Foreign Object Debris Detection on Wireless Electric Vehicle Charging Pad Using Machine Learning Approach

Foreign object debris (FOD) includes any unwanted and unintentional material lying on the charging lane or parking lots, posing a risk to the wireless charging system, the vehicle, or the people inside. FOD in an Electric Vehicle (EV) wireless charging system can cause problems, including decreased charging efficiency, safety risks, charging system damage, communication issues, and health risks. To address this problem, this paper proposes the deep learning object detection network approach of using YOLOv4 (You Only Look Once), which is a single-shot detector. Additionally, for real-time implementation, YOLOv4-Tiny is suggested, which is a compressed version of YOLOv4 designed for devices with low computational power. YOLOv4-Tiny enables faster inferences and facilitates the deployment of FOD detectors on edge devices. The algorithm is trained using the FOD dataset, consisting of images of common debris on runways or taxiways. Furthermore, utilizing the concept of transfer learning, the last few layers of the pre-trained YOLOv4 model are modified using the COCO (Common Objects in Context) dataset to transfer features to the new network and retrain the model on the FOD dataset. The results obtained using this YOLOv4 model yielded a precision rate of 99.05%, while the results from YOLOv4-Tiny achieved a precision rate of 97.74%, with an average inference time of 150 ms under the ambient light and weather conditions.

Policy Compression for Intelligent Continuous Control on Low-Power Edge Devices.

Interest in deploying deep reinforcement learning (DRL) models on low-power edge devices, such as Autonomous Mobile Robots (AMRs) and Internet of Things (IoT) devices, has seen a significant rise due to the potential of performing real-time inference by eliminating the latency and reliability issues incurred from wireless communication and the privacy benefits of processing data locally. Deploying such energy-intensive models on power-constrained devices is not always feasible, however, which has led to the development of model compression techniques that can reduce the size and computational complexity of DRL policies. Policy distillation, the most popular of these methods, can be used to first lower the number of network parameters by transferring the behavior of a large teacher network to a smaller student model before deploying these students at the edge. This works well with deterministic policies that operate using discrete actions. However, many real-world tasks that are power constrained, such as in the field of robotics, are formulated using continuous action spaces, which are not supported. In this work, we improve the policy distillation method to support the compression of DRL models designed to solve these continuous control tasks, with an emphasis on maintaining the stochastic nature of continuous DRL algorithms. Experiments show that our methods can be used effectively to compress such policies up to 750% while maintaining or even exceeding their teacher's performance by up to 41% in solving two popular continuous control tasks.

Energy-efficiency optimization for heterogeneous computing-assisted NOMA-MEC edge AI tasks

Edge artificial intelligence (AI) is an emerging paradigm that leverages edge computing to pave the last-mile delivery of AI. To satisfy the increasing demand for high-performance computing and low latency of edge service, heterogeneous computing accelerators, especially Neural Processor Units (NPUs), are widely deployed on edge nodes. However, the edge AI heterogeneous computing system encounters the challenge of energy constraints. In this paper, we investigate the energy efficiency (EE) optimization problem in heterogeneous computing-assisted non-orthogonal multiple access mobile edge computing (NOMA-MEC) network model. Edge devices are configured with CPU-NPU heterogeneous computing processors to improve the AI task processing, and NOMA is applied to edge task offloading to enable massive connectivity. The system-centric energy efficiency maximization problem is studied. We formulate a system-centric energy efficiency maximization problem. To solve this problem, we decouple the original problem into two subproblems, i.e., the optimization problems of heterogeneous computing workload splitting and task offloading. A heuristic algorithm and binary search algorithm are proposed to optimize NOMA user pairing and task offloading delay, respectively. Furthermore, we propose a multi-objective alternative iterative algorithm to optimize system energy efficiency. Numerical results show that the proposed scheme can significantly improve the energy efficiency of heterogeneous computing-assisted NOMA-MEC networks compared to the existing baselines.

Edge Computing Architecture Sensor-based Flood Monitoring System: Design and Implementation

The purpose of this research is to develop and execute a system for monitoring floods using sensors and edge computing architecture. The goal is to make flood detection and prediction more accurate and faster. The growing frequency and severity of flood disasters in different parts of the world has prompted the necessity for a better system to track these events. The primary goal of this study is to design a system that can reduce network load and latency by processing sensor data locally at edge devices before sending it to the cloud. To detect and anticipate flood events, the research method incorporates several environmental sensors that measure things like soil moisture, water level, and rainfall. These readings are subsequently processed by edge nodes using machine learning algorithms. Compared to more conventional methods that depend only on cloud computing, the results demonstrate that the system can deliver quicker and more accurate predictions. Results showed a detection and prediction accuracy of 98.95% for floods. Edge computing also succeeded in drastically cutting down on bandwidth consumption and communication latency. This research concludes that edge computing architecture based on sensors can effectively monitor floods and has excellent potential for use in many different areas prone to flooding. Improving the prediction algorithm and investigating its potential integration with a more thorough early warning system should be the focus of future research.