Edge Computing Platform Research Articles

Optimizing the Agricultural Internet of Things (IoT) with Edge Computing and Low-Altitude Platform Stations

Using low-altitude platform stations (LAPSs) in the agricultural Internet of Things (IoT) enables the efficient and precise monitoring of vast and hard-to-reach areas, thereby enhancing crop management. By integrating edge computing servers into LAPSs, data can be processed directly at the edge in real time, significantly reducing latency and dependency on remote cloud servers. Motivated by these advancements, this paper explores the application of LAPSs and edge computing in the agricultural IoT. First, we introduce an LAPS-aided edge computing architecture for the agricultural IoT, in which each task is segmented into several interdependent subtasks for processing. Next, we formulate a total task processing delay minimization problem, taking into account constraints related to task dependency and priority, as well as equipment energy consumption. Then, by treating the task dependencies as directed acyclic graphs, a heuristic task processing algorithm with priority selection is developed to solve the formulated problem. Finally, the numerical results show that the proposed edge computing scheme outperforms state-of-the-art works and the local computing scheme in terms of the total task processing delay.

Automated image-based fire detection and alarm system using edge computing and cloud-based platform

To tackle the increasing wildfire challenges, this paper presents an automated image-based fire detection and alarm system utilizing edge computing and a cloud-based platform, specifically designed for urban building fire detection. The system captures both RGB and infrared images from thermal cameras and employs existing computer vision techniques to detect fire characteristics such as flames and smoke. By integrating edge computing, the system minimizes latency to enhance the accuracy of fire detection and alarm activation. The cloud platform supports extensive data storage, analysis, and remote monitoring, which can ensure data accessibility and scalable data management. The proposed system descriptions include a detailed system architecture design, data collection, and the selection and application of detection algorithms that leverage both RGB and thermal image data for fire detection. Using the campus building and surrounding risk-prone areas as a testbed, the proposed system demonstrated desired fire detection capabilities and a robust solution to quickly identify and respond to fire incidents within the urban area. The proposed system functionalities can be scaled and adapted to other fire risk-prone areas.

A Lightweight Model of Underwater Object Detection Based on YOLOv8n for an Edge Computing Platform

The visual signal object detection technology of deep learning, as a high-precision perception technology, can be adopted in various image analysis applications, and it has important application prospects in the utilization and protection of marine biological resources. While the marine environment is generally far from cities where the rich computing power in cities cannot be utilized, deploying models on mobile edge devices is an efficient solution. However, because of computing resource limitations on edge devices, the workload of performing deep learning-based computationally intensive object detection on mobile edge devices is often insufficient in meeting high-precision and low-latency requirements. To address the problem of insufficient computing resources, this paper proposes a lightweight process based on a neural structure search and knowledge distillation using deep learning YOLOv8 as the baseline model. Firstly, the neural structure search algorithm was used to compress the YOLOv8 model and reduce its computational complexity. Secondly, a new knowledge distillation architecture was designed, which distills the detection head output layer and NECK feature layer to compensate for the accuracy loss caused by model reduction. When compared to YOLOv8n, the computational complexity of the lightweight model optimized in this study (in terms of floating point operations (FLOPs)) was 7.4 Gflops, which indicated a reduction of 1.3 Gflops. The multiply–accumulate operations (MACs) stood at 2.72 G, thereby illustrating a decrease of 32%; this saw an increase in the AP50, AP75, and mAP by 2.0%, 3.0%, and 1.9%, respectively. Finally, this paper designed an edge computing service architecture, and it deployed the model on the Jetson Xavier NX platform through TensorRT.

Deep Learning and Self-Powered Sensors Enabled Edge Computing Platform for Predicting Microclimate at Urban Blocks

With the development of urban construction and the continuous innovation of science and technology, the progress of smart cities is imperative. Urban local climate prediction is very important in the construction of smart city, but the existing climate prediction methods cannot accurately predict the urban local climate. Therefore, this work provides accurate climate reference for people’s travel and outdoor activities, thus improving the accuracy of local climate prediction in the city. Firstly, the current situation of climate prediction is discussed, and the main indexes of climate prediction and the basic algorithm of climate index prediction are determined. Then, the deep Learning algorithm is introduced and applied to the edge computing (EC) platform, and self-powered sensors are integrated to predict the microclimate index of urban blocks. Using self-powered sensors can achieve zero contact work and improve the research efficiency. Finally, experiments are carried out in two sample cities, and the prediction results are compared with the actual detection results to verify the performance of the prediction model. The results show that the error value of the microclimate index predicted by the deep learning model under EC platform is -6~5, and the climate index with the largest error is the air age. The air age error of city A is about ± 5 s, and that of city B is -6 ~ 4 s. Practice has proved that the deep learning technology has achieved ideal results in the microclimate index prediction of urban blocks on the EC platform and is practical. This work provides technical support for the prediction of urban block climate indicators and helps to improve the accuracy of climate prediction.

Stateful Serverless Application Placement in MEC With Function and State Dependencies

Serverless computing is emerging as an enabling technology for elastic and low-cost AI applications in the edge of core networks. It allows AI developers to decompose a complex training and time-sensitive inference task into multiple functions with dependency, and upload the task to a Multi-access Edge Computing platform (MEC) for execution. Serverless computing adopts a popular design principle: the disaggregation of storage and computation, making the functions ‘stateless’. However, most AI applications are ‘stateful’ and rely on an external storage service to manage their states (ephemeral data). This will incur a prohibitively long delay for delay-sensitive AI applications if external services storing the states are far from the serverless functions. Motivated by this critical issue, in this paper we investigate a fundamental problem in serverless computing – the stateful serverless application placement problem, for which, we first propose an efficient heuristic algorithm, and then devise an approximation algorithm with a provable approximation ratio for one of its special cases. We also consider the online version of the problem, and develop an online learning-driven algorithm with a bounded regret. The crux of the online algorithm is the adoption of the multi-armed bandits technique for dynamic admissions of inference requests, under the uncertainty of both data volumes of requests and network delays. We finally evaluate the performance of the proposed algorithms through experimental simulations. Simulation results show that the proposed algorithms outperform their counterparts, reducing at least <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$32\%$</tex-math></inline-formula> in the total cost and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$27\%$</tex-math></inline-formula> of the average delay.

Using N-BEATS ensembles to predict automated guided vehicle deviation

AbstractA novel AGV (Automated Guided Vehicle) control architecture has recently been proposed where the AGV is controlled remotely by a virtual Programmable Logic Controller (PLC), which is deployed on a Multi-access Edge Computing (MEC) platform and connected to the AGV via a radio link in a 5G network. In this scenario, we leverage advanced deep learning techniques based on ensembles of N-BEATS (state-of-the-art in time-series forecasting) to build predictive models that can anticipate the deviation of the AGV’s trajectory even when network perturbations appear. Therefore, corrective maneuvers, such as stopping the AGV, can be performed in advance to avoid potentially harmful situations. The main contribution of this work is an innovative application of the N-BEATS architecture for AGV deviation prediction using sequence-to-sequence modeling. This novel approach allows for a flexible adaptation of the forecast horizon to the AGV operator’s current needs, without the need for model retraining or sacrificing performance. As a second contribution, we extend the N-BEATS architecture to incorporate relevant information from exogenous variables alongside endogenous variables. This joint consideration enables more accurate predictions and enhances the model’s overall performance. The proposed solution was thoroughly evaluated through realistic scenarios in a real factory environment with 5G connectivity and compared against main representatives of deep learning architectures (LSTM), machine learning techniques (Random Forest), and statistical methods (ARIMA) for time-series forecasting. We demonstrate that the deviation of AGVs can be effectively detected by using ensembles of our extended N-BEATS architecture that clearly outperform the other methods. Finally, a careful analysis of a real-time deployment of our solution was conducted, including retraining scenarios that could be triggered by the appearance of data drift problems.

An Improved Lightweight Real-Time Detection Algorithm Based on the Edge Computing Platform for UAV Images

Unmanned aerial vehicle (UAV) image detection algorithms are critical in performing military countermeasures and disaster search and rescue. The state-of-the-art object detection algorithm known as you only look once (YOLO) is widely used for detecting UAV images. However, it faces challenges such as high floating-point operations (FLOPs), redundant parameters, slow inference speed, and poor performance in detecting small objects. To address the above issues, an improved, lightweight, real-time detection algorithm was proposed based on the edge computing platform for UAV images. In the presented method, MobileNetV3 was used as the YOLOv5 backbone network to reduce the numbers of parameters and FLOPs. To enhance the feature extraction ability of MobileNetV3, the efficient channel attention (ECA) attention mechanism was introduced into MobileNetV3. Furthermore, in order to improve the detection ability for small objects, an extra prediction head was introduced into the neck structure, and two kinds of neck structures with different parameter scales were designed to meet the requirements of different embedded devices. Finally, the FocalEIoU loss function was introduced into YOLOv5 to accelerate bounding box regression and improve the localization accuracy of the algorithm. To validate the performance of the proposed improved algorithm, we compared our algorithm with other algorithms in the VisDrone-Det2021 dataset. The results showed that compared with YOLOv5s, MELF-YOLOv5-S achieved a 51.4% reduction in the number of parameters and a 38.6% decrease in the number of FLOPs. MELF-YOLOv5-L had 87.4% and 47.4% fewer parameters and FLOPs, respectively, and achieved higher detection accuracy than YOLOv5l.

Rate-Splitting and Common Message Decoding in Hybrid Cloud/Mobile Edge Computing Networks

This paper proposes, and evaluates the benefits of, a hybrid central cloud (CC) and mobile edge computing (MEC) platform, especially introduced to balance the network resources for joint communication and computation. The transmission is further empowered by splitting the users’ messages into private and common parts, to mitigate the interference within the CC and MEC platforms. While several power-hungry, computationally-limited unmanned aerial vehicles (UAVs) are deployed at the cell-edge to boost the CC connectivity and relieve part of its computation burden, the CC connects to the base-stations via capacity-limited fronthauls. The paper then considers the problem of maximizing the weighted sum-rate subject to fronthaul and computation capacity, achievable rates, power, delay, and data-split constraints. Thereby determining the beamforming vectors associated with the private and common messages, the computation allocations, and the data-split factors. Such intricate non-convex optimization problem is tackled using an iterative algorithm that relies on well-chosen discrete relaxation, successive convex approximation, and fractional programming, and can be compellingly implemented in a distributed fashion. The simulations illustrate the proposed algorithm’s capabilities for empowering joint communication and computation, and highlight the pronounced role of rate-splitting and common message decoding in alleviating large-scale interference in hybrid CC/MEC networks.

A Self-Adaptive Trajectory Optimization Algorithm Using Fuzzy Logic for Mobile Edge Computing System Assisted by Unmanned Aerial Vehicle

The advancement of the Internet of Things (IoT) and the availability of wide cloud services have led to the horizon of edge computing paradigm which demands for processing the data at the edge of the network. The development of 5G technology has led to the increased usage of IoT-based devices and the generation of a large volume of data followed by increased data traffic, which is difficult to process by the mobile edge computing (MEC) platform. The latest inventions related to unmanned aerial vehicles (UAVs) helps to assist and replace the edge servers used for MEC. In the present work, the objective is to develop self-adaptive trajectory optimization algorithm (STO) which is a multi-objective optimization algorithm used to solve the vital objectives associated with the above scenario of a UAV-assisted MEC system. The objectives identified are minimizing the energy consumed by the MEC and minimizing the process emergency indicator, where the process emergency indicator implies the urgency level of a particular process. Finding the optimal values for these conflicting objectives will help to further efficiently apply UAV for MEC systems. A self-adaptive multi-objective differential evolution-based trajectory optimization algorithm (STO) is proposed, where a pool of trial vector generation strategies is extended. The strategies and the crossover rate associated with a differential evolution (DE) algorithm are self-adapted using fuzzy systems to improve the population diversity. The experimentation is planned to be conducted on hundreds of IoT device instances considered to be fixed on the ground level and to evaluate the performance of the proposed algorithm for a single unmanned aerial vehicle-assisted mobile edge computing system.

Hierarchical federated learning with mobile edge computing in the Internet of Vehicles

Federated Learning is a distributed machine learning framework, which can be used in the Internet of Vehicles to train deep learning models without directly accessing the original data of mobile edge vehicle nodes. ECS can access massive data, but it has the characteristics of high latency and high communication overhead. However, mobile edge computing (MEC) platform can directly and efficiently communicate with mobile edge vehicle nodes. Combining the advantages of the two, a three-layer federated learning system of edge car network edge server cloud server is used. This system is supported by the HierFedProx algorithm and aggregates the model output of the edge car to the edge server to improve the model learning efficiency and reduce the global communication frequency. The experimental results show that the system can reduce the training time and improve the accuracy of the model compared with the federated learning without introducing the edge server.

Federated learning for 5G-enabled infrastructure inspection with UAVs

Electricity infrastructures include assets that require frequent maintenance, as they are exposed into heavy use, in order to produce energy that satisfies customer demands. Such maintenance is currently performed by specialized personnel that is scaffolding to spot damages or malfunctioning equipment. Scaffolding is time-consuming and incurs accident risks. To tackle this challenges, grid operators are gradually using Unmanned Aerial Vehicles (UAVs). UAV trajectories are observed by a centralized operation center engineers for identifying electrical assets. Moreover, asset identification can be further automated through the use of Artificial Intelligence (AI) models. However, centralized training of AI models with UAV images may cause inspection delays when the network is overloaded and requires Cloud environments with enough processing power for model training on the operation center. This imposes privacy concerns as sensitive data is stored and processed externally from the infrastructure facility. This article proposes a federated learning method for UAV-based inspection that leverages a Multi-access Edge Computing platform installed in edge nodes to train UAV data and improve the overall inspection autonomy. The method is applied for the inspection of the Public Power Corporation’s Innovation Hub. Experiments are performed with the proposed method as well as with a centralized AI inspection method and demonstrate the federated learning benefits in reliability, AI model processing time and privacy conservation.

Edge Computational Offloading for Corrosion Inspection in Industry 4.0

The fifth generation (5G) mobile networks are a game changer for the 4th industrial revolution (Industry 4.0). The difficulty of developing vertical applications for mobile networks is however still challenging. This article proposes a new paradigm solution to significantly facilitate deploying network applications (NetApp) closer to the end user and improving the user experience, facilitated by 5G. This new paradigm leverages flexible service onboarding and composition end to end, and computation offloading to a Multi-access Edge Computing (MEC) platform to meet the demanding requirements of the NetApps. A functional prototype is developed and empirical comparison is conducted to validate and evaluate the performance of a corrosion detection NetApp using artificial intelligence (AI) in three different scenarios where the application is hosted. The prototype runs this NetApp over an Unmanned Aerial Vehicle (UAV) connected to a 5G network testbed, and the results offer insights into this performance improvement and elaborates how opting for this NetApp solution can help the industry. In this case, the use of this NetApp can achieve a 55.47 percent reduction in computation time and improve corrosion identification processing time by 25.5 percent.

Dynamic Online Trajectory Planning for a UAV-Enabled Data Collection System

Due to the maneuverability and flexibility of unmanned aerial vehicles (UAVs), the UAV-enabled data collection systems for wireless sensor networks (WSN) have received widespread attention. However, the state-of-the-art UAV trajectory designs mainly focus on static environments, which are not applicable in the practical scenarios considered in this work, e.g., mobile nodes, decommissioning of existing nodes, and new emergency nodes. This article proposes a two-level deep reinforcement learning (DRL) framework to solve this challenge. The first-level deep neural network (DNN) is applied to model the dynamic changing environment. In the second level, we employ a deep Q-learning network to plan a trajectory online according to the environment features from the first level DNN. Besides, online trajectory planning is performed by a low-power UAV edge computing platform. To enable online planning on the power-constraint UAV edge-computing platform, all networks adopt a lightweight low-complexity optimization design. According to simulation results, the proposed system achieves higher data acquisition success rates when compared to existing state-of-the-art methods. We also perform field tests on the proposed UAV edge computing platform, which also demonstrates high data acquisition performance.

Implementation of EDGE Computing Platform in Feeder Terminal Unit for Smart Applications in Distribution Networks with Distributed Renewable Energies

Under the plan of net-zero carbon emissions in 2050, the high penetration of distributed renewable energies in distribution networks will cause the operation of more complicated distribution networks. The development of edge computing platforms will help the operator to monitor and compute the system status timely and locally, and it can ensure the security operation of the system. In this paper, a novel EDGE computing platform that is implemented by a graphics processing unit in the existing feeder terminal unit (FTU) is proposed for smart applications in distribution networks with distributed renewable energies and loads. This platform makes timely forecasts of the feeder status for the next seven days in accordance with historical weather, sun, and loading data. The forecast solver uses the machine learning long short-term memory (LSTM) method. Thereafter, the power calculation analyzers transform feeder topology into the circuit model for transient-state, steady-state, and symmetrical component analyses. An important-factor explainer parses the LSTM model into the concise value of each historical datum. All information transports to remote devices via the internet for the real-time monitor feature. The software stack of the EDGE platform consists of the database archive file system, time-series forecast solver, power flow analyzers, important-factor explainer, and message queuing telemetry transport (MQTT) protocol communication. All open-source software packages, such as SQLite, LSTM, ngspyce, Shapley Additive Explanations, and Paho-MQTT, form the aforementioned function. The developed EDGE forecast and power flow computing platform are helpful for achieving FTU becoming an Internet of Things component for smart operation in active distribution networks.

Construction of Edge Computing Platform Using 3D LiDAR and Camera Heterogeneous Sensing Fusion for Front Obstacle Recognition and Distance Measurement System

This research aims to utilise heterogeneous sensor fusion using 3D Light Detection and Ranging (LiDAR) and cameras, combined with an object recognition system and a ranging system, to construct an edge computing platform such that a vehicle equipped with the platform can perform computations offline in real time. This work comprises two main sections: the first is heterogeneous fusion, and the second is obstacle recognition and ranging detection. To achieve heterogeneous sensor fusion, 3D–3D point matching was used to find rigid body transformation between two sensors and finally project the LiDAR 3D point cloud image onto the 2D image. For object recognition, YOLOv4-Tiny was used as the detection network. A lightweight network architecture and high computational speed could be effectively used on edge computing hardware with limited performance. Further, by drawing the bounding box, we could detect the point cloud within the bounding box to estimate the distance to the obstacle. For detecting distance, we conducted experiments in two ways: ‘minimum point in box’ and ‘median point in box’ and compared the results. With heterogeneous sensor fusion, object recognition and the ranging system, detecting the category and distance of obstacles ahead of the vehicle was possible in real time. Furthermore, integrating the edge computing platform architecture enabled moving the entire system offline, making it an independent system that returns results in real time. Finally, a dynamic test was conducted on a road. The experiment showed that the detection speed of YOLOv4-Tiny in the dynamic test was higher than 60 FPS, and the accuracy rate surpassed 70%. Furthermore, the distance detection error of the 3D LiDAR was less than 3 cm, which is sufficiently accurate to be applied to complex environments on roads.

Cost-Efficient Server Configuration and Placement for Mobile Edge Computing

Computing resource configuration and site selection of edge servers (ESs) are two critical steps to build up a mobile edge computing (MEC) platform. In this paper, the joint optimization problem of configuration and placement for ES in the MEC environment is investigated. First, we treat each ES as an M/G/m queueing model, and establish mathematical models to characterize the MEC environment, such that the performance and operational expenditures (OPEX) of the system can be calculated analytically. Then, we design a two-stage method and develop a series of algorithms based on bisection algorithm and genetic algorithm (GA) to obtain the optimal configuration scheme and sub-optimal placement scheme (including the deployment quantity) of ESs, with the goal of minimizing OPEX while maintaining system performance at a predetermined level. Finally, we conduct experiments based on a real base station dataset provided by Shanghai Telecom to show the effectiveness of the proposed algorithms. To the best of our knowledge, this work is the first research of the joint optimization problem of configuration and placement for ES in the MEC environment, where the main objective is to increase the cost efficiency.

Flow Assignment and Processing on a Distributed Edge Computing Platform

The evolution of telecommunication networks toward the fifth generation of mobile services (5G), along with the increasing presence of cloud-native applications, and the development of Cloud and Mobile Edge Computing (MEC) paradigms, have opened up new opportunities for the monitoring and management of logistics and transportation. We address the case of distributed streaming platforms with multiple message brokers to develop an optimisation model for the real-time assignment and load balancing of event streaming generated data traffic among Edge Computing facilities. The performance indicator function to be optimised is derived by adopting queuing models with different granularity (packet- and flow-level) that are suitably combined. A specific use case concerning a logistics application is considered and numerical results are provided to show the effectiveness of the optimisation procedure, also in comparison to a “static” assignment proportional to the processing speed of the brokers.

Activities and Results of Field Trials—Network Edge Computing Platform

Activities and Results of Field Trials—Network Edge Computing Platform

Optimal Task Allocation Algorithm Based on Queueing Theory for Future Internet Application in Mobile Edge Computing Platform.

For 5G and future Internet, in this paper, we propose a task allocation method for future Internet application to reduce the total latency in a mobile edge computing (MEC) platform with three types of servers: a dedicated MEC server, a shared MEC server, and a cloud server. For this platform, we first calculate the delay between sending a task and receiving a response for the dedicated MEC server, shared MEC server, and cloud server by considering the processing time and transmission delay. Here, the transmission delay for the shared MEC server is derived using queueing theory. Then, we formulate an optimization problem for task allocation to minimize the total latency for all tasks. By solving this optimization problem, tasks can be allocated to the MEC servers and cloud server appropriately. In addition, we propose a heuristic algorithm to obtain the approximate optimal solution in a shorter time. This heuristic algorithm consists of four algorithms: a main algorithm and three additional algorithms. In this algorithm, tasks are divided into two groups, and task allocation is executed for each group. We compare the performance of our proposed heuristic algorithm with the solution obtained by three other methods and investigate the effectiveness of our algorithm. Numerical examples are used to demonstrate the effectiveness of our proposed heuristic algorithm. From some results, we observe that our proposed heuristic algorithm can perform task allocation in a short time and can effectively reduce the total latency in a short time. We conclude that our proposed heuristic algorithm is effective for task allocation in a MEC platform with multiple types of MEC servers.

LNNet: Lightweight Nested Network for motion deblurring

Motion deblurring methods based on convolutional neural networks (CNN) have recently demonstrated their advantages over conventional methods. However, repetitions of scaling or slicing operations of these methods on the input images inevitably lead to spatial information loss. Meanwhile, some recent methods based on complex models inevitably bring a large model size and huge computing cost. It is still challenging to balance the deblurring performance and the cost. To this end, we propose a lightweight nested network (LNNet) for the motion-deblurring task. Our LNNet leverages several simple yet efficient sub-networks to process motion deblurring features at each stage. We design a nested connection, which is conducive to the model size reduction when connecting sub-networks so as to reuse deblurring information and facilitate deblurring information diversity. Meanwhile, we introduce the feature-fusion module to improve deblurring performance further. We perform extensive experiments on a workstation platform and an embedded mobile edge computing (MEC) platform to evaluate our LNNet as well as other existing methods. Extensive experimental results demonstrate that our LNNet achieves superior deblurring performance than state-of-the-art methods with a small model size within a short running time. Moreover, experimental results also show that our model is quite suitable for other embedded devices.