2023 IEEE International Conference on Edge Computing and Communications (EDGE)最新文献

iEDGE 2023 Symposium Chairs Message iEDGE 2023研讨会主席致辞

2023 IEEE International Conference on Edge Computing and Communications (EDGE) Pub Date : 2023-07-01 DOI: 10.1109/edge60047.2023.00011

引用次数: 0

Performance Analysis of Real-Time Video Surveillance Application Leveraging Edge and Cloud 基于边缘和云的实时视频监控应用性能分析

2023 IEEE International Conference on Edge Computing and Communications (EDGE) Pub Date : 2023-07-01 DOI: 10.1109/EDGE60047.2023.00039

Priyal Thakkar, Ashish Singh Patel, Gaurav Shukla, A. Kherani, B. Lall

{"title":"Performance Analysis of Real-Time Video Surveillance Application Leveraging Edge and Cloud","authors":"Priyal Thakkar, Ashish Singh Patel, Gaurav Shukla, A. Kherani, B. Lall","doi":"10.1109/EDGE60047.2023.00039","DOIUrl":"https://doi.org/10.1109/EDGE60047.2023.00039","url":null,"abstract":"With the advent of the Edge and Cloud server in the 5G system, an application needs to be designed to have multiple components, where a part of it (Data Intensive Component (DIC)) is executed on the Edge server while the other part (Computation Intensive Component (CIC)) is executed on the Cloud server. Such deployment of the applications’ components into the Edge and Cloud server opens up opportunities for managing the Edge, Cloud, and network resources. In this work, performance aspects of the simultaneous deployment of a video surveillance application on the Edge and Cloud server are explored. Furthermore, application placement approach at the Edge and Cloud server based on the service time requirement of an application is demonstrated. In addition, an adaptive data transmission mechanism at the Edge server is presented, where the components that run at the Edge server use a scaled-down version of the video based on Initial Analysis, reducing the bandwidth consumption between the Edge server and UE. As a use-case, a surveillance application to identify traffic violations (jumping signal) is deployed. The performance of the simultaneous deployment of video surveillance application (Edge-cloud approach) is evaluated by demonstrating bandwidth preserved and end-to-end bandwidth requirement in comparison with the different Cloud only approaches. To simulate actual deployments, the surveillance application is deployed on an ETSI-compliant 5G MEC testbed with the Edge and Cloud server.","PeriodicalId":369407,"journal":{"name":"2023 IEEE International Conference on Edge Computing and Communications (EDGE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126086842","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

MECBench: A Framework for Benchmarking Multi-Access Edge Computing Platforms MECBench:多访问边缘计算平台的基准测试框架

2023 IEEE International Conference on Edge Computing and Communications (EDGE) Pub Date : 2023-07-01 DOI: 10.1109/EDGE60047.2023.00024

Omar Naman, H. Qadi, M. Karsten, S. Al-Kiswany

引用次数: 0

eDashA: Edge-based Dash Cam Video Analytics eDashA:基于边缘的行车记录仪视频分析

2023 IEEE International Conference on Edge Computing and Communications (EDGE) Pub Date : 2023-07-01 DOI: 10.1109/EDGE60047.2023.00040

Jayden King, Young Choon Lee

引用次数: 0

Big Data Analytics from the Rich Cloud to the Frugal Edge 从富云到节俭边缘的大数据分析

2023 IEEE International Conference on Edge Computing and Communications (EDGE) Pub Date : 2023-07-01 DOI: 10.1109/EDGE60047.2023.00054

Feras M. Awaysheh, Riccardo Tommasini, Ahmed Awad

{"title":"Big Data Analytics from the Rich Cloud to the Frugal Edge","authors":"Feras M. Awaysheh, Riccardo Tommasini, Ahmed Awad","doi":"10.1109/EDGE60047.2023.00054","DOIUrl":"https://doi.org/10.1109/EDGE60047.2023.00054","url":null,"abstract":"Modern systems and applications generate and consume an enormous amount of data from different sources, including mobile edge computing and IoT systems. Our ability to locate and analyze these massive amounts of data will shape the future, building next-generation Big Data Analytics (BDA) and artificial intelligence systems in critical domains. Traditionally, big data materialize in a centralized repository (e.g., the cloud) for running sophisticated analytics using decent computation. Nevertheless, many modern applications and critical domains require low-latency data analysis with the right decision at the right time standard for building trust. With the advent of edge computing, that traditional deployment model shifted closer to the data sources at the network’s edge. Such a shift was motivated by minimized latency, increased uptime, and enhanced efficiencies. This paper studies the BDA building blocks, analyzes the deployment requirements for edge-based BDA QoS, and drafts future trends. It also discusses critical open issues and further research directions for the next step of edge-based BDA.","PeriodicalId":369407,"journal":{"name":"2023 IEEE International Conference on Edge Computing and Communications (EDGE)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114094354","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

EDGE 2023 Committees EDGE 2023委员会

2023 IEEE International Conference on Edge Computing and Communications (EDGE) Pub Date : 2023-07-01 DOI: 10.1109/edge60047.2023.00012

引用次数: 0

Improved Knowledge Distillation for Crowd Counting on IoT Devices 物联网设备上人群计数的改进知识蒸馏

2023 IEEE International Conference on Edge Computing and Communications (EDGE) Pub Date : 2023-07-01 DOI: 10.1109/EDGE60047.2023.00041

Zuo Huang, R. Sinnott

{"title":"Improved Knowledge Distillation for Crowd Counting on IoT Devices","authors":"Zuo Huang, R. Sinnott","doi":"10.1109/EDGE60047.2023.00041","DOIUrl":"https://doi.org/10.1109/EDGE60047.2023.00041","url":null,"abstract":"Manual crowd counting for real-world problems is impossible or results in wildly inaccurate estimations. Deep learning is one area that has been applied to address this issue. Crowd counting is a computationally intensive task. Therefore, many crowd counting models employ large-scale deep convolutional neural networks (CNN) to achieve higher accuracy. However, these are typically at the cost of performance and inference speed. This makes such approaches difficult to apply in real-world settings, e.g., on Internet-of-Things (IoT) devices. To tackle this problem, one method is to compress models using pruning and quantization or use of lightweight model backbones. However, such methods often result in a significant loss of accuracy. To address this, some studies have explored knowledge distillation methods to extract useful information from large state-of-the-art (teacher) models to guide/train smaller (student) models. However, knowledge distillation methods suffer from the problem of information loss caused by hint-transformers. Furthermore, teacher models may have a negative impact on student models. In this work, we propose a method based on knowledge distillation that uses self-transformed hints and loss functions that ignore outliers to tackle real-world and challenging crowd counting tasks. Based on our approach, we achieve a MAE of 77.24 and a MSE of 276.17 using the JHU-CROWD++ [1] test set. This is comparable to state-of-the-art deep crowd counting models, but at a fraction of the original model size and complexity, thus making the solution suitable for IoT devices. The source code is available at https://github.com/huangzuo/effcc_distilled.","PeriodicalId":369407,"journal":{"name":"2023 IEEE International Conference on Edge Computing and Communications (EDGE)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127562908","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Probabilistic Error Reasoning on IoT Edge Devices 物联网边缘设备的概率错误推理

2023 IEEE International Conference on Edge Computing and Communications (EDGE) Pub Date : 2023-07-01 DOI: 10.1109/EDGE60047.2023.00031

Charles Qing Cao, Yunhe Feng

{"title":"Probabilistic Error Reasoning on IoT Edge Devices","authors":"Charles Qing Cao, Yunhe Feng","doi":"10.1109/EDGE60047.2023.00031","DOIUrl":"https://doi.org/10.1109/EDGE60047.2023.00031","url":null,"abstract":"Existing IoT applications are increasingly using sensors to collect real-world measurements to make decisions. Such measurements are inherently limited by the accuracy of ADC devices, hence, introduce noise and errors. However, application developers often choose scalar data to represent sensor readings without regard to the errors associated with such data. This gives the illusion that the measurements are error-free, leading to error accumulation and false positive results. In this paper, we present a new type of programming abstraction for modeling errors and performing inference tasks in measurements of the physical world on resource-constrained IoT devices, which we call approximation variables (approxes). Using approxes does not require any changes to the programming language itself. Instead, it is designed as a suite of library functions that can be integrated directly into existing programming practices. We demonstrate how to use it in C programs. This framework makes decisions about the distributions of parameter values and inherently supports sampling and hypothesis testing to evaluate the accuracy of computational results. We compare its use to traditional programming practices and show how the library can be used to reveal uncertainty to the user, so that it can handle errors, reduce false positive results, and lead to better decision-making. These benefits make approxes a compelling and promising solution for programming with noisy sensor measurements for modern IoT applications.","PeriodicalId":369407,"journal":{"name":"2023 IEEE International Conference on Edge Computing and Communications (EDGE)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125933180","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Realising the Power of Edge Intelligence: Addressing the Challenges in AI and tinyML Applications for Edge Computing 实现边缘智能的力量:解决边缘计算的人工智能和微型ml应用中的挑战

2023 IEEE International Conference on Edge Computing and Communications (EDGE) Pub Date : 2023-07-01 DOI: 10.1109/EDGE60047.2023.00056

Michael Gibbs, E. Kanjo

{"title":"Realising the Power of Edge Intelligence: Addressing the Challenges in AI and tinyML Applications for Edge Computing","authors":"Michael Gibbs, E. Kanjo","doi":"10.1109/EDGE60047.2023.00056","DOIUrl":"https://doi.org/10.1109/EDGE60047.2023.00056","url":null,"abstract":"The edge computing paradigm has become increasingly popular due to its benefits over cloud computing, particularly in the context of AI and IoT applications. Its harmonising with AI to form Edge intelligence (EI) has opened up possible application areas for further development. Tiny machine learning (tinyML) is a specific focus within EI that targets machine learning algorithms deployed to constrained edge devices such as microcontrollers. However, despite the potential advantages of EI and tinyML, there are several challenges that researchers often overlook, especially when deploying on microcontrollers. These challenges include programming language choice, lack of support for development boards, neglect of preprocessing, choice of sensors, and insufficient labelled data. This paper assesses these previously unaddressed challenges, highlights their issues with a particular focus on microcontroller deployment, and offers potential solutions. By addressing these challenges, researchers can design more effective and efficient tinyML systems, pushing the boundaries of edge AI faster than before.","PeriodicalId":369407,"journal":{"name":"2023 IEEE International Conference on Edge Computing and Communications (EDGE)","volume":"10 6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127280250","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A Brief History of Liquid Software 液体软件简史

2023 IEEE International Conference on Edge Computing and Communications (EDGE) Pub Date : 2023-07-01 DOI: 10.1109/EDGE60047.2023.00058

C. Pautasso

引用次数: 0