Raju Imandi , Arijit Roy , Yong-Guk Kim , Pavan Kumar B.N.
{"title":"在支持雾的无人机即服务中构建可持续的联邦学习模型,用于航空图像分类","authors":"Raju Imandi , Arijit Roy , Yong-Guk Kim , Pavan Kumar B.N.","doi":"10.1016/j.suscom.2025.101133","DOIUrl":null,"url":null,"abstract":"<div><div>The Fog-enabled UAV-as-a-Service (FU-Serve) platform leverages distributed fog nodes to enable real-time data processing for multiple concurrent applications. However, the computational limitations of these fog nodes significantly hamper the execution of resource-intensive deep learning (DL) algorithms, compromising both operational performance and energy sustainability. To address these challenges, we integrated Federated Learning (FL) within the FU-Serve platform, coupled with the development of three specialized FL-based models tailored for on-device image classification. First, we introduced a sustainable adaptation of MobileNetV2 that synergizes Transfer Learning (TL) with FL principles. This model achieves 97.68% accuracy with an 8.64 MB footprint by distributing pre-trained weights to optimize bandwidth efficiency. To further address resource constraints of fog nodes, we designed FUSERNet—a lightweight DL architecture employing separable convolutions and skip connections, which reduces computational overhead while preserving critical feature representations. This model achieves 97.47% accuracy with an ultra-compact footprint of 237 KB, demonstrating a 98.59% reduction in size compared to state-of-the-art models. Finally, our third model, FusionNet, combines the strengths of MobileNetV2 and FUSERNet to deliver a balanced solution, achieving 97.75% accuracy with moderate resource requirements (8.86 MB). We evaluated our models on the AIDER and NDD disaster response datasets, our models demonstrate superior performance in classifying critical natural disaster scenarios. Notably, FusionNet matches SOTA accuracy levels while reducing memory consumption by 50%, and FUSERNet’s 0.23 MB size enables deployment on even the most resource-constrained UAVs. These contributions enhance the FU-Serve platform’s real-time decision-making capabilities, balancing computational efficiency and mission-critical accuracy for sustainable disaster response.</div></div>","PeriodicalId":48686,"journal":{"name":"Sustainable Computing-Informatics & Systems","volume":"46 ","pages":"Article 101133"},"PeriodicalIF":3.8000,"publicationDate":"2025-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Building sustainable federated learning models in Fog-enabled UAV-as-a-Service for aerial image classification\",\"authors\":\"Raju Imandi , Arijit Roy , Yong-Guk Kim , Pavan Kumar B.N.\",\"doi\":\"10.1016/j.suscom.2025.101133\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><div>The Fog-enabled UAV-as-a-Service (FU-Serve) platform leverages distributed fog nodes to enable real-time data processing for multiple concurrent applications. However, the computational limitations of these fog nodes significantly hamper the execution of resource-intensive deep learning (DL) algorithms, compromising both operational performance and energy sustainability. To address these challenges, we integrated Federated Learning (FL) within the FU-Serve platform, coupled with the development of three specialized FL-based models tailored for on-device image classification. First, we introduced a sustainable adaptation of MobileNetV2 that synergizes Transfer Learning (TL) with FL principles. This model achieves 97.68% accuracy with an 8.64 MB footprint by distributing pre-trained weights to optimize bandwidth efficiency. To further address resource constraints of fog nodes, we designed FUSERNet—a lightweight DL architecture employing separable convolutions and skip connections, which reduces computational overhead while preserving critical feature representations. This model achieves 97.47% accuracy with an ultra-compact footprint of 237 KB, demonstrating a 98.59% reduction in size compared to state-of-the-art models. Finally, our third model, FusionNet, combines the strengths of MobileNetV2 and FUSERNet to deliver a balanced solution, achieving 97.75% accuracy with moderate resource requirements (8.86 MB). We evaluated our models on the AIDER and NDD disaster response datasets, our models demonstrate superior performance in classifying critical natural disaster scenarios. Notably, FusionNet matches SOTA accuracy levels while reducing memory consumption by 50%, and FUSERNet’s 0.23 MB size enables deployment on even the most resource-constrained UAVs. These contributions enhance the FU-Serve platform’s real-time decision-making capabilities, balancing computational efficiency and mission-critical accuracy for sustainable disaster response.</div></div>\",\"PeriodicalId\":48686,\"journal\":{\"name\":\"Sustainable Computing-Informatics & Systems\",\"volume\":\"46 \",\"pages\":\"Article 101133\"},\"PeriodicalIF\":3.8000,\"publicationDate\":\"2025-05-15\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Sustainable Computing-Informatics & Systems\",\"FirstCategoryId\":\"94\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S221053792500054X\",\"RegionNum\":3,\"RegionCategory\":\"计算机科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"COMPUTER SCIENCE, HARDWARE & ARCHITECTURE\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Sustainable Computing-Informatics & Systems","FirstCategoryId":"94","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S221053792500054X","RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"COMPUTER SCIENCE, HARDWARE & ARCHITECTURE","Score":null,"Total":0}
Building sustainable federated learning models in Fog-enabled UAV-as-a-Service for aerial image classification
The Fog-enabled UAV-as-a-Service (FU-Serve) platform leverages distributed fog nodes to enable real-time data processing for multiple concurrent applications. However, the computational limitations of these fog nodes significantly hamper the execution of resource-intensive deep learning (DL) algorithms, compromising both operational performance and energy sustainability. To address these challenges, we integrated Federated Learning (FL) within the FU-Serve platform, coupled with the development of three specialized FL-based models tailored for on-device image classification. First, we introduced a sustainable adaptation of MobileNetV2 that synergizes Transfer Learning (TL) with FL principles. This model achieves 97.68% accuracy with an 8.64 MB footprint by distributing pre-trained weights to optimize bandwidth efficiency. To further address resource constraints of fog nodes, we designed FUSERNet—a lightweight DL architecture employing separable convolutions and skip connections, which reduces computational overhead while preserving critical feature representations. This model achieves 97.47% accuracy with an ultra-compact footprint of 237 KB, demonstrating a 98.59% reduction in size compared to state-of-the-art models. Finally, our third model, FusionNet, combines the strengths of MobileNetV2 and FUSERNet to deliver a balanced solution, achieving 97.75% accuracy with moderate resource requirements (8.86 MB). We evaluated our models on the AIDER and NDD disaster response datasets, our models demonstrate superior performance in classifying critical natural disaster scenarios. Notably, FusionNet matches SOTA accuracy levels while reducing memory consumption by 50%, and FUSERNet’s 0.23 MB size enables deployment on even the most resource-constrained UAVs. These contributions enhance the FU-Serve platform’s real-time decision-making capabilities, balancing computational efficiency and mission-critical accuracy for sustainable disaster response.
期刊介绍:
Sustainable computing is a rapidly expanding research area spanning the fields of computer science and engineering, electrical engineering as well as other engineering disciplines. The aim of Sustainable Computing: Informatics and Systems (SUSCOM) is to publish the myriad research findings related to energy-aware and thermal-aware management of computing resource. Equally important is a spectrum of related research issues such as applications of computing that can have ecological and societal impacts. SUSCOM publishes original and timely research papers and survey articles in current areas of power, energy, temperature, and environment related research areas of current importance to readers. SUSCOM has an editorial board comprising prominent researchers from around the world and selects competitively evaluated peer-reviewed papers.