[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-125255-en":3,"doc-seo-125255-105":29,"detail-sidebar-cat-0-en-105":89},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":11},125255,687197207057,"Sage","https://ap-avatar.wpscdn.com/davatar_29158cc5080c5b710cf443261637dec0",8,"Research & Report","Distributed Predictive Maintenance Architecture for Edge Sensors Networks - Optimal Regression Based Machine Learning Model","Predictive maintenance on edge sensor hardware is difficult due to strict power constraints that limit computation on CPUs and microcontrollers. This work evaluates using FPGAs for hardware acceleration, emphasizing their ability to outperform conventional and even GPU approaches under low-precision data while improving performance per watt. A methodology is proposed to identify, train, and adapt suitable regression-based ML models for FPGA deployment using lightweight model techniques and high-level synthesis (HLS). Power estimation and implementation considerations guide the design toward efficient edge execution.","School of Physics, Engineering and Computer Science  \nDistributed Predictive Maintenance Architecture for Edge Sensors Networks: An Optimal Regression Based Machine Learning Model  \nJacob Greasley 1,2*, Oluyomi Simpson 2 , Iosif Mporas 2  \n1MBDA  \n2 University of Hertfordshire  \n*corresponding author: [j.greasley2@herts.ac.uk](j.greasley2@herts.ac.uk)  \nIn predictive maintenance (PdM), implementing machine learning models (ML) on edge sensor hardware is particularly challenging. This is due to power constraints which significantly reduce computational performance in conventional embedded processors such as central processing units (CPUs) and microcontroller units (MCUs) . However, Field Programmable Gate Arrays (FPGAs) have been identified as an ideal processing unit to overcome this, providing hardware acceleration of models on the edge. With low-precision data, FPGAs have been shown to outperform conventional processing units both in terms of giga-operations-per-second (GOPS) and power consumption. This research seeks to establish an effective methodology for implementing high-level ML regression models on FPGAs within edge sensors.  \nKeywords: FPGAs; machine learning; edge sensors; hardware acceleration; regression models.  \nIntroduction  \nAdvancements in modern machine learning (ML) techniques have been incorporated with predictive maintenance (PdM) to great effect, allowing data analysts and maintenance engineers to uncover hidden anomalies and interpret their meanings in ways never before possible [1] . However, machine learning alone cannot account for the rise of PdM. Its great potential is enabled by advancements in a whole host of other interdependent technologies. Today, high-level computational tools are more powerful than ever before, allowing for greater precision and accuracy in simulation and analysis of engineering systems through physicsbased modelling. Telecommunication technologies are also more powerful, facilitating faster, higher-capacity interconnectivity between systems, end-users, and databases-the concept referred to as the internet-of-things (IoT) [2] .  \nUnder the IoT umbrella, so-called edge sensors used for condition monitoring (CM) are also ongoing vast technological advancements, whereby embedded computing enables a smart gateway between the physical world and the digital infrastructure of PdM. Now, in distributed architectures machine learning models can be pushed to the edge to overcome common challenges such as latency and bandwidth requirements associated with networking, cloud computing and the wider IoT [3] . As a result, edge sensors themselves need to evolve, to improve symbiosis with machine learning technology.  \nImplementing ML models on sensor hardware is challenging, as their complexity and the inherent low power constraints of edge devices, greatly reduce computational performance in common embedded processors such as CPUs and MCUs. Graphics processing units (GPUs) demonstrate greater performance in the training and validation of ML models than CPUs and MCUs, while also comfortably handling floating point datatypes, but their high-power consumption and physical sizes make them unsuitable for low-power embedded application. Hence, they are not well suited to typical remote sensors and IoT nodes. On the other hand, research has shown that for low precision data, FPGAs outperform GPUs both in terms of giga-operations-per-second (GOPS) and power consumption, with a better performance per watt in the execution of mathematically complex operations. They also have better form factor and heat dissipation.  \nFor speed, ASICs are the best option for hardware acceleration, however they are the most costly and time consuming to develop [5] [6] . In comparison, FPGAs are considerably cheaper and quicker to develop, and still  \nSchool of Physics, Engineering and Computer Science  \nprovide superior speed to CPUs, MCUs and GPUs due to their parallelism. This parallelism especially thrives in perf","cbCaio2NqqLcpJfG","https://ap.wps.com/l/cbCaio2NqqLcpJfG","pdf",467242,1,3,"English","en",105,"# Introduction\n# Experimental/Simulation\n# Results and discussion\n# Conclusion","[{\"question\":\"Why are machine learning models challenging to deploy on edge sensors for predictive maintenance?\",\"answer\":\"Edge devices face strict power constraints that reduce computational performance on common embedded processors like CPUs and microcontrollers.\"},{\"question\":\"What role do FPGAs play in the proposed predictive maintenance architecture?\",\"answer\":\"FPGAs are used as ideal processing units for hardware acceleration, offering better performance per watt and suitable form factor for edge execution, especially with low-precision data.\"},{\"question\":\"How does the research plan to implement regression-based machine learning on FPGAs?\",\"answer\":\"It focuses on selecting regression algorithms for edge deployment, investigating lightweight ML libraries/platforms, and using high-level synthesis (HLS) while optimizing code conversion to RTL for FPGA execution.\"}]","Distributed Predictive Maintenance Architecture for Edge Sensors Networks - Optimal Regression Based Machine Learning Model | PDF",1785897742,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":84,"head_meta":86,"extra_data":88,"updated_unix":28},"distributed-predictive-maintenance-architecture-for-edge-sensors-networks-optimal-regression-based-machine-learning-model","",{"@graph":35,"@context":83},[36,52,66],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,49],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":21},"https://docshare.wps.com/document/research-report/",{"item":50,"name":13,"@type":42,"position":51},"https://docshare.wps.com/document/distributed-predictive-maintenance-architecture-for-edge-sensors-networks-optimal-regression-based-machine-learning-model/125255/",4,{"url":50,"name":13,"@type":53,"author":54,"headline":13,"publisher":56,"fileFormat":59,"inLanguage":23,"description":14,"dateModified":60,"datePublished":60,"encodingFormat":59,"isAccessibleForFree":61,"interactionStatistic":62},"DigitalDocument",{"name":9,"@type":55},"Person",{"url":40,"name":57,"@type":58},"DocShare","Organization","application/pdf","2026-08-05",true,{"@type":63,"interactionType":64,"userInteractionCount":4},"InteractionCounter",{"@type":65},"ViewAction",{"@type":67,"mainEntity":68},"FAQPage",[69,75,79],{"name":70,"@type":71,"acceptedAnswer":72},"Why are machine learning models challenging to deploy on edge sensors for predictive maintenance?","Question",{"text":73,"@type":74},"Edge devices face strict power constraints that reduce computational performance on common embedded processors like CPUs and microcontrollers.","Answer",{"name":76,"@type":71,"acceptedAnswer":77},"What role do FPGAs play in the proposed predictive maintenance architecture?",{"text":78,"@type":74},"FPGAs are used as ideal processing units for hardware acceleration, offering better performance per watt and suitable form factor for edge execution, especially with low-precision data.",{"name":80,"@type":71,"acceptedAnswer":81},"How does the research plan to implement regression-based machine learning on FPGAs?",{"text":82,"@type":74},"It focuses on selecting regression algorithms for edge deployment, investigating lightweight ML libraries/platforms, and using high-level synthesis (HLS) while optimizing code conversion to RTL for FPGA execution.","https://schema.org",{"og:url":50,"og:type":85,"og:title":13,"og:site_name":57,"og:description":14},"article",{"robots":87,"canonical":50},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":90},[91,95,99,103,108,113,118,121,126,129,133],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":92,"show_sort_weight":93,"slug":94},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":96,"show_sort_weight":97,"slug":98},"Literature",80,"literature",{"id":51,"doc_module":4,"doc_module_name":45,"category_name":100,"show_sort_weight":101,"slug":102},"Exam",70,"exam",{"id":104,"doc_module":4,"doc_module_name":45,"category_name":105,"show_sort_weight":106,"slug":107},5,"Comic",60,"comic",{"id":109,"doc_module":4,"doc_module_name":45,"category_name":110,"show_sort_weight":111,"slug":112},6,"Technology",50,"technology",{"id":114,"doc_module":4,"doc_module_name":45,"category_name":115,"show_sort_weight":116,"slug":117},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":119,"slug":120},30,"research-report",{"id":122,"doc_module":4,"doc_module_name":45,"category_name":123,"show_sort_weight":124,"slug":125},9,"Religion & Spirituality",20,"religion-spirituality",{"id":124,"doc_module":4,"doc_module_name":45,"category_name":127,"show_sort_weight":124,"slug":128},"World Cup","world-cup",{"id":130,"doc_module":4,"doc_module_name":45,"category_name":131,"show_sort_weight":130,"slug":132},10,"Lifestyle","lifestyle",{"id":134,"doc_module":4,"doc_module_name":45,"category_name":135,"show_sort_weight":104,"slug":136},19,"General","general"]