[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-119551-en":3,"doc-seo-119551-105":30,"detail-sidebar-cat-0-en-105":91},{"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":29},119551,1374391974564,"Clementine","https://ap-avatar.wpscdn.com/avatar/14000253aa45c000a9e?x-image-process=image/resize,m_fixed,w_180,h_180&k=1779874745381141002",8,"Research & Report","Diffusion-Based Machine Learning Method for Accelerating Quantum Transport Simulations in Nanowire Transistors","Numerical device simulations of nanoscale transistors are essential for semiconductor research, yet they are costly in runtime and require many self-consistent iterations. This work introduces an image-processing machine learning approach to accelerate quantum transport simulations for nanowire transistors. The proposed ML diffusion model leverages a U-Net-based architecture with research-specific modifications to speed up simulation steps while improving convergence behavior. Implemented in the in-house NESS code, the method achieves up to 60% faster performance.","Diffusion-Based Machine Learning Method for Accelerating Quantum Transport Simulations in  \nNanowire Transistors  \nPreslav Aleksandrov  \nDepartment of Computer Science University of Cambridge  \nEmail: [pa511@cam.ac.uk](pa511@cam.ac.uk)  \nPranav Acharya  \nJames Watt School of Engineering University of Glasgow  \nEmail: [2670931A@student.gla.ac.uk](2670931A@student.gla.ac.uk)  \nVihar Georgiev  \nJames Watt School of Engineering University of Glasgow  \nEmail: [vihar.georgiev@glasgow.ac.uk](vihar.georgiev@glasgow.ac.uk)  \nAbstract—Numerical device simulations of nano-scale transistors are a vital part of semiconductor research. Recent advancements in machine learning are slowly finding their way into the nanoelectronic device simulation community. Machine learning (ML) supported methods hold the promise of significantly reducing the cost and wall-clock time of simulations.  \nIn this work, we present a new way to utilise advanced image processing machine learning techniques to accelerate quantum transport simulations of nanowire transistors. Our method uses a ML based diffusion model to accelerate the speed of the numerical simulations and reduce the self-consistent numerical iterations. The ML diffusion model is based on the popular UNet architecture together with application specific modifications, which we have developed for the purpose of our research. Our ML based method improves the speed and the performance of our in-house code, called NESS, by up to 60%.  \nI. INTRODUCTION  \nNanowires play a pivotal role in the advancement of nanoelectronics, offering unique properties that are essential for the development of high-performance electronic devices [1] . These one-dimensional structures are crucial in applications ranging from transistors to sensors, where their quantum mechanical properties can be leveraged for enhanced functionality [2] . Hence, accurate and efficient simulation methods are critical for exploring and optimising nanowire-based devices in order to decrease fabrication time and cost [3] .  \nOne way to reduce the simulation time and increase the speed of the simulations is to combine numerical simulations with ML methods. For example, our previous work introduced the ML-NEGF method, which utilised convolutional neural networks (CNNs) to accelerate non-equilibrium Green’s function (NEGF) simulations [4]. While this approach significantly improved convergence speed, there remained room for further enhancements. In this follow-up study, we present a novel model architecture that replaces CNNs with diffusion models, which have demonstrated superior performance in capturing complex physical phenomena [5] . Diffusion models, with their ability to better model stochastic processes, provide a more accurate and efficient framework for our simulations. Additionally, we have developed an improved location map, which significantly enhances the spatial resolution and accuracy of  \nFig. 1: Schematic representation of the nanowire structure. (a) Cross-sectional view showing the dimensions: 12 nm height, 3 nm width, and 1 nm gate oxide thickness on each side. (b) Three-dimensional view of the nanowire, illustrating its 22 nm length. The red region represents the silicon semiconductor channel, while the blue regions indicate the gate silicon oxide surrounding the channel.  \nour simulations. This advancement allows us to significantly reduce simulation time by paving the way for more effective exploration and development of the next-generation nanoelectronic devices.  \nII. DEVICE STRUCTURE  \nTo validate our new simulation methodology, we have designed a nanosheet transistor structure relating to the so-called 3 nm technology node [6] . Figure 1 illustrates the geometry created using our in-house structure generator implemented in NESS [7] . The device features a N-type gate-all-around (GAA) configuration with a Si channel length of 16 nm and source/drain lengths of 3 nm each, resulting in a total device length of 22 nm. The rectangu","cbCaihEPorH6qeU8","https://ap.wps.com/l/cbCaihEPorH6qeU8","pdf",718946,1,5,"English","en",105,"# Introduction\n# Device Structure","[{\"question\":\"Why are quantum transport simulations for nanowire transistors computationally demanding?\",\"answer\":\"They require accurate numerical device modeling and repeated self-consistent iterations, leading to high cost and long wall-clock time.\"},{\"question\":\"How does the proposed method accelerate simulations?\",\"answer\":\"It uses an ML-based diffusion model to reduce self-consistent numerical iterations and increase simulation speed compared with the baseline approach.\"},{\"question\":\"What neural network architecture underpins the diffusion model in this work?\",\"answer\":\"The diffusion model is based on the UNet architecture, combined with application-specific modifications developed for the authors' research.\"}]","Diffusion-Based Machine Learning Method for Accelerating Quantum Transport Simulations in Nanowire Transistors | PDF",1785724917,13,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"diffusion-based-machine-learning-method-for-accelerating-quantum-transport-simulations-in-nanowire-transistors","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/diffusion-based-machine-learning-method-for-accelerating-quantum-transport-simulations-in-nanowire-transistors/119551/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-03",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why are quantum transport simulations for nanowire transistors computationally demanding?","Question",{"text":75,"@type":76},"They require accurate numerical device modeling and repeated self-consistent iterations, leading to high cost and long wall-clock time.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the proposed method accelerate simulations?",{"text":80,"@type":76},"It uses an ML-based diffusion model to reduce self-consistent numerical iterations and increase simulation speed compared with the baseline approach.",{"name":82,"@type":73,"acceptedAnswer":83},"What neural network architecture underpins the diffusion model in this work?",{"text":84,"@type":76},"The diffusion model is based on the UNet architecture, combined with application-specific modifications developed for the authors' research.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,109,114,119,122,127,130,134],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":21,"doc_module":4,"doc_module_name":46,"category_name":106,"show_sort_weight":107,"slug":108},"Comic",60,"comic",{"id":110,"doc_module":4,"doc_module_name":46,"category_name":111,"show_sort_weight":112,"slug":113},6,"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":46,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":21,"slug":137},19,"General","general"]