[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-290773-105":59,"doc-detail-290773-en":131},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":124,"head_meta":126,"extra_data":128,"updated_unix":130},105,"en","real-time-tcp-packet-loss-prediction-using-machine-learning","Real-Time TCP Packet Loss Prediction Using Machine Learning","","Congestion and resulting packet loss in TCP connections can degrade performance and reduce the end users’ Quality of Experience. Traditional TCP congestion control mainly reacts when packet loss signals congestion, often after senders have increased their rate until drops occur. The work investigates replacing heuristic backoff decisions with machine learning that predicts impending packet loss in real time. Two ML models for Reno and Cubic flows are trained and used to trigger sender-side ECN signaling to reduce rate before loss happens, improving real-time behavior.",{"@graph":69,"@context":123},[70,84,106],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/real-time-tcp-packet-loss-prediction-using-machine-learning/290773/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/real-time-tcp-packet-loss-prediction-using-machine-learning/290773.png","ImageObject",300,407,{"name":92,"@type":93},"Rowan","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-11-04","2026-09-17",true,{"@type":102,"interactionType":103,"userInteractionCount":105},"InteractionCounter",{"@type":104},"ViewAction",12,{"@type":107,"mainEntity":108},"FAQPage",[109,115,119],{"name":110,"@type":111,"acceptedAnswer":112},"Why does TCP packet loss harm performance and user experience?","Question",{"text":113,"@type":114},"Congestion and packet loss degrade connection performance and reduce end-user Quality of Experience (QoE).","Answer",{"name":116,"@type":111,"acceptedAnswer":117},"How does the proposed approach predict packet loss in real time?",{"text":118,"@type":114},"It trains two machine learning models for Reno and Cubic flows using sender-available delay/throughput-related metrics, then predicts when packet loss would happen before it occurs.",{"name":120,"@type":111,"acceptedAnswer":121},"How is the prediction used to prevent packet loss?",{"text":122,"@type":114},"The TCP sender is informed to reduce its sending rate by setting a local, artificially generated ECN signal at the sender prior to the predicted loss event.","https://schema.org",{"og:url":83,"og:type":125,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":127,"canonical":83},"index,follow",{"doc_id":129,"site_id":62},290773,1790247200,{"code":4,"msg":5,"data":132},{"doc_id":129,"user_id":133,"nickname":92,"user_avatar":134,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":135,"file_id":136,"file_url":137,"file_type":138,"file_size":139,"view_count":105,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":140,"language":141,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":142,"faqs":143,"seo_title":144,"seo_description":67,"update_tm":145,"read_time":146,"preview_page_count":4},1099514067415,"https://ap-avatar.wpscdn.com/avatar/100002539d78ffe74a7?x-image-process=image/resize,m_fixed,w_180,h_180&k=1779092875211072502","Received 6 October 2024, accepted 28 October 2024, date of publication 30 October 2024, date of current version 8 November 2024. Digital Object Identifier 10.1109/ACCESS.2024.3488511  \nReal-Time TCP Packet Loss Prediction Using Machine Learning  \nMICHAEL WELZL1,(Member, IEEE), SAFIQUL ISLAM2,(Senior Member, IEEE),  \nAND MAXIMILIAN VON STEPHANIDES3  \n1Department of Informatics, University of Oslo, 0316 Oslo, Norway  \n2Department of Computer Science, Oslo Metropolitan University, 0130 Oslo, Norway  \n3Dintero AS, 0484 Oslo, Norway  \nCorresponding author: Michael Welzl ([michawe@ifi.uio.no](michawe@ifi.uio.no))  \nABSTRACT Congestion and resulting packet loss in TCP connections can lead to performance degradation and reduce the Quality of Experience (QoE) for end users. Many common TCP congestion control algorithms therefore adjust their sending rate proactively, using heuristics based on measured delay and/or throughput. We have investigated whether it would be possible to replace such heuristics with Machine Learning (ML), such that Internet hosts could meaningfully react when a packet loss is predicted to happen. For example, such a reaction could involve the use of Forward Erasure Correction (FEC) to protect against the loss, or to dynamically tune parameters of TCP. We present two ML models that were trained for Reno and Cubic flows, respectively, and show that they can be beneficially applied in real time. We do this by informing a TCP sender to reduce its rate via a local, artificially generated Explicit Congestion Notification (ECN) signal at the sender before packet loss would have happened.  \nINDEX TERMS TCP, congestion control, machine learning, ECN.  \nI. INTRODUCTION  \nTraditional congestion control algorithms such as Reno [1] and Cubic [2] adjust the sending rate reactively when they see a congestion signal in the form of packet loss. This leads to a behavior where the sender keeps increasing its rate until congestion occurs, eventually causing packets to be dropped. Congestion and resulting packet loss can lead to performance degradation and reduced Quality of Experience (QoE) for end users.  \nOne way to circumvent these detrimental effects is to react proactively and avoid packet loss. By allowing the sender to reduce its rate before congestion occurs, packet loss and resulting retransmissions are reduced, which reduces latency and improves the QoE for end users. This can be done by monitoring throughput or delay (as seen via returning acknowledgments (ACKs))—a strategy that is incorporated in many modern congestion control mechanisms, such as BBR [3], Low Extra Delay Background Transport (LEDBAT) [4], Verus [5], COPA [6],  \nThe associate editor coordinating the review of this manuscript and approving it for publication was Wai-Keung Fung .  \nNetwork-Assisted Dynamic Adaptation (NADA) [7] and Self-Clocked Rate Adaptation for Multimedia (SCReAM) [8], to name but a few. It can also be done with Explicit Congestion Notification (ECN) [9], where routers set a bit in the IP header upon impending congestion, and the receiver uses another bit in the TCP header to tell the sender that it should back off.  \nWith this paper, we do not participate in the competition for better congestion control, nor do we take a side in it regarding the choice of input heuristic. Instead, we try to answer the following question: Since heuristics based on sender-side monitored delay and throughput metrics can be used to determine when a sender should back off, could we also use ML on these metrics to predict, in real time, when a packet loss would happen? Such a prediction could be used in various ways, [e.g. as](e.g. as) input for a new congestion control mechanism, or by amending an existing one, e.g. to increase more cautiously or back off early; adding Forward Erasure Correction (FEC) to the data [10], but only at the right moment, to minimize unnecessary overhead; deciding whether starting a new connection should be slightly delayed; or ad","cbCaioMg5ihr8hiL","https://ap.wps.com/l/cbCaioMg5ihr8hiL","pdf",1627099,13,"English","# Introduction\n## Proactive packet loss avoidance\n## Using ML for real-time prediction\n# Contributions\n## XGBoost models for Reno and Cubic\n## Feature and metric selection\n## Real-time inference evaluation","[{\"question\":\"Why does TCP packet loss harm performance and user experience?\",\"answer\":\"Congestion and packet loss degrade connection performance and reduce end-user Quality of Experience (QoE).\"},{\"question\":\"How does the proposed approach predict packet loss in real time?\",\"answer\":\"It trains two machine learning models for Reno and Cubic flows using sender-available delay/throughput-related metrics, then predicts when packet loss would happen before it occurs.\"},{\"question\":\"How is the prediction used to prevent packet loss?\",\"answer\":\"The TCP sender is informed to reduce its sending rate by setting a local, artificially generated ECN signal at the sender prior to the predicted loss event.\"}]","Real-Time TCP Packet Loss Prediction Using Machine Learning | PDF",1789642544,33]