[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-86127-en":3,"doc-seo-86127-105":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":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":13,"seo_description":14,"update_tm":28,"read_time":29},86127,962075114765,"Quinn","https://ap-avatar.wpscdn.com/davatar_a8503ba1806abce46bf441b54a3ca4cd",8,"Research & Report","Link Adaptation Using Joint Thompson Sampling","Link adaptation selects modulation and coding scheme (MCS) to maximize expected throughput while meeting an error-rate constraint under varying wireless channel conditions at the MAC layer. The approach uses ACK/NACK statistics and CQI feedback, modeling MCS choice as a multi-armed bandit where each MCS is a Bernoulli arm with an unknown transmission success probability. Joint-Thompson Sampling (Joint-TS) replaces independent Beta priors with a multivariate ordered Beta prior to preserve monotonic success-probability structure. Simulations show superior, robust throughput compared with prior MAB strategies in challenging scenarios.","arXiv :2607 . 1 1075v 1 [ cs .LG] 13 Jul 2026  \nLink Adaptation Using Joint-Thompson Sampling  \nVignatha Vinjam∗ Manjunath Kolavennu∗ Myna Vajha∗ Karthik Periyapattana Narayanaprasad†  \n∗ Department of EE, IIT Hyderabad † Department of AI, IIT Hyderabad  \n[ee23resch14002@iith.ac.in](ee23resch14002@iith.ac.in), [ee25mtech02002@iith.ac.in](ee25mtech02002@iith.ac.in), [mynav@ee.iith.ac.in](mynav@ee.iith.ac.in), [pnkarthik@ai .iith.ac.in](pnkarthik@ai .iith.ac.in)  \nAbstract  \nThe choice of Modulation and Coding (MCS) type for a particular channel condition is made through link adaptation (LA) algorithms that operate at the MAC layer. These algorithms rely on the ACK/NACK statistics and the channel quality index (CQI) feedback. Several existing works model LA as a multi-armed bandit (MAB) problem across cellular and Wi-Fi links. In the MAB formulation, each available MCS is a Bernoulli arm parameterized by its transmission success probability, and the goal is to design a selection strategy that accrues maximum reward. Several popular MAB algorithms, such as upper confidence bound (UCB) and Thompson Sampling (TS), have been proposed in the literature. Using the fact that MCS success probabilities are ordered, we propose the Joint-Thompson Sampling (Joint-TS) algorithm. Unlike classical TS, which assumes independent Beta distributions for each arm, Joint-TS utilizes a multivariate ordered Beta distribution as the prior to preserve the inherent monotonicity of success probabilities. Our simulation results show that while existing MAB algorithms fail in specific scenarios, Joint-TS delivers competitive throughput with robust, consistent performance in all scenarios.  \nIndex Terms  \nThompson Sampling, multi-armed bandits, link adaptation, Wi-Fi, 5G, 6G, multivariate ordered Beta distribution  \nI. INTRODUCTION  \nLink adaptation (LA) is an essential MAC layer algorithm that selects link parameters such as Modulation and Coding Scheme (MCS), power control, number of layers, and precoders for dynamic wireless channels. LA focusing on MCS selection is widely known as Adaptive Modulation and Coding (AMC) or as Adaptive Data Rate (ADR) in WiFi. The goal of MCS selection is to maximize expected throughput under an error rate constraint for a given channel state—the Signal-to-Interference-plus-Noise Ratio (SINR) . Mathematically,  \nRmax (c) = max si · (1 − BLER(i, c)),  \ni: BLER(i,c) ≤ ptarget  \nwhere si is the spectral efficiency corresponding to MCS i, determined as product of number of bits per modulation symbol times the rate of forward error correction (FEC) code used (cf. Table I), and BLER(i, c) is the Block Error Rate at channel state (or effective SINR) c upon using MCS i. Effective SINR [1] maps the sub-carrier SINR vector to a single scalar, enabling the use of Additive White Gaussian Noise (AWGN) BLER tables. While the BLER target, ptarget , is typically set to 0.1 [2], SINRdependent BLER targets result in better performance of ARQ schemes at the MAC layer. [3] If the throughput-maximizing MCS always satisfies this BLER constraint, then the condition BLER(i, c) ≤ ptarget can be ignored. We observe this holds true for the 5G NR waveform simulations conducted over the AWGN channel using the MATLAB 5G toolbox.  \n\n| MCS (i) | Mi | Fi | si |\n| --- | --- | --- | --- |\n| 0 | 2 | 120 | 0.2344 |\n| 1 | 2 | 157 | 0.3066 |\n| 2 | 2 | 193 | 0.3770 |\n| 3 | 2 | 251 | 0.4902 |\n| .\u003Cbr>. | .\u003Cbr>. | .\u003Cbr>. | .\u003Cbr>. |\n| . | . | . | . |\n| 27 | 6 | 910 | 5.3320 |\n| 28 | 6 | 948 | 5.5547 |\n\nTABLE I: 5G NR MCS configurations [4, 5.1.3.1 Table 1], where Mi is bits/symbol, Fi is FEC code rate (×1024), and si = Mi · Fi /1024 is the spectral efficiency.  \nWhile uplink MCS selection is straightforward using the SINR estimates, the downlink selection is a complex task since the gNB lacks direct channel knowledge. As a remedy, a quantized MCS suggestion, called Channel Quality Index (CQI), can be requested from the UE. See [4, [5.2.2.1](5.2.2.1)] for a list of 5G ","cbCaileCOigKZmFG","https://ap.wps.com/l/cbCaileCOigKZmFG","pdf",1470492,3,1,7,"English","en",105,"# Introduction\n## Link adaptation background\n## MCS selection and BLER constraints\n## Prior work: ILLA and OLLA\n## MAB-based link adaptation and Thompson Sampling","[{\"question\":\"What problem does link adaptation aim to solve in this work?\",\"answer\":\"It aims to choose link parameters—especially the modulation and coding scheme (MCS)—to maximize expected throughput under an error-rate (BLER) constraint for a given channel state.\"},{\"question\":\"How is link adaptation for MCS selection formulated in the multi-armed bandit view?\",\"answer\":\"Each available MCS is treated as an arm that generates Bernoulli rewards based on transmission success probability, and the algorithm learns from ACK/NACK and CQI feedback to maximize cumulative reward.\"},{\"question\":\"What is the key difference between Joint-TS and classical Thompson Sampling in this paper?\",\"answer\":\"Joint-TS uses a multivariate ordered Beta prior to preserve the inherent monotonic ordering of MCS success probabilities, unlike classical TS that assumes independent Beta distributions per arm.\"}]",1784208699,18,{"code":4,"msg":31,"data":32},"ok",{"site_id":25,"language":24,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"link-adaptation-using-joint-thompson-sampling","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,50],{"item":41,"name":42,"@type":43,"position":21},"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":20},"https://docshare.wps.com/document/research-report/",{"item":51,"name":13,"@type":43,"position":52},"https://docshare.wps.com/document/link-adaptation-using-joint-thompson-sampling/86127/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":24,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":41,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-26","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does link adaptation aim to solve in this work?","Question",{"text":75,"@type":76},"It aims to choose link parameters—especially the modulation and coding scheme (MCS)—to maximize expected throughput under an error-rate (BLER) constraint for a given channel state.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How is link adaptation for MCS selection formulated in the multi-armed bandit view?",{"text":80,"@type":76},"Each available MCS is treated as an arm that generates Bernoulli rewards based on transmission success probability, and the algorithm learns from ACK/NACK and CQI feedback to maximize cumulative reward.",{"name":82,"@type":73,"acceptedAnswer":83},"What is the key difference between Joint-TS and classical Thompson Sampling in this paper?",{"text":84,"@type":76},"Joint-TS uses a multivariate ordered Beta prior to preserve the inherent monotonic ordering of MCS success probabilities, unlike classical TS that assumes independent Beta distributions per arm.","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,119,122,127,130,134],{"id":21,"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":52,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":22,"doc_module":4,"doc_module_name":46,"category_name":116,"show_sort_weight":117,"slug":118},"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":106,"slug":137},19,"General","general"]