[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-83599-en":3,"doc-seo-83599-105":30,"detail-sidebar-cat-0-en-105":92},{"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},83599,8796095360427,"Lucas Martin","https://ap-avatar.wpscdn.com/davatar_994ba38a5ba835b3df7d355c54d3ed8d",8,"Research & Report","Enabling Real-Time AI in O-RAN Deploying and Measuring AI Inside a Near-RT RIC xApp","Open Radio Access Network (O-RAN) architectures use programmable Near-Real-Time RAN Intelligent Controllers (Near-RT RICs) with xApps to enable closed-loop control at 10 ms–1 s. This work validates an AI-enabled network state classification xApp running inside a live OpenAirInterface (OAI) and FlexRIC testbed, rather than relying on offline or external inference paths. Logistic regression and a shallow MLP are exported as deterministic C inference modules, achieving 1–5 µs and 10–25 µs measured inference latency, while keeping end-to-end service latency under 4 ms. Model comparisons, noise ablation, and latency profiling confirm meeting the 10 ms budget for over 95% of loop executions.","Enabling Real-Time AI in O-RAN: Deploying and Measuring AI Inside a Near-RT RIC xApp  \nLawrence Obiuwevwi, Krzysztof J. Rechowicz, Sampath Jayarathna, Safdar Hussain Bouk, Fahmida Afrin, C. Nicolas Barati, Neda Moghim, Valentina Nannou, Muhammad Enayetur Rahman, and Sachin Shetty  \nOld Dominion University, Norfolk, VA, USA  \n[lobiu001@odu.edu](lobiu001@odu.edu), [krechowi@odu.edu](krechowi@odu.edu), [sampath@cs.odu.edu](sampath@cs.odu.edu),  \n[sbouk@odu.edu](sbouk@odu.edu), [fafri002@odu.edu](fafri002@odu.edu), [cbaratin@odu.edu](cbaratin@odu.edu),  \n[nmoghim@odu.edu](nmoghim@odu.edu), [vtebo001@odu.edu](vtebo001@odu.edu), [mrahm011@odu.edu](mrahm011@odu.edu), [sshetty@odu.edu](sshetty@odu.edu)  \narXiv :2607 .0 1583v2 [ cs .NI ] 7 Jul 2026  \nAbstract—Open Radio Access Network (O-RAN) architectures introduce programmable Near-Real-Time RAN Intelligent Controllers (Near-RT RICs) that support closed-loop control at 10 ms–1 s timescales through xApps. Although AI has been widely studied for RAN optimization, fewer works demonstrate measured AI inference embedded directly inside the Near-RT RIC software loop on a live testbed. This paper presents an AI-enabled network state classification xApp implemented on an OpenAirInterface (OAI) and FlexRIC testbed. The xApp is trained and evaluated on a structured synthetic dataset that emulates cross-layer RAN states using MAC, RLC, PDCP, GTP, and UE-count features. These results validate embedding and execution feasibility, rather than production-level generalization. Logistic regression and a shallow multi-layer perceptron (MLP) are exported as deterministic C inference modules and compiled into the xApp binary, eliminating external ML runtime dependencies. Measured inference latency is 1–5 µs for logistic regression and 10–25 µs for the MLP, while end-to-end service latency remains below 4 ms. A six-model comparison shows that supervised models cluster within a narrow 0.88–0.90 accuracy range, indicating that the LR–MLP similarity reflects the proxy problem structure rather than insufficient model exploration. Noise ablation, confusion matrix analysis, and CDF-based latency characterization show that both embedded models satisfy the 10 ms Near-RT budget for over 95% of projected loop executions. These results demonstrate that lightweight AI can be embedded within Near-RT RIC timing constraints while preserving deterministic execution. We also release RIC Workbench, a lightweight orchestration dashboard for reproducing the testbedon commodity hardware.  \nIndex Terms—O-RAN, Near-RT RIC, xApp, AI, machine learning, network state classification, logistic regression, MLP, OpenAirInterface, FlexRIC  \nI. INTRODUCTION  \nThe evolution of cellular networks toward 5G and beyond has increased the complexity of radio access network (RAN) management. Modern networks must support enhanced mobile broadband (eMBB), ultra-reliable low-latency communication (URLLC), and massive machine-type communication (mMTC), each with distinct and often conflicting requirements. These competing objectives make static, rule-based control mechanisms insufficient in dynamic environments where traffic demand, mobility, interference, and resource availability change rapidly [1]–[3] .  \nThe O-RAN Alliance addresses these challenges through an open, disaggregated, and programmable RAN architec-  \nture with standardized interfaces and two RAN Intelligent Controllers (RICs) . The Non-Real-Time RIC operates above one second and supports policy management, long-term optimization, and ML training through rApps. The Near-RealTime RIC operates at 10 ms–1 s timescales and hosts xApps for monitoring, inference, and closed-loop control over E2-connected RAN nodes [4]–[8] .  \nThis programmability creates an opportunity to embed AI directly into the RAN control loop. However, Near-RT RIC inference must satisfy strict timing constraints, consume live service model indications, avoid excessive runtime overhead, and integrate reproducibly wi","cbCaieTId6gADDfH","https://ap.wps.com/l/cbCaieTId6gADDfH","pdf",1542841,5,1,9,"English","en",105,"# Introduction\n# Background and Related Work\n## O-RAN Architecture and Near-RT RIC","[{\"question\":\"What is the main goal of the presented Near-RT RIC xApp approach?\",\"answer\":\"Embed lightweight AI inside a Near-RT RIC xApp with deterministic timing so inference latency and feasibility can be measured on a live OAI/FlexRIC testbed.\"},{\"question\":\"How is the AI model integrated into the xApp to avoid runtime overhead?\",\"answer\":\"Logistic regression and a shallow MLP are exported as deterministic C inference modules and compiled into the xApp binary, removing external machine-learning runtime dependencies.\"},{\"question\":\"What latency targets and results are reported for embedded inference and end-to-end service?\",\"answer\":\"Measured inference latency is 1–5 µs for logistic regression and 10–25 µs for the MLP, while end-to-end service latency remains below 4 ms. The study’s latency characterization shows the 10 ms Near-RT budget is met for over 95% of projected loop executions.\"}]",1784189126,23,{"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":87,"head_meta":89,"extra_data":91,"updated_unix":28},"enabling-real-time-ai-in-o-ran-deploying-and-measuring-ai-inside-a-near-rt-ric-xapp","",{"@graph":36,"@context":86},[37,54,69],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"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":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/enabling-real-time-ai-in-o-ran-deploying-and-measuring-ai-inside-a-near-rt-ric-xapp/83599/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":24,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-07-25","2026-07-16",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"What is the main goal of the presented Near-RT RIC xApp approach?","Question",{"text":76,"@type":77},"Embed lightweight AI inside a Near-RT RIC xApp with deterministic timing so inference latency and feasibility can be measured on a live OAI/FlexRIC testbed.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"How is the AI model integrated into the xApp to avoid runtime overhead?",{"text":81,"@type":77},"Logistic regression and a shallow MLP are exported as deterministic C inference modules and compiled into the xApp binary, removing external machine-learning runtime dependencies.",{"name":83,"@type":74,"acceptedAnswer":84},"What latency targets and results are reported for embedded inference and end-to-end service?",{"text":85,"@type":77},"Measured inference latency is 1–5 µs for logistic regression and 10–25 µs for the MLP, while end-to-end service latency remains below 4 ms. The study’s latency characterization shows the 10 ms Near-RT budget is met for over 95% of projected loop executions.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":93},[94,98,102,106,110,115,120,123,127,130,134],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":20,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},"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":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":22,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},"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":20,"slug":137},19,"General","general"]