Enabling Real-Time AI in O-RAN: Deploying andMeasuring AI Inside a Near-RT RIC xApp

2026-07-01

AI Summary

This work addresses the challenge of deploying AI inference under stringent 10-ms latency constraints in O-RAN near-real-time RIC by proposing an embedded lightweight AI xApp. The approach exports logistic regression and shallow MLP models as deterministic C code, which is directly compiled into the xApp binary—eliminating dependencies on external machine learning runtimes. A synthetic dataset is constructed using cross-layer features including MAC, RLC, PDCP, GTP, and UE count. Experimental evaluation on OpenAirInterface and FlexRIC demonstrates inference latencies of only 1–25 microseconds, end-to-end service latency below 4 ms, model accuracy between 0.88 and 0.90, and over 95% of execution cycles meeting the 10-ms deadline. This study presents the first validation of deterministic embedded AI within a near-real-time RIC closed loop and releases the RIC Workbench to enable reproducible research.

Abstract

Open Radio Access Network (O-RAN) architectures introduce programmable Near-Real-Time RAN Intelligent Controllers (Near-RT RICs) that support closed-loop control through xApps at timescales from 10 ms to 1 s. Although AI has been widely studied for RAN optimization, fewer works demonstrate measured AI inference embedded directly within 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. The results validate embedding and execution feasibility rather than production-level generalization. Logistic regression and a shallow multilayer perceptron (MLP) are exported as deterministic C inference modules and compiled into the xApp binary, eliminating external machine-learning runtime dependencies. Measured inference latency is 1 to 5 microseconds for logistic regression and 10 to 25 microseconds for the MLP, while end-to-end service latency remains below 4 ms. A six-model comparison shows that supervised models achieve similar accuracy, ranging from 0.88 to 0.90, indicating that LR and MLP similarity reflects the proxy problem structure rather than limited 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 more than 95% of projected loop executions. These results demonstrate that lightweight AI can operate within Near-RT RIC timing constraints while preserving deterministic execution. We also release RIC Workbench, a lightweight orchestration dashboard for reproducing the testbed on commodity hardware.

Problem

• O-RAN
• Near-RT RIC
• AI inference
• real-time AI
• xApp

Innovation

• Near-RT RIC
• xApp
• embedded AI inference
• deterministic execution
• O-RAN