One of the major transformative factors in 6G will be the integration of Artificial Intelligence (AI) to become a native part of Radio Access Network (RAN). While most physical-layer AI features have so far been evaluated in isolation using link-level simulations, their combined behavior in a realistic multi-cell, multi-UE deployment has remained largely unexplored. In this paper, we present system-level performance results when multiple uplink AI features are enabled together, achieved by integrating accurate link-level and system-level simulators. To infer state-of-the-art deep-learning-aided Multiple Input Multiple Output (MIMO) receivers under the dynamic allocations produced by a realistic uplink scheduler, we propose a mirrored data augmentation method that decouples receiver performance from scheduled allocation size. In addition to these Physical Layer (PHY) receiver features, we combine several recent advances in deep reinforcement learning to train uplink power control and link adaptation that outperform a heuristic baseline and further boost the gains obtainable from the AI receiver alone. The system-level results show that the combined AI features improve the mean uplink user throughput by roughly 27% compared to a non-AI baseline, confirming that the individual PHY and Medium Access Control (MAC) AI features provide complementary gains when deployed jointly.
Figures & tables
Fig. 1 : The considered DeepRx AI receiver model architecture [ 4 ] .
Parameter
DeepLA
DeepTPC
Learning rate
0.0003
0.0001
Weight initialization
uniform [-0.2, 0.2]
uniform [-0.2, 0.2]
Discount factor
0
0.95
Target entropy C
0.5
0.9
Mini-batch size
64
64
Hidden layers
2
2
TABLE I : SACD Hyper-parameters
Parameter
Value
Deployment scenario
3GPP Urban Macro, 500m ISD [ 20 ] ,
3D channel model in [ 21 ]
Number of BSs
21 (7 sites, 3 sectors/site)
Number of UEs
210
UE velocity
30 km/h
BS antenna config
(M, N, P, Mg, Ng, Mp, Np) =
TABLE II : Simulation configuration.
Fig. 2 : Mean performance scores of DeepLA and DeepTPC training with ±2σ range across base stations. DeepLA’s performance is visualized with kbits/RB scoring while DeepTPC’s performance score is effective SINR driven.
Fig. 3 : Full-buffer uplink UPT when individual AI features are enabled on top of the no- AI baseline (features are not cumulatively stacked), except all-in AI , which enables all features jointly. The DeepRx Detector has the single largest individual impact. Note that DeepLA and DeepTPC are trained to be used with DeepRx, while here used with plain LMMSE-IRC.
Fig. 4 : Full-buffer uplink UPT under feature ablation, where one AI feature at a time is removed from the all-in AI configuration. Comparing against the all-in AI and no- AI bounds isolates the marginal contribution of each feature. The DeepRx Detector is the most significant feature and partially compensates for the removal of the DeepRx Denoiser; viewed in this light, the latter proves to be the least significant one.
Fig. 5 : User perceived throughput with FTP Model 3 traffic. The importance of the AI features remains, or even slightly increases, with bursty traffic at ∼50 % RB load. Removing all AI features results in a 24% drop in mean throughput, which corresponds to the 32% benefit provided by AI .
Cellular research and development (R&D) is throttled by six structural processes that each consume months of manual engineering work per iteration: (i) synthesizing new features from standards or research papers into production code; (ii) conformance and interoperability testing; (iii) hardening against field anomalies and diverse deployment environments; (iv) data-driven optimization of network functionalities; (v) discovering and prototyping novel waveforms, functionalities, and capabilities for future standards; and (vi) securing the stack against vulnerabilities. Although Large Language Models (LLMs) have compressed comparable R&D work in general software engineering from days to minutes, their known pitfalls worsen on Radio Access Network (RAN) use cases: they hallucinate Application Programming Interfaces (APIs) and mis-read specifications, which kills interoperability of RAN components at the first mistake, and they heavily rely on simulations for designing algorithms, which is notorious for breaking when transferred to real hardware. To address these challenges, we present GENESIS, an agentic Artificial Intelligence (AI) framework that converts intents (e.g., a specification clause, a telemetry anomaly, or a research hypothesis) into solutions validated with over-the-air experiments, fed back into a persistent knowledge base. GENESIS is built on three composable primitives (agents, skills, hooks) and a knowledge layer (SYNAPSE) that doubles as the source of ground truth and the recipient of every artifact the framework produces, making capabilities compound across runs.
Tamerlan Aghayev, Maxime Elkael, Michele Polese +11
Institute for Intelligent Networked Systems at Northeastern University, Boston, MA.
Outer-loop link adaptation (OLLA) is widely deployed in 5G NR to track channel variations, yet its reliance on first-order, single-bit feedback degrades performance significantly under high-mobility and fast-varying channels. This paper presents LOLLA (Learned Outer-Loop Link Adaptation), a deep reinforcement learning framework that replaces the conventional OLLA staircase with a learned, continuous SINR offset conditioned on rich PHY/MAC telemetry inaccessible to OLLA. The offset modulates the SINR-to-MCS lookup table, preserving 3GPP-compliant MCS selection and provably subsuming the conventional OLLA update rule. A Proximal Policy Optimization (PPO) policy trained under a Lagrangian block error rate (BLER) constraint automatically enforces tunable reliability targets from 1% to 15% without manual penalty calibration. The framework is realized as the first closed-loop AI-native control dApp on a GPU-accelerated 5G NR stack, achieving end-to-end control latencies under 500 microseconds. Evaluations under 3GPP TDL channel models demonstrate 15% to 92% throughput gains over OLLA across Doppler frequencies up to 400 Hz, while attaining a Pareto frontier that strictly dominates OLLA across all evaluated reliability targets. The learned policy generalizes to unseen channel models and scales to eight concurrent UEs under shared-resource scheduling. In the uplink formulation, the gNB directly observes decoding outcomes, enabling simulation-to-deployment parity.
Industrial 6G networks require ultra-reliable, low-latency, and energy-efficient connectivity in dynamic and blockage-prone environments, where conventional terrestrial deployments often fail to ensure stable coverage. Hence, in this paper, we propose a RIS-enabled Open-RAN framework for integrated terrestrial/non-terrestrial (TN/NTN) industrial 6G networks, in which UAVs-mounted reconfigurable intelligent surfaces (RISs) cooperate with ground radio units and a high-altitude platform (HAP) to enhance connectivity for dense industrial IoT devices. Owing to the high dimensionality and strong coupling among decision variables, conventional optimization techniques become computationally intractable. To overcome this limitation, the joint optimization problem of data rates, latency, and energy consumptions is formulated as a decentralized partially observable Markov decision process (Dec-POMDP) and solved using a multi-agent deep reinforcement learning framework. Simulation results show improvements of up to 75% in data rate, 25% latency reduction, and 16% energy savings compared with state-of-the-art learning-based and non-RIS baselines, demonstrating the effectiveness of RIS-assisted Open-RAN intelligence for industrial 6G networks.
Marwan Dhuheir, Thang X. Vu, Symeon Chatzinotas
The Interdisciplinary Centre for Security, Reliability and Trust (SnT), University of Luxembourg, Luxembourg.