Pith. sign in

REVIEW 5 major objections 8 minor 29 references

Proactive AI-and-RAN Workload Orchestration in O-RAN Architectures for 6G Networks

T0 review · 5 major / 8 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read The paper claims that a proactive O-RAN orchestrator can keep RAN demand fulfillment near 99 percent while reclaiming idle GPU capacity for AI workloads.

desk verdict CAORA is a plausible integration blueprint for AI-RAN, but the headline completion numbers rest on an unvalidated RNTI-to-MIG scaling and the evaluation needs work. read the letter →

arxiv 2507.09124 v1 pith:X25RIWJ6 submitted 2025-07-12 cs.NI

Syed Danial Ali Shah , Maryam Hafeez , Abdelaziz Salama , Syed Ali Raza Zaidi This is my paper · ORCID
classification cs.NI
keywords O-RANAI-RANconvergenceresourceorchestrationSoftActor-CriticLSTMforecastingMulti-InstanceGPU6Gnetworkstrace-drivensimulation
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper proposes CAORA, a four-layer O-RAN architecture in which real-time RAN functions and compute-heavy AI workloads share one GPU/MIG pool. Its claim is that a proactive orchestrator—an LSTM forecaster called SpikeAwareLSTM feeding a Soft Actor-Critic (SAC) reinforcement learning agent—can anticipate RAN demand spikes and shift idle capacity to AI tasks without degrading RAN service. Using real 5G traffic traces from three Barcelona districts, the authors report RAN demand fulfillment of 95.9–98.6 percent and AI workload completion of 78.5–81.5 percent, above fixed 50/50 and 70/30 baselines. The reason to care is economic: if CAORA is right, a 6G operator can buy one accelerated pool instead of dedicating underused GPUs to AI and separate capacity to RAN.

What carries the argument

The load-bearing machinery is the coupling of a SpikeAwareLSTM forecaster with a Soft Actor-Critic (SAC) agent inside an end-to-end orchestrator. SpikeAwareLSTM is a two-layer LSTM trained with a composite loss, mean-squared error for demand prediction plus binary cross-entropy for spike detection, whose internal spike probability $\hat{s}^{\mathrm{int}}_t$ down-weights forecasted demand during anomalies. The SAC agent's state includes current and predicted RAN/AI demands and the previous allocation, and its reward combines current QoS with anticipated QoS while imposing a super-linear overprovisioning penalty and a predictive contention penalty. The resource constraint allows reservations up to $R_{\max} + \sum_\delta \alpha_\delta \hat{\Delta}_{\mathrm{free}}(t+\delta)$, so predicted freed resources can be spent before they appear. The framework partitions the physical GPU into Multi-Instance GPU (MIG) slices, each an isolated compute instance, and exchanges KPIs between a monitoring xApp in the near-real-time RIC and the orchestrator over the Y1 interface.

What would settle it

Re-run the same trace-driven simulator with RAN demand derived from measured GPU/MIG utilization of real distributed-unit signal processing under the same RNTI loads, instead of the normalized RNTI-to-capacity rule of Eq. (12). If RAN completion falls materially below the reported 95.9–98.6 percent, or if peak demand exceeds the 21-MIG capacity, then the near-99 percent claim does not survive.

Watch

Extended reading notes

Core claim

The paper's central claim is that RAN and AI workloads do not need separate dedicated GPU pools. CAORA ties a monitoring xApp inside the near-real-time RIC to an end-to-end orchestrator over the Y1 interface; the orchestrator's SpikeAwareLSTM forecasts demand and flags spikes, and the SAC agent translates those signals into rate-limited, priority-scaled MIG allocations that first protect RAN and lend spare capacity to AI. In trace-driven simulations over three Barcelona 5G traces, the authors report RAN demand fulfillment of 95.9–98.6 percent and AI workload completion of 78.5–81.5 percent, with both above the fixed 50/50 balanced and 70/30 RAN-priority baselines. The paper presents CAORA as a blueprint for a converged 6G platform in which one shared accelerated pool carries radio and AI work.

Load-bearing premise

The load-bearing premise is that RNTI counts (radio network temporary identifiers, a proxy for active users), after being min-max normalized to [0,1], linearly represent the fraction of the 21-MIG GPU pool that the RAN actually needs, with the busiest observed count set equal to full capacity; if that mapping is wrong, the reported fulfillment percentages change.

Editorial extensions

If this is right

  • If CAORA is right, the Y1 interface can carry radio analytics from a near-real-time RIC to an external orchestrator fast enough for closed-loop AI/RAN co-allocation.
  • Operators could run AI training, inference, or generative-AI workloads on RAN GPU capacity during off-peak hours while keeping RAN demand fulfillment above 95 percent in the tested scenarios.
  • In the football-match trace, the reported 98.3 percent RAN completion with 81.5 percent AI completion would come from proactive spike-aware allocation rather than from fixed service priorities.
  • Since the results hold across stadium-spike, nightlife-cyclical, and residential-steady demand traces, the same control loop could serve as a common resource manager for a city-scale converged 6G platform.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The RNTI-to-MIG demand model in Eq. (12) is the least supported link in the chain: it scales the busiest observed RNTI count to exactly $R_{\max}$, so a perfectly accurate forecaster would satisfy demand almost by construction. I would not transfer the 99 percent figure to a deployment without calibrating that mapping against real GPU utilization.
  • A clean ablation, replacing SpikeAwareLSTM with a vanilla LSTM while keeping the SAC agent unchanged, would isolate whether joint spike detection is what produces the adaptability gain; the paper does not perform that separation.
  • The simulations cover a single orchestration site under one SAC agent; a multi-cell or multi-RIC extension would face Y1 monitoring latency, distributed state, and competing allocations, none of which the current results address.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

5 major / 8 minor

Summary. The paper proposes CAORA, a four-layer O-RAN-compliant architecture for sharing GPU/MIG resources between RAN and AI workloads. The E2E orchestrator consumes radio analytics from NRT-RIC xApps via the Y1 interface, uses a SpikeAwareLSTM for multi-horizon demand forecasting and spike detection, and employs a Soft Actor-Critic (SAC) agent to adjust MIG allocations under rate and capacity constraints. The authors evaluate the framework in trace-driven simulations using Barcelona 5G RNTI traces from three areas (Les Corts-Camp Nou, Poble Sec, El Born) plus a synthetic periodic AI-demand model, comparing against balanced (50/50) and RAN-priority (70/30) static baselines on completion rate and adaptability. They report RAN completion rates of 95.9-99.3% and AI completion rates of about 81-90%, concluding that predictive LSTM-SAC orchestration is a viable blueprint for AI-RAN converged 6G systems.

Significance. The architectural contribution is timely: integrating monitoring xApps with an E2E orchestrator over the Y1 interface and dynamically partitioning MIG resources between RAN and AI workloads directly addresses an open problem in AI-RAN convergence. The SpikeAwareLSTM multi-task design is a reasonable mechanism for coupling forecasting with spike detection. If the demand model were independently calibrated, the paper would offer a useful blueprint for future 6G resource management. However, the central quantitative claim rests on an unvalidated RNTI-to-MIG scaling, and the evaluation lacks the ablations and statistical reporting needed to attribute the observed gains to proactive forecasting or SAC adaptation. The work is therefore of moderate-to-good significance but currently does not support 'near 99% fulfillment' as a measured system-level property.

major comments (5)
  1. [Section VII-A, Eq. (12)] The RAN demand proxy d_RAN(t) = (RNTI_t - min(RNTI))/(max(RNTI) - min(RNTI) + epsilon) is min-max normalized and then interpreted as a fraction of R_max = 21 MIGs. This makes the peak RNTI of each trace coincide exactly with full system capacity by construction. No evidence is provided that RNTI counts (numbers of active user identifiers) map linearly, or at all, to GPU/MIG compute demand of an O-DU. Because the optimization objective, reward, and baselines all consume this same normalized demand, the reported completion rates and the CAORA-vs-baseline gap may be artifacts of the chosen denominator rather than measured properties of the architecture. Please calibrate the mapping on an O-DU/MIG profiling study, provide a sensitivity analysis over plausible scaling laws, and state explicitly that the completion-rate numbers are conditional on this proxy.
  2. [Section IV-D2, Eq. (10); Algorithm 1, line 10] The reward definitions are inconsistent. Eq. (10) defines R'_t as sum_x [C_x(t)/d_x(t) + mu * C_hat_x(t+1)/d_x(t)] minus penalty terms, with C_hat_x(t+1) never defined, while Algorithm 1 uses p_x(t)*C_x(t)/d_x(t) + mu * min(p_x(t)*r_x(t), d_hat_x(t+1))/d_x(t). The priority weighting p_x appears only in the pseudocode, and C_hat_x never appears in the state or allocation equations. Since the SAC agent is trained on this reward, the discrepancy makes the actual optimized objective ambiguous and affects reproducibility.
  3. [Section VII-C.2, Figures 9-11] The attribution of the performance gain to proactive LSTM forecasting and SAC adaptation is not supported by the present experiment set. The baselines are static (50/50 and 70/30) with no learning or lookahead, and there is no ablation comparing SAC with and without the LSTM forecasts, or against a reactive adaptive policy. Add such ablations, or soften the conclusion that 'predictive orchestration significantly improves' performance.
  4. [Section VII-B, Evaluation Metrics; Figures 9-12] The evaluation reports only point estimates. No error bars, confidence intervals, number of random seeds, or variance across training runs are provided, so it is impossible to assess whether the reported differences are statistically significant. The text calls the adaptability result 'statistically significant' without performing or citing a significance test. Add repeated-run statistics and a significance test for the completion-rate and adaptability comparisons.
  5. [Section IV-B, Eqs. (4)-(5); Section VII-A] Several load-bearing quantities are under-specified. In Eq. (4), Delta(k,tau) is used but never defined, and the relationship between C_pred, Delta_total, r(k,t), and d_x(t) is not established, making it impossible to verify the dimensional consistency of the contention factor and the freed-resource prediction in Eq. (5). In addition, the numerical values of the reward weights mu, lambda, eta, the future discount beta, the rate limits nu_max_x, and the borrowing coefficients alpha_delta in constraint (7) are not reported in Table III or the text. These parameters determine the agent's behavior, so their absence blocks reproduction.
minor comments (8)
  1. [Abstract; Section VII-C.2, Figure 11] The abstract claims 'near 99% fulfillment of RAN demands,' but the Poble Sec scenario reports a RAN completion rate of 95.9%; please rephrase to report the actual range accurately.
  2. [Section VII-A, Eq. (13)] Eq. (13) is not a full-wave rectified sine; it is a shifted sine with a DC offset. Either correct the formula or change the terminology to avoid a misleading description.
  3. [Table III] Table III lists R_max = 21 MIGs with the MIG profile 1g.5gb providing 7 MIG instances per A100; clarify whether 21 MIGs corresponds to three GPUs and describe how discrete MIG partitioning is modeled in the simulator.
  4. [Algorithm 1; Section V-A] Algorithm 1 outputs {Delta_r_RAN, Delta_r_AI, nu_RAN, nu_AI}, while Section V-A states |A| = 2 and Section VII-A describes actions as {Delta_r_RAN, Delta_r_AI}; align the action-space definition across the paper.
  5. [Figures 3-8] In the submitted PDF, Figures 3-8 contain corrupted font encoding in the axis labels and legends, so the forecast and allocation plots cannot be visually verified; please regenerate the figures.
  6. [Section IV-B, Eq. (4)] The text describes the demand fusion as 'Bayesian averaging,' but Eq. (4) uses a simple weighted average of current and predicted demands; rename this to avoid an unjustified Bayesian claim.
  7. [Table II; Section VI] The paper says the LSTM is trained and tested on separate CSV files, but it does not report the exact evaluation day for each trace or confirm that no training data overlaps the test period; provide this information.
  8. [General] The paper does not state whether code or processed data will be released; given that the raw traces are public (refs. [27], [28]), a reproducibility statement would strengthen the work.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity found; the evaluation metric is aligned with the RL reward by design, but the reported results are empirical simulation outcomes and the main weak point (Eq. 12 scaling) is an unvalidated assumption, not a circular derivation.

full rationale

The paper's derivation chain is self-contained: real RNTI traces are normalized in Eq. (12), used to define RAN demand, fed to an LSTM forecaster and SAC allocator, and then evaluated by completion rates in Figures 9-11. The completion-rate metric (Section VII-B-1) does coincide with the first term of the reward function in Eq. (10), but this is standard reward-objective alignment in RL, not a case of the result being identical to the input by construction. The SAC agent is not mathematically guaranteed to satisfy demand, and the static baselines demonstrably achieve lower completion, so the comparison has independent content. The RNTI-to-MIG conversion in Eq. (12) is an externally unvalidated modeling assumption rather than a fitted-input-called-prediction step: it shapes what 'demand' means, but it does not by itself force the reported policy outcomes or the gap versus baselines. Self-citations ([2], [5], [7], [8]) appear in background and related-work contexts and are not load-bearing for the architecture's correctness or for any uniqueness claim. No self-citation chain, imported uniqueness theorem, ansatz smuggled via citation, or renaming of a known result was found. Concerns about the physical validity of the RNTI-to-GPU mapping belong to correctness risk, not circularity.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

The paper introduces no new physical entities, forces, or conserved quantities. It does introduce an architectural design (CAORA) and a model variant (SpikeAwareLSTM), but these are engineering constructs, not invented entities in the graviton sense. The load-bearing assumptions are the demand-scaling proxy and the synthetic AI workload.

free parameters (4)
  • Reward weights (μ, λ, η) and horizon (H, αδ) = not reported
    These control the tradeoff between anticipation, overprovisioning penalty, and contention; their values affect the reported completion rates but are not listed in Table III.
  • Spike threshold τ_spike = 90th percentile
    Chosen by hand; changes LSTM spike labels and thus the forecast-gated demand signals in Eq. 4.
  • SAC temperature α = 0.2
    Standard RL hyperparameter from Table III, not grounded in the problem or tuned against a validation objective.
  • Allocation rate limit νmax_x = not reported
    Constraint (8) depends on νmax, but no value is given; it directly shapes how quickly the SAC can adapt.
assumptions (4)
  • domain assumption Normalized RNTI count represents RAN resource demand
    Eq. 12 maps min-max scaled RNTI to a fraction of Rmax; the entire simulation depends on this being a valid proxy for actual GPU/MIG demand.
  • ad hoc to paper AI demand follows a deterministic full-wave rectified sinusoid
    Eq. 13; no real AI trace is used, so all results for AI support rest on this synthetic pattern.
  • domain assumption MIG resources are a single divisible pool with no interference
    The simulation treats 21 MIGs as fungible capacity and ignores GPU memory/bandwidth constraints and interference between co-located instances.
  • domain assumption LSTM predictions and completion probabilities are sufficiently accurate for proactive allocation
    The SAC relies on ĥ_demand and Δ_free forecasts; the paper does not analyze how prediction errors propagate into allocation decisions or completion rates.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Proactive AI-and-RAN Workload Orchestration in O-RAN Architectures for 6G Networks." pith.science (2026). https://pith.science/paper/X25RIWJ6

@misc{pith2026250709124,
  author       = {Pith},
  title        = {Pith review of: Proactive AI-and-RAN Workload Orchestration in O-RAN Architectures for 6G Networks},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/X25RIWJ6}},
  note         = {Machine review of arXiv:2507.09124}
}
read the original abstract

The vision of AI-RAN convergence, as advocated by the AI-RAN Alliance, aims to unlock a unified 6G platform capable of seamlessly supporting AI and RAN workloads over shared infrastructure. However, the architectural framework and intelligent resource orchestration strategies necessary to realize this vision remain largely unexplored. In this paper, we propose a Converged AI-and-ORAN Architectural (CAORA) framework based on O-RAN specifications, enabling the dynamic coexistence of real-time RAN and computationally intensive AI workloads. We design custom xApps within the Near-Real-Time RAN Intelligent Controller (NRT-RIC) to monitor RAN KPIs and expose radio analytics to an End-to-End (E2E) orchestrator via the recently introduced Y1 interface. The orchestrator incorporates workload forecasting and anomaly detection modules, augmenting a Soft Actor-Critic (SAC) reinforcement learning agent that proactively manages resource allocation, including Multi-Instance GPU (MIG) partitioning. Using real-world 5G traffic traces from Barcelona, our trace-driven simulations demonstrate that CAORA achieves near 99\% fulfillment of RAN demands, supports dynamic AI workloads, and maximizes infrastructure utilization even under highly dynamic conditions. Our results reveal that predictive orchestration significantly improves system adaptability, resource efficiency, and service continuity, offering a viable blueprint for future AI-and-RAN converged 6G systems.

Figures

Figures reproduced from arXiv: 2507.09124 by the authors.

Figure 1
Figure 1. The proposed converged AI-and-ORAN architectural framework [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. LSTM-SAC resource orchestration framework for AI-and-RAN systems [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Resource demand prediction for Les Corts-Camp Nou [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figures from the paper (7 more)
Figure 4
Figure 4. Figure 4: Resource demand prediction for Poble Sec [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]
Figure 5
Figure 5. Figure 5: Resource demand prediction for El Born VII. PERFORMANCE EVALUATION A. Experimental Setup and Implementation The experimental evaluation of the proposed framework was conducted using Python, leveraging the O-RAN 7.2x split [1]. We simulated realistic O-RAN infrastructur…
Figure 6
Figure 6. Figure 6: Resource allocation during a football match (LCCM), prioritizing [PITH_FULL_IMAGE:figures/full_fig_p010_6.png]
Figure 7
Figure 7. Figure 7: Dynamic resource balancing for touristic and nightlife-driven demand [PITH_FULL_IMAGE:figures/full_fig_p010_7.png]
Figure 9
Figure 9. Figure 9: Completion rate comparison in Les Corts Proposed SAC Balanced RAN Priority 0 20 40 60 80 100 Completion Rate (%) 98.6% 97.4% 99.3% 81.0% 70.5% 56.1% 90.1% 84.5% 78.5% RAN (%) AI (%) Combined (%) [PITH_FULL_IMAGE:figures/full_fig_p011_9.png]
Figure 10
Figure 10. Figure 10: Completion rate comparison in El Born failing to reallocate unused RAN resources. The proposed ap￾proach maintained strong performance across varying demand scenarios, particularly excelling during rapid demand transi￾tions where static strategies struggled with delay…
Figure 11
Figure 11. Figure 11: Completion rate comparison in Poble Sec Les Corts Poble Sec El Born 0 20 40 60 80 100 Adaptability Ratio (%) 90.1% 86.1% 90.6% 70.4% 74.7% 72.5% 59.2% 69.4% 62.3% Proposed SAC Balanced RAN Priority [PITH_FULL_IMAGE:figures/full_fig_p011_11.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

29 extracted references · 26 canonical work pages

  1. [5]

    The Interplay of AI-and-RAN: Dynamic Resource Allocation for Converged 6G Platform,

    S. D. A. Shah, Z. Nezami, M. Hafeez, and S. A. R. Zaidi, “The Interplay of AI-and-RAN: Dynamic Resource Allocation for Converged 6G Platform,” arXiv preprint arXiv:2503.07420 , 2025

    arXiv 2025

  2. [1]

    Under- standing O-RAN: Architecture, Interfaces, Algorithms, Security, and Research Challenges,

    M. Polese, L. Bonati, S. D’oro, S. Basagni, and T. Melodia, “Under- standing O-RAN: Architecture, Interfaces, Algorithms, Security, and Research Challenges,” IEEE Communications Surveys & Tutorials , vol. 25, no. 2, pp. 1376–1411, 2023

    work page 2023

  3. [2]

    Dynamic AI-Driven Network Slicing with O-RAN for Continuous Connectivity in Connected Vehicles and Onboard Consumer Electronics,

    S. D. A. Shah, A. K. Bashir, Y . D. Al-Otaibi, M. M. Al Dabel, and F. Ali, “Dynamic AI-Driven Network Slicing with O-RAN for Continuous Connectivity in Connected Vehicles and Onboard Consumer Electronics,” IEEE Transactions on Consumer Electronics, vol. 71, no. 1, pp. 720–733, Feb. 2025

    work page 2025

  4. [3]

    AI-RAN Alliance Vision and Mission White Paper,

    “AI-RAN Alliance Vision and Mission White Paper,” AI-RAN Al- liance, Tech. Rep., 2024, available online: https://ai-ran.org/wp-content/ uploads/2024/12/AI-RAN_Alliance_Whitepaper.pdf. Accessed: Jan. 8, 2025

    work page 2024

  5. [4]

    Integrating AI/ML in Open-RAN: Overcoming Challenges and Seizing Opportunities,

    “Integrating AI/ML in Open-RAN: Overcoming Challenges and Seizing Opportunities,” AI-RAN Alliance, Tech. Rep., 2024, available online: https://ai-ran.org/publications/. Accessed: Jan. 5, 2025

    work page 2024

  6. [6]

    Enhancing vehicular networks with hierarchical O-RAN slicing and federated DRL,

    B. Hazarika, P. Saikia, K. Singh, and C.-P. Li, “Enhancing vehicular networks with hierarchical O-RAN slicing and federated DRL,” IEEE Transactions on Green Communications and Networking , vol. 8, no. 3, pp. 1099–1117, Sep. 2024

    work page 2024

  7. [7]

    Artificial Intelligence-Defined Wireless Networking for Computational Offloading and Resource Allocation in Edge Computing Networks,

    S. D. A. Shah, M. Gregory, F. Bouhafs, and F. Den Hartog, “Artificial Intelligence-Defined Wireless Networking for Computational Offloading and Resource Allocation in Edge Computing Networks,” IEEE Open Journal of the Communications Society , vol. 5, pp. 2039–2057, Mar. 2024

    work page 2024

  8. [8]

    A Distributed Control Plane Architecture for Handover Management in MEC-enabled Vehicular Networks,

    S. D. A. Shah, M. A. Gregory, and S. Li, “A Distributed Control Plane Architecture for Handover Management in MEC-enabled Vehicular Networks,” in 2021 31st International Telecommunication Networks and Applications Conference (ITNAC). IEEE, 2021, pp. 188–191

    work page 2021

Show all 29 references
  1. [9]

    Toward Sustainable O-RAN Deployment: An In- Depth Analysis of Power Consumption,

    G. Baldini, R. Bolla, R. Bruschi, A. Carrega, F. Davoli, C. Lombardo, and R. Rabbani, “Toward Sustainable O-RAN Deployment: An In- Depth Analysis of Power Consumption,” IEEE Transactions on Green Communications and Networking , 2024

    2024

  2. [10]

    Meta Reinforcement Learning Approach for Adaptive Resource Optimization in O-RAN,

    F. Lotfi and F. Afghah, “Meta Reinforcement Learning Approach for Adaptive Resource Optimization in O-RAN,” arXiv preprint arXiv:2410.03737, 2024

    2024 arXiv

  3. [11]

    Adaptive Resource Allo- cation for Virtualized Base Stations in O-RAN with Online Learning,

    M. Kalntis, G. Iosifidis, and F. A. Kuipers, “Adaptive Resource Allo- cation for Virtualized Base Stations in O-RAN with Online Learning,” IEEE Transactions on Communications , 2024

    2024

  4. [12]

    Open AI Cellular (OAIC): An Open Source 5G O-RAN Testbed for Design and Testing of AI-Based RAN Management Algorithms,

    P. S. Upadhyaya, N. Tripathi, J. Gaeddert, and J. H. Reed, “Open AI Cellular (OAIC): An Open Source 5G O-RAN Testbed for Design and Testing of AI-Based RAN Management Algorithms,” IEEE Network , vol. 37, no. 5, pp. 7–15, Sep. 2023

    2023

  5. [13]

    An Open, Programmable, Multi-Vendor 5G O-RAN Testbed with NVIDIA ARC and OpenAirInterface,

    D. Villa, I. Khan, F. Kaltenberger, N. Hedberg, R. S. Da Silva, A. Kelkar, C. Dick, S. Basagni, J. M. Jornet, T. Melodia et al. , “An Open, Programmable, Multi-Vendor 5G O-RAN Testbed with NVIDIA ARC and OpenAirInterface,” in Proc. IEEE Conference on Computer Communications Wo...

    2024

  6. [14]

    The OpenAir- Interface 5G New Radio Implementation: Current Status and Roadmap,

    F. Kaltenberger, G. De Souza, R. Knopp, and H. Wang, “The OpenAir- Interface 5G New Radio Implementation: Current Status and Roadmap,” in Proc. 23rd International ITG Workshop on Smart Antennas (WSA) . VDE, 2019, pp. 1–5

    2019

  7. [15]

    ColO- RAN: Developing Machine Learning-Based xApps for Open RAN Closed-Loop Control on Programmable Experimental Platforms,

    M. Polese, L. Bonati, S. D’Oro, S. Basagni, and T. Melodia, “ColO- RAN: Developing Machine Learning-Based xApps for Open RAN Closed-Loop Control on Programmable Experimental Platforms,” IEEE Transactions on Mobile Computing , vol. 22, no. 10, pp. 5787–5800, 2022

    2022

  8. [16]

    Federated Deep Rein- forcement Learning for Resource Allocation in O-RAN Slicing,

    H. Zhang, H. Zhou, and M. Erol-Kantarci, “Federated Deep Rein- forcement Learning for Resource Allocation in O-RAN Slicing,” in GLOBECOM 2022-2022 IEEE Global Communications Conference . IEEE, 2022, pp. 958–963

    2022

  9. [17]

    Open RAN LSTM Traffic Prediction and Slice Management using Deep Reinforcement Learning,

    F. Lotfi and F. Afghah, “Open RAN LSTM Traffic Prediction and Slice Management using Deep Reinforcement Learning,” in 2023 57th Asilomar Conference on Signals, Systems, and Computers. IEEE, 2023, pp. 646–650

    2023

  10. [18]

    Delay-Sensitive Resource Allocation for IoT Systems in 5G O-RAN Networks,

    R. Firouzi and R. Rahmani, “Delay-Sensitive Resource Allocation for IoT Systems in 5G O-RAN Networks,” Internet of Things , vol. 26, p. 101131, 2024

    2024

  11. [19]

    Actor-Critic Network for O-RAN Resource Allocation: xApp Design, Deployment, and Analysis,

    M. Kouchaki and V . Marojevic, “Actor-Critic Network for O-RAN Resource Allocation: xApp Design, Deployment, and Analysis,” in 2022 IEEE Globecom Workshops (GC Wkshps) . IEEE, 2022, pp. 968–973

    2022

  12. [20]

    REAL: Re- inforcement Learning-Enabled xApps for Experimental Closed-Loop Optimization in O-RAN with OSC RIC and srsRAN,

    R. Barker, A. E. Dorcheh, T. Seyfi, and F. Afghah, “REAL: Re- inforcement Learning-Enabled xApps for Experimental Closed-Loop Optimization in O-RAN with OSC RIC and srsRAN,” arXiv preprint arXiv:2502.00715, 2025

    2025

  13. [21]

    Peer- to-Peer Federated Learning Based Anomaly Detection for Open Radio Access Networks,

    D. Attanayaka, P. Porambage, M. Liyanage, and M. Ylianttila, “Peer- to-Peer Federated Learning Based Anomaly Detection for Open Radio Access Networks,” in ICC 2023-IEEE International Conference on Communications. IEEE, 2023, pp. 5464–5470

    2023

  14. [22]

    Miti- gating xApp Conflicts for Efficient Network Slicing in 6G O-RAN: A Graph Convolutional-Based Attention Network Approach,

    S. Bakri, I. Dey, H. Siljak, M. Ruffini, and N. Marchetti, “Miti- gating xApp Conflicts for Efficient Network Slicing in 6G O-RAN: A Graph Convolutional-Based Attention Network Approach,” arXiv preprint arXiv:2504.17590, 2025

    2025 arXiv

  15. [23]

    Federated Learning for Anomaly Detection in Open RAN: Security Architecture Within a Digital Twin,

    Y . Rumesh, D. Attanayaka, P. Porambage, J. Pinola, J. Groen, and K. Chowdhury, “Federated Learning for Anomaly Detection in Open RAN: Security Architecture Within a Digital Twin,” in 2024 Joint European Conference on Networks and Communications & 6G Summit (EuCNC/6G Summit). ...

    2024

  16. [24]

    H. A. Tahir, W. Alayed, W. u. Hassan, and T. D. Do, “Optimizing Open Radio Access Network Systems with LLAMA V2 for Enhanced Mobile Broadband, Ultra-Reliable Low-Latency Communications, and Massive Machine-Type Communications: A Framework for Efficient Network Slicing and Real...

    2024

  17. [25]

    LLM-xApp: A Large Language Model Empowered Radio Resource Management xApp for 5G O-RAN,

    X. Wu, J. Farooq, Y . Wang, and J. Chen, “LLM-xApp: A Large Language Model Empowered Radio Resource Management xApp for 5G O-RAN,” in Symposium on Networks and Distributed Systems Security (NDSS), Workshop on Security and Privacy of Next-Generation Networks (FutureG 2025), San...

    2025

  18. [26]

    NVIDIA Multi-Instance GPU User Guide,

    “NVIDIA Multi-Instance GPU User Guide,” NVIDIA Corporation, Tech. Rep., 2025, available online: https://docs.nvidia.com/datacenter/ tesla/mig-user-guide/index.html. Accessed: Jan. 10, 2025

    2025

  19. [27]

    Federated-time-series-forecasting,

    vperifan, “Federated-time-series-forecasting,” accessed: Mar. 21, 2025. [Online]. Available: https://github.com/vperifan/ Federated-Time-Series-Forecasting/tree/main/dataset

    2025

  20. [28]

    Federated Learning for 5G Base Station Traffic Forecasting,

    V . Perifanis, N. Pavlidis, R.-A. Koutsiamanis, and P. S. Efraimidis, “Federated Learning for 5G Base Station Traffic Forecasting,” Computer Networks, vol. 235, p. 109950, 2023

    2023

  21. [29]

    How ACK Edge Solves Challenges in Elasticity for LLM Inference Services,

    L. Chen, “How ACK Edge Solves Challenges in Elasticity for LLM Inference Services,” Alibaba Cloud, Tech. Rep., 2024, available online: https://www.alibabacloud.com/blog/ how-ack-edge-solves-challenges-in-elasticity-for-llm-inference-services_ 602236. Accessed: Mar. 01, 2025. S...

    2024

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.