Shape the Future of AI-Powered RAN

Shape the Future of AI-Powered RAN – Submit Your Innovation Proposal by May 30!

The AI-RAN Alliance invites cutting-edge proposals focused on the transformative potential of Artificial Intelligence (AI) in Radio Access Networks (RAN). As the telecommunications industry moves towards more intelligent and efficient systems, AI integration into RAN is no longer optional—it is essential.

AI integration promises to unlock lower operational costs and new monetization avenues. The AI-RAN Alliance is at the forefront of this evolution, leading efforts in testing and benchmarking AI/ML algorithms within RAN environments.

We’re calling on everyone passionate about telecom and AI — from academia, startups, research labs, and industry — to submit their ideas. Proposals should aim to solve real-world AI-RAN challenges with innovative and impactful solutions.

Topics of Interest (click on each to learn more)

See What to Include/Submission Guidelines

Submission Deadline: May 30, 2025

Please note: If you encounter issues with this Google submittal form, please submit your proposal directly to: [email protected]

Join an AI-RAN Alliance Public Info Session – May 8

Learn more about the AI-RAN Alliance Call for Innovation during one of our public information sessions on Wednesday, 8 May. We’ll be hosting two sessions to accommodate different time zones:

– 7:00 AM PT | 10:00 AM ET | 4:00 PM CEST  | 11:00 PM JST
– 8:00 PM PT | 11:00 PM ET | 5:00 AM CEST (9 May) | 12:00 PM JST (9 May)

Join either session via Zoom:
https://us06web.zoom.us/j/86786028086?pwd=3Nd1ryIho7FlQtah56da8ypi78COux.1

Why Participate?

Participants will access key AI-RAN challenges prioritized by the industry, collaborating with top telecom operators, vendors, researchers, and ecosystem partners. This offers a direct platform to present ideas to the AI-RAN Alliance, with selected proposals eligible for fast-track benchmarking, increasing visibility, networking, and potential for future collaboration, funding, or real-world deployment.

Topics of Interest

We welcome proposals that delve into various facets of AI application in RAN, including, but not limited to:

Optimization of RAN Performance:
Examples including:

  • Novel signal processing techniques for extra-large MIMO (X-MIMO, i.e. 256 – 1024 antenna elements) and Occupied Bandwidth (OBW) of 400MHz and more (e.g. in FR3)
  • Reducing RAN latency, especially for simultaneous channel access of a large number of devices to enable mMTC deployment at scale.
  • Joint scheduling for distributed MIMO (D-MIMO) and MIMO layer-to-user mapping for mMIMO systems.
  • Uplink performance enhancements with AI/ML.
  • Mobility management for ultra reliable wireless access (uRLLC)
  • Beamforming steering for high spectral efficiency
  • Highly accurate UE positioning
  • Admission control for higher levels of security and lower latency
  • Innovative channel coding and decoding
  • Site specific modulation techniques that are learned using AI/ML
  • Improvements in random access of the channel

User Experience Enhancement:

Employing AI to optimize and personalize user experiences and predict user behavior for better service provisioning of existing and new applications. The aim of the proposals should be to demonstrate novel AI/ML techniques to deliver better user experience for these applications on the existing 5G RAN while also showing the path forward for enhanced experiences in the upcoming 6G RAN. A few examples are:

  • AI enabled XR with split compute
  • AI enabled critical applications for healthcare
  • AI enabled security applications for intrusion detection of aerial or ground perimeters
  • AI enabled remote control of industrial UAVs/Drones/AGVs
  • AI enabled Autonomous Vehicle steering
  • AI enabled immersive communication and gaming
  • AI and GenAI enabled collaboration across teams
  • AI / GenAI enabled tactile internet applications
  • AI enabled holographic communications

Use of AI or GenAI for Beyond Shannon Communication:

  • Goal oriented communication: conveying meaning rather than just accurately transmitting symbols or data.
  • Semantic communication: on achieving specific objectives through communication.

Sensing Aided Communication:
Use of AI and RF sensing (e.g. ISAC) and multi modal sensing (mix of RF and non-RF sensing modalities such as LiDAR, Radar, Camera) to perform RF resource management tasks to significantly improve spectral efficiency.

Predictive Maintenance and Network Automation:
Utilizing machine learning algorithms for forecasting potential network disruptions and maintenance needs. This could be based on real-time network digital twins as well as any SON / OSS / RIC platforms.

Security and Privacy:
AI-enabled solutions for safeguarding RAN against cyber threats and ensuring data privacy.

Joint orchestration of AI and RAN workflows:
Proposals that address advanced workflow balancing and orchestration mechanisms within AI-RAN environments. Of particular interest are AI-driven solutions that enable dynamic, real-time allocation of computing and communication resources, maximizing infrastructure utilization and supporting multi-tenant, concurrent RAN and AI workloads through unified orchestration platforms.

Radio Interface Enhancements:
Proposals that address new radio interfaces or enhancements for existing ones in order to deploy AI and Generative AI (GenAI) applications across consumer, enterprise, government sectors, and potentially other verticals. These enhancements could include new information elements that need to be shared between third party AI applications and RAN for real-time or near-real-time inferencing. For e.g., if an application relies on real-time access to radio environment information for joint sensing and communication type of applications, the proposals could study what kind of information the application needs to receive, and the related radio interface enhancements required.

APIs and integration enablers:
Contributions that expose or consume APIs relevant to AI-RAN use cases. This includes (but is not limited to) prototyping new APIs, demonstrating API-driven workflows, enabling plug-and-play components (e.g., inference modules, RAN functions, orchestration), using APIs to integrate third party models into simulators, making digital twin environments API-accessible, or showcasing integration across ecosystem actors. Emphasis should be on open interfaces, ease of experimentation, and enabling faster innovation.

What to Include in Your Proposal:

  • Concise problem statement identifying a specific AI-RAN challenge
  • Description of your proposed solution, including methodology and any early results
  • Rationale outlining impact, feasibility, and innovation

Submission Guidelines:

Proposals must present original and unpublished work, such as arXiv papers, prototypes, or GitHub repositories. Each submission should include a clear development plan outlining next steps, as well as defined expected outcomes. These outcomes may include a working prototype, performance evaluation results, or benchmarking-ready code—preferably hosted on GitHub.

Please note: If you encounter issues with this Google submittal form, please submit your proposal directly to: [email protected]

Please feel free to share this Call for Innovation Proposals announcement with anyone you think would be a great fit to participate or help spread the word.

We look forward to seeing your ideas push the boundaries of what’s possible in AI-powered RAN.

If you have questions, please contact: [email protected]