3 Credit Hours

(also offered as CSC 591)

VS

Vijay Shah

Assistant Professor

vijay.shah@ncsu.edu

This course provides a comprehensive introduction to AI-native future-generation telecom networks, with a particular emphasis on Open Radio Access Networks (Open RAN or O-RAN) and emerging AI-RAN architectures. Students will explore the evolution of wireless networks toward open, intelligent, and autonomous systems through the principles of disaggregation, interoperability, open interfaces, cloud- native design, and artificial intelligence. The course examines how AI is transforming network planning, deployment, optimization, orchestration, and operations across the Radio Access Network (RAN), transport, and core domains.
Students will gain both theoretical foundations and practical experience in designing, implementing, and evaluating intelligent network solutions, including RAN Intelligent Controllers (RICs), xApps/rApps/dApps, AI agents, digital twins, and AI/ML-driven network automation frameworks. Topics include Open RAN architectures and standards, AI-RAN frameworks, generative AI and large language models (LLMs) and time-series foundational models (TSFMs) for network management, autonomous network operations, spectrum intelligence, edge AI, cloud-native network functions, and emerging applications in IoT, Industry 4.0, and XR/VR systems.
Through hands-on laboratories and projects, students will develop the knowledge and skills required to design, deploy, and innovate within next-generation AI-native telecom ecosystems.

Prerequisites

Students should have completed a networking course (e.g., ECE 407 engineering principles of computer communications and networking) or have equivalent background knowledge.

In addition, students are expected to:

• Be proficient in programming (e.g., Python, C/C++).
• Be comfortable learning new software frameworks, development environments, and simulation tools.

Prerequisites may be waived with instructor approval.

Learning Outcomes

Upon successful completion of this course, students will be able to:

• Explain the architecture, standards, and key technologies underlying Open RAN, AI-RAN, and AI-native telecom networks.
• Design and develop intelligent network applications using Open RAN interfaces, RIC platforms, AI/ML models, AI agents, and cloud-native technologies.
• Analyze and evaluate the role of artificial intelligence in network automation, optimization, orchestration, and autonomous operations across RAN, transport, and core networks.
• Implement and test AI-driven telecom solutions using Open RAN software tools, digital twins,AI/ML frameworks, and network automation platforms.
• Assess emerging trends and challenges in AI-native networking, including generative AI, foundational models, edge intelligence, spectrum optimization, autonomous networks, and 6G architectures.
• Apply Open AI-RAN concepts to real-world use cases in mobile broadband, private networks, IoT, Industry 4.0, and future wireless systems.

Course Outline

1. Introduction to Open RAN, AI-RAN, and AI-native Telecom Networks
   a. Evolution of cellular networks from 2G through 5G and towards 6G
   b. Evolution of RAN architectures: Traditional RAN, virtualized RAN (vRAN), Open RAN, and AI-RAN
   c. Introduction to the O-RAN ecosystem and standards
   d. AI-native network vision and autonomous networks
   e. Representative use cases, including, QoE optimization, RAN slicing, Dynamic spectrum sharing, Industrial IoT etc.
2. Open RAN and AI-RAN Architecture
   a. O-RAN reference architecture, Service Management and Orchestration (SMO), Non-Real-Time RIC and Near-Real-Time RIC
   b. O-Cloud and cloud-native network functions
   c. Open interfaces: O1, O2, A1, E2, Open Fronthaul, AIE1 and AI-related interfaces, Relevant 3GPP interfaces
   d. AI-RAN architectural frameworks
   e. Integration of AI workloads and telecom workloads
   f. Distributed AI inference and training across RAN infrastructure
3. AI and ML for Open, AI-native RAN networks
   a. Fundamentals of AI/ML for wireless systems
   b. Supervised, unsupervised, reinforcement, and federated learning
   c. Generative AI and Large Language Models (LLMs) for telecom operations
   d. AI model lifecycle management (MLOps/AIOps)
   e. Data collection, feature engineering, and model deployment
   f. AI governance, explainability, robustness, and trustworthiness
   g. Telecom digital twins and simulation environments
4. Intelligent Applications: xApps, rApps and dApps/ μApps
   a. Various Near-RT RIC xApp use cases, such as, KPIMON xApp, RAN Slicing xApp, ML-based xApps
   b. Various Non-RT rApp use cases, such as, traffic steering rApp, QoE rApp
   c. Various dApps/ μApps, such as, RB scheduling μApp and Beamforming dApp.
5. AI-RAN Paradigm and Emerging Use Cases
   a. AI-for-RAN: AI optimizing network performance
   b. AI-on-RAN: AI services running on RAN infrastructure
   c. AI-and-RAN: Converged AI and RAN services
   d. AI-Grid and distributed compute fabrics
   e. Resource orchestration between RAN and compute resources
6. O-RAN and AI-RAN Research and Development Initiatives
   a. Industry-led initiatives, such as, O-RAN ALLIANCE, TIP, ONF, and major telecom vendors and operators,
   b. Academia led initiatives like OAIC, Colosseum, AERPAW and 5GENESIS,
   c. Government led initiatives like NTIA.
7. Future Directions/Roadmap
   a. Open AI RAN evolution towards 6G
   b. Native AI architecture in 6G systems
   c. Integrated Sensing and Communication
   d. Roadmap towards fully autonomous, self-evolving telecom networks

Course Requirements

• 20% homework/labs (Coding/programming -based assignment using srsRAN/OAI 5G testbed, AERPAW testbed, Colosseum emulator or at least, ns-O-RAN simulator/emulator)
• 20% class presentations (on O-RAN and AI-RAN topics)
• 20% exam
• 40% class project

Textbook

Nishith D. Tripathi and Vijay K. Shah, Fundamentals of O-RAN, Wiley/IEEE Press (Link: https://www.wiley.com/en-ie/Fundamentals+of+O-RAN-p-9781394206803)

Updated: 05/26/2026