Pre-Training Survey (Adv Security and AI for 5G & Beyond)
Training Date
Name
Job Title / Department
Mobile No.
Years of experience in telecommunications networks:
Years of experience specifically in 5G (planning / deployment / operations):
Years of experience in cybersecurity / pentesting:
Years of experience in applied AI / ML (model training, automation, pipelines, inference):
Current responsibilities (select one or more):
Please enter a message
Participation goals (Top 3 desired learning outcomes)
1. Rate your practical, hands on competence for each topic on a scale 1 to 5 (1=Some to None, 2=Basic, 3=Intermediate, 4=Advanced, 5=Expert) - Evolution 4G→5G Security & ZTA
2. Identifying RAN vulnerabilities
3. Core (UPF/AMF/SMF/NRF) threat assessment
4. Transport / SDN / NFV / MEC security issues
5. API / OAuth / Service Exposure security
6. Threat modeling & risk frameworks (MITRE ATT&CK, NESAS)
7. Building / operating lab with srsRAN, Open5GS, UERANSIM, Amarisoft
8. Packet / signaling analysis (Wireshark, S1AP/NAS)
9. Packet / signaling analysis (Wireshark, S1AP/NAS)
10. Network slicing security & isolation tests
11. 5G control plane attack simulation
12. Rogue gNB / IMSI catcher techniques
13. Diameter / HTTP/2 signaling attacks
14. User plane (GTP-U, DNS tunneling)
15. Kubernetes / CNF hardening & RBAC multi-tenancy
16. AI-based threat detection / ML IDS deployment
17. Incident response & forensics for 5G attacks
18. Red/Blue Team simulation in 5G context
19. Designing 5G AI data pipelines (collection→inference→retraining)
20. GPU cluster / dimensioning for AI workloads
21. Scaling PoC to production AI deployments
22. Closed-loop / self-healing automation design
23. Federated learning in distributed network nodes
24. Architectures for L4/L5 autonomy & policy enforcement
25. Vendor neutral autonomous architectures / scaling challenges
26. Key AI use cases (network slicing, predictive maintenance, anomaly detection, cybersecurity)
27. Working with LLMs / SLMs in telecom ops
28. Model deployment strategies & supervised vs. unsupervised approaches
29. AI validation: accuracy, fairness, reliability testing
30. Digital twins & simulation for AI validation
31. Governance & regulatory / compliance frameworks
32. Securing AI-native telecom systems (containers, workloads)
33. AI-specific threat mitigation & model hardening
34. Planning for 6G native AI & zero-trust evolution
1. Indicate familiarity on Tooling & Platform: 0=Never Used, 1=Some Exposure, 2=Used in Lab/PoC, 3=Operational Use, 4=Can Optimize/Troubleshoot - srsRAN / Open5GS testbed setup
2. UERANSIM / Amarisoft in lab
3. Wireshark (S1AP/NAS)
4. Kubernetes for CNFs (RBAC, multi-tenancy)
5. ML-based IDS / anomaly detection tooling
6. SOAR / automated incident response
7. GPU infrastructure for AI training / retraining
8. Federated learning frameworks
9. Digital twin / simulation platforms
10. AI model pentesting / hardening
List any additional specialized tools / frameworks you expect to leverage during the course.
Scenario-Based Baseline (Short Answers) - You notice anomalous signaling bursts in the control plane possibly tied to HTTP/2 abuse (reference to signaling attacks) . Outline your initial triage & tools.
A new network slice exhibits performance degradation; isolation bypass suspected . What tests do you run?
Design a minimal lab to reproduce a user plane tunneling exploit (referencing GTP-U vulnerabilities) . Which components & datasets are essential?
You must introduce closed-loop remediation for RAN anomalies (self-healing workflows). What telemetry & decision logic do you prioritize first?
Draft first 3 validation gates before deploying an ML model detecting IMSI-catcher patterns (rogue gNB techniques) referencing model validation needs .
Hardening an AI-powered network function prior to production (hardening & securing AI workloads) : list top 3 control categories you implement
Planning for Level 4 autonomy: identify one policy enforcement challenge and one scalability concern (policy enforcement & scalability roadmaps).
1. Knowledge Self Assessment (Quick Check) / Select the best option; this gauges baseline only (not graded) - Primary objective of Zero Trust in 5G context (related to ZTA)
2. Most appropriate data pipeline stage to implement feature normalization for anomaly detection (pipeline stages)
3. Which element is directly part of a federated learning design in distributed 5G? (federated learning mention)
4. Key risk when scaling PoC AI to production (scaling strategies)
5. Primary reason for using digital twins in AI validation (digital twins context)
Open Feedback - Anything you do not want covered (already expert in)?
Additional comments / expectations / questions.
Thank you for submitting your pre-training survey. We shall review and endeavour to deliver a training that meets your learning objectives. Looking foward to meeting you soon!