18_test_plan_mobility_site_to_core
Mobility 4G/5G Holistic Security Test Plan
Objective
Evaluate a mobility environment end-to-end as a security engineer, from the antenna site and radio layer through transport, core, cloud, and management systems. This plan combines the Mobility/ and open5gs_lab/ source material with current public guidance from 3GPP, GSMA, NIST, CISA, ENISA, and O-RAN Alliance.
Scope
In scope
- Antenna site and cabinet security
- 4G LTE and 5G NSA/SA radio access
- Fronthaul, midhaul, backhaul, and timing/sync paths
- EPC / 5GC control and user plane interfaces
- SBI, PFCP, GTP, Diameter, DNS/PKI, and roaming boundaries
- OSS/BSS/EMS/NMS, virtualization, Kubernetes, and CI/CD dependencies
- Monitoring, logging, detection, and resilience controls
Out of scope unless separately authorized
- Live-spectrum active transmission against production networks
- Interference, jamming, or unauthorized subscriber interception
- Any third-party roaming or interconnect actions without written approval
Engagement guardrails
- Use passive collection for production RF work unless you have explicit spectrum and operator authorization.
- Keep active air-interface testing inside a Faraday cage or licensed test environment using your own lab SIMs and devices.
- Treat antenna sites, transport POPs, NOCs, and interconnects as critical infrastructure with change control and evidence-handling requirements.
Typical operator architecture to assess
The plan assumes a modern operator footprint with these major components:
- Antenna site / RAN edge: antennas, RET/AISG, RRU/RRH or O-RU, BBU or DU, site router/switch, PTP/GNSS timing, rectifiers, batteries, generator transfer, environmental controller, cameras, and badge access.
- Transport: fiber or microwave fronthaul/backhaul, L2/L3 aggregation, MPLS/IP transport, timing distribution, OOB management, and encryption domains.
- Core and edge: EPC for 4G, 5GC for 5G SA, MEC or regional UPF breakout, DNS, PKI, API gateways, and interconnect security controls.
- Management and cloud: OSS/BSS, EMS/NMS, SIEM, identity providers, NFV, Kubernetes/OpenShift, image registries, GitOps/CI/CD, and vendor remote access.
Access-level model
Use this legend in every test case to state the level of access the assessor has.
| Level | Access you have | Typical examples |
|---|---|---|
| AL0 | Public or untrusted external vantage | OSINT, passive observation from outside facilities, public attack surface review |
| AL1 | Authorized subscriber-side access only | Your test handset, test SIM, passive RF collection, drive/walk testing |
| AL2 | Authorized physical site access | Escorted cabinet inspection, console access, patch panel review, power and lock checks |
| AL3 | Internal read-only network or operations access | SPAN/TAP, NMS/OSS read-only, transport diagrams, inventory, logs, config review |
| AL4 | Lab or pre-production admin access to telecom functions | Open5GS, UERANSIM, srsRAN, NF config changes, controlled protocol validation |
| AL5 | Cloud/platform/admin access | Kubernetes, hypervisor, CI/CD, secrets, PKI, image registry, IAM |
| AL6 | Interconnect or roaming security authority | SEPP, Diameter/GRX/IPX, partner trust, external signaling gateways |
Architecture and attack-location diagrams
Diagram 1: End-to-end mobility trust boundaries
graph LR
classDef external fill:#d62828,stroke:#264653,color:#fff
classDef internal fill:#2a9d8f,stroke:#264653,color:#fff
classDef thirdparty fill:#e76f51,stroke:#264653,color:#fff
classDef data fill:#264653,stroke:#2a9d8f,color:#fff
subgraph EXT["External / Subscriber Edge"]
UE["UE / handset / SIM"]:::external
RF["Air interface
Uu / NR"]:::external
end
subgraph SITE["Antenna Site / Cell Edge"]
ANT["Antennas / RET / RRU or O-RU"]:::internal
DU["BBU / DU / site router"]:::internal
PWR["Power, battery, GNSS/PTP, cabinet access"]:::internal
end
subgraph TRANS["Transport"]
FH["Fronthaul / midhaul"]:::thirdparty
BH["Backhaul / MPLS / microwave / fiber"]:::thirdparty
OOB["OOB management"]:::thirdparty
end
subgraph CORE["Regional Edge / Core"]
EPC["4G EPC
MME/HSS/SGW/PGW/PCRF"]:::internal
SA["5G Core
AMF/SMF/UPF/AUSF/UDM/NRF"]:::internal
MEC["MEC / UPF breakout / DNS / PKI"]:::data
end
subgraph MGMT["Management / Platform"]
OSS["OSS/BSS/EMS/NMS"]:::internal
K8S["Kubernetes / NFV / CI-CD / IAM"]:::internal
SIEM["SIEM / telemetry / detections"]:::data
end
UE -->|"attach / registration / paging"| RF
RF -->|"attack area: rogue cell, paging, downgrade"| ANT
ANT -->|"attack area: cabinet, console, AISG, GNSS"| DU
DU -->|"attack area: fronthaul / sync / management"| FH
FH -->|"attack area: transport exposure"| BH
BH -->|"attack area: signaling and user plane reachability"| EPC
BH -->|"attack area: signaling and user plane reachability"| SA
EPC -->|"Diameter / GTP / policy"| MEC
SA -->|"SBI / PFCP / GTP-U"| MEC
OSS -->|"config / monitoring / remote admin"| DU
OSS -->|"config / monitoring / remote admin"| EPC
OSS -->|"config / monitoring / remote admin"| SA
K8S -->|"platform and secret control"| SA
K8S -->|"platform and secret control"| MEC
SIEM -.->|"telemetry / alerting coverage"| PWR
SIEM -.->|"telemetry / alerting coverage"| BH
SIEM -.->|"telemetry / alerting coverage"| SADiagram 2: Antenna site attack locations
graph TB
classDef external fill:#d62828,stroke:#264653,color:#fff
classDef internal fill:#2a9d8f,stroke:#264653,color:#fff
classDef detect fill:#264653,stroke:#2a9d8f,color:#fff
ATT["Assessor / attacker position"]:::external
ANT["Panel antennas / massive MIMO"]:::internal
RRU["RRU / O-RU / TMA / RET"]:::internal
CAB["Cabinet / DU / router / switch"]:::internal
TIME["GNSS / PTP / sync"]:::internal
PWR["Rectifier / battery / generator transfer"]:::internal
MGMT["Local craft / serial / OOB mgmt"]:::internal
DET["CCTV / access logs / tamper alarms"]:::detect
ATT -->|"passive RF collection"| ANT
ATT -->|"physical tamper / cable access"| RRU
ATT -->|"badge / lock / cabinet access"| CAB
ATT -->|"timing spoof / sync dependency review"| TIME
ATT -->|"power outage / resiliency review"| PWR
ATT -->|"console / mgmt exposure review"| MGMT
DET -.->|"detects or records activity"| ANT
DET -.->|"detects or records activity"| CAB
DET -.->|"detects or records activity"| MGMTDiagram 3: Control, user, and management plane attack map
graph TD
classDef attack fill:#d62828,stroke:#264653,color:#fff
classDef internal fill:#2a9d8f,stroke:#264653,color:#fff
classDef detect fill:#264653,stroke:#2a9d8f,color:#fff
K1["Radio plane
UE <-> eNB/gNB"]:::attack
K2["Site plane
RRU/O-RU <-> DU/CU"]:::attack
K3["Transport plane
fronthaul / backhaul / timing"]:::attack
K4["Control plane
NAS / NGAP / S1AP / Diameter / SBI / PFCP"]:::attack
K5["User plane
GTP-U / N3 / S1-U / N6 breakout"]:::attack
K6["Management plane
OSS/BSS / EMS / IAM / K8s / vendor access"]:::attack
D1["Packet capture, logs, timing alarms, auth events"]:::detect
C1["Defense objective:
isolate, authenticate, encrypt, detect"]:::internal
K1 --> K2 --> K3 --> K4 --> K5 --> K6
D1 -.->|"coverage required across every layer"| K1
D1 -.->|"coverage required across every layer"| K3
D1 -.->|"coverage required across every layer"| K6
C1 -.->|"control intent applies to all planes"| K2
C1 -.->|"control intent applies to all planes"| K4
C1 -.->|"control intent applies to all planes"| K5Required equipment
Core assessment kit
| Equipment | Required | Used for |
|---|---|---|
| Linux laptop or workstation with Docker and packet tools | Yes | Open5GS/UERANSIM/srsRAN lab, captures, config review |
| Wireshark/tshark/tcpdump | Yes | NGAP, S1AP, NAS, GTP, PFCP, Diameter, HTTP/2 SBI analysis |
| Test Android handset plus known-good test SIM/eSIM profile | Yes | Subscriber-side validation, field measurements, privacy and fallback checks |
| Documentation kit (diagrams, inventory, IP plan, standards mapping) | Yes | Site-to-core traceability and evidence collection |
RF and field kit
| Equipment | Required when | Used for |
|---|---|---|
| Passive RF scanner or handset-based measurement apps | AL1 field testing | Cell inventory, PCI/TAC/neighbor mapping, serving-cell changes |
| SDR such as USRP B210 or equivalent | AL4 lab-only active testing | Controlled attach, registration, and protocol validation in isolated lab |
| Faraday cage or licensed RF-shielded environment | Any active radio testing | Prevents live-network interference |
| Directional and omni antennas, attenuators, calibrated cables | SDR lab work | Safe signal shaping and repeatable lab scenarios |
| Portable battery pack, safety PPE, flashlight, console cable set | AL2 site work | Cabinet review and safe field inspection |
Site, transport, and platform kit
| Equipment | Required when | Used for |
|---|---|---|
| Read-only NMS/OSS/EMS access | AL3 | Config, alarms, inventory, remote-access review |
| SPAN/TAP or capture container | AL3-AL5 | Control-plane and management-plane evidence |
| Time sync monitoring tools | AL3 | GNSS/PTP drift, holdover, and alarm validation |
Kubernetes tools (kubectl, helm, image scanning, secrets review) |
AL5 | CNF/NFV and platform hardening checks |
| Secure evidence repository | All phases | PCAPs, screenshots, logs, config snapshots, site photos |
Test execution waves
| Wave | Goal | Primary access |
|---|---|---|
| W1 | External and subscriber-side mapping | AL0-AL1 |
| W2 | Antenna site and transport review | AL2-AL3 |
| W3 | Core signaling and management validation | AL3-AL5 |
| W4 | Resilience, isolation, and detection validation | AL3-AL6 |
Test case matrix
| ID | Domain | What you validate | Where the attack would happen | Access level | Key equipment | Expected evidence / pass condition |
|---|---|---|---|---|---|---|
| TC-01 | RF inventory | Build a cell/site map, identify LTE/NR bands, TAC/PCI reuse, and unusual neighbor relations | Air interface and site perimeter | AL1 | Test UE, passive RF tools | Repeatable cell inventory and no unexplained rogue or stale neighbors |
| TC-02 | Subscriber privacy | Confirm 5G identities use SUCI where expected and document NSA/LTE fallback exposure | Registration path between UE and AMF/MME | AL1 + AL4 | Test UE, Open5GS lab, Wireshark | No plaintext permanent identity in expected 5G flows; fallback paths documented |
| TC-03 | Downgrade control review | Check whether network design, UE policy, and monitoring detect or prevent fallback abuse | Radio control plane | AL1 + AL4 | Test UE, lab stack, captures | Downgrade conditions are logged, bounded, and tied to policy |
| TC-04 | Paging and location privacy | Assess whether paging identifiers and timing can expose subscriber presence patterns | Air interface and mobility management | AL1 + AL4 | Passive RF tools, captures | Paging identifiers rotate appropriately; monitoring exists for anomalies |
| TC-05 | Antenna site physical security | Review locks, cabinet tamper controls, cable exposure, local craft ports, and third-party access processes | Tower, rooftop, or cabinet site | AL2 | PPE, console kit, checklist, camera | No uncontrolled cabinet access, exposed ports, or undocumented vendor pathways |
| TC-06 | Site management exposure | Verify craft, serial, web, and OOB interfaces are authenticated, segmented, and logged | Site router, DU/BBU, environmental controller | AL2 + AL3 | Console cables, read-only NMS, packet capture | No default creds, clear segmentation, and complete admin logging |
| TC-07 | Timing and sync resilience | Review GNSS/PTP dependency, holdover, alarming, and fallback design | Site timing chain and transport | AL3 | Timing dashboards, architecture docs | Timing sources are redundant, monitored, and not single-point dependent |
| TC-08 | Power resilience | Validate battery runtime, generator transfer assumptions, rectifier alarms, and restoration procedures | Antenna site power plane | AL2 + AL3 | Site docs, alarm history | Site power failure modes are documented and alerting works end-to-end |
| TC-09 | Fronthaul/backhaul security | Verify segmentation, encryption, QoS separation, and OOB isolation on fiber/microwave paths | Fronthaul, midhaul, and backhaul transport | AL3 | Network diagrams, read-only config, captures | Signaling, user, timing, and management traffic are separated and protected |
| TC-10 | 4G core baseline | Validate attach, EPS-AKA, bearer setup, and logging in the EPC | MME/HSS/SGW/PGW path | AL4 | Open5GS, srsRAN, Wireshark | Clean attach flow, expected Diameter/GTP behavior, and preserved evidence |
| TC-11 | 5G core baseline | Validate registration, 5G-AKA, PDU session establishment, and NF registration | gNB/AMF/SMF/UPF/NRF path | AL4 | Open5GS, UERANSIM, Wireshark | Clean registration and PDU session flow with expected SBI activity |
| TC-12 | Diameter, GTP, and PFCP hardening | Review reachability, source validation, exposure boundaries, and detection coverage | EPC/5GC control and user plane interfaces | AL3 + AL4 | Captures, configs, firewall rules | No unnecessary exposure; filtering and logging exist for key interfaces |
| TC-13 | SBI security | Verify mTLS, service auth, cert lifecycle, rate limits, and least-privilege NF access | 5G SBI between AMF/SMF/UDM/AUSF/NRF/PCF | AL4 + AL5 | Open5GS lab, PKI records, API review | NF communication is authenticated and unauthorized discovery is blocked |
| TC-14 | Slice and tenant isolation | Confirm unauthorized slice requests fail and shared UPF/compute boundaries are controlled | NSSF/AMF/SMF/UPF and platform layer | AL4 + AL5 | UERANSIM, Open5GS, K8s tooling | Slice policy is enforced and no cross-slice leakage is observed |
| TC-15 | O-RAN or split-RAN review | Assess O-RU/O-DU/O-CU, E2, A1, O1, and vendor remote access security | Open RAN and disaggregated RAN interfaces | AL3 + AL5 | Architecture docs, O-RAN configs, mgmt review | Open interfaces are authenticated, segmented, and monitored |
| TC-16 | MEC and local breakout | Review edge UPF breakout, DNS, service exposure, and local application trust boundaries | MEC / regional edge / N6 breakout | AL3 + AL5 | Edge diagrams, capture, app inventory | Breakout paths are isolated and edge services are not implicitly trusted |
| TC-17 | OSS/BSS and admin identity | Review privileged access, break-glass accounts, MFA, PAM, and vendor jump paths | Management plane | AL3 + AL5 | IAM, PAM, bastion logs | Privileged access is controlled, approved, and attributable |
| TC-18 | CNF/NFV/Kubernetes posture | Validate image provenance, secret handling, pod security, east-west policy, and backup integrity | Telecom cloud platform | AL5 | kubectl, registry review, policy tools |
CNFs follow hardened runtime and secret-management controls |
| TC-19 | Detection and telemetry | Confirm alarms correlate across RF, transport, core, and cloud for the same incident | SIEM/NMS/telemetry pipeline | AL3 + AL5 | SIEM, dashboards, alert rules | A cross-layer incident can be traced from site to core without blind spots |
| TC-20 | Roaming and interconnect governance | Review SEPP, Diameter/GRX/IPX trust boundaries, partner filtering, and contractual security controls | External interconnect boundary | AL6 | Interconnect diagrams, policies, read-only gateway logs | Partner paths are filtered, authenticated, and formally governed |
Priority test cases to run first
If time is limited, start with these eight because they expose the largest end-to-end risk concentration:
- TC-05 Antenna site physical security
- TC-06 Site management exposure
- TC-09 Fronthaul/backhaul security
- TC-11 5G core baseline
- TC-12 Diameter, GTP, and PFCP hardening
- TC-13 SBI security
- TC-17 OSS/BSS and admin identity
- TC-19 Detection and telemetry
Evidence checklist by layer
| Layer | Evidence to collect |
|---|---|
| RF / UE | Serving-cell screenshots, band and TAC inventory, registration captures, fallback behavior |
| Site | Photos, lock and tamper observations, local-port inventory, access logs, power and environmental alarm history |
| Transport | L2/L3 diagrams, encryption and VRF/VLAN policy, sync topology, firewall/filter rules |
| Core | PCAPs for NGAP/S1AP, Diameter, PFCP, GTP, HTTP/2 SBI; NF configs and route policy |
| Platform | K8s manifests, image sources, IAM role maps, cert lifecycle records, secret storage method |
| Detection | Correlated alerts, dashboards, retention settings, on-call runbooks, incident replay notes |
Pass/fail criteria
- Pass: the layer is authenticated, segmented, minimally exposed, monitored, and operationally recoverable.
- Needs improvement: controls exist but do not cover the full trust boundary, are not logged, or depend on a single point of failure.
- Fail / high risk: unauthenticated management paths, flat transport, missing SBI or interconnect protections, weak site access control, or no cross-layer visibility.
Recommended reporting structure
For each executed test case, record:
- Access level used
- Asset or interface tested
- Attack location
- Evidence gathered
- Security impact if abused
- Existing control
- Gap severity
- Recommended remediation owner
Lab mapping to your existing material
Use the open5gs_lab/ content for controlled validation of protocol and core behaviors:
17_test_plan_4g_5g_holistic.mdfor baseline functional and security domains08_threat_model_4g.mdand09_threat_model_5g.mdfor EPC and 5GC attack surfacesTP-00_resources_and_setup.mdfor lab build requirements
Use the Mobility/ material to extend beyond the lab:
1. RF_Air_Interface.mdfor air-interface and site-edge attack surfaces2. RRC_NAS_Signaling.mdfor registration, identity, and mobility signaling exposure3. Core_Network_Diameter_GTP.mdfor core transport and interconnect risks4. UE_Android_SIM.mdfor subscriber-side validation5. 5G_Specific.mdfor SBA, slicing, O-RAN, and PFCP focus
External references
- 3GPP TS 33.401 - LTE/EPS security architecture
- 3GPP TS 33.501 - 5G security architecture
- 3GPP TS 29.244 - PFCP for N4
- 3GPP TS 29.510 - NRF and service discovery
- GSMA FS.40 v3.0 (2024) - 5G security guide
- GSMA FS.19 / FS.20 - Diameter and GTP security guidance
- NIST SP 800-187 - LTE security guidance
- NIST SP 1800-33 - 5G cybersecurity practice guide
- NIST CSWP 36E - 5G network security design principles
- CISA 5G Security and Resilience guidance
- ENISA 5G Security Controls Matrix
- O-RAN Alliance security specifications for open interfaces and management security
Outcome
This plan gives you a practical way to assess mobility holistically: where the attack happens, what access the assessor has, what equipment is required, and what evidence proves the control either works or fails. It is designed so you can execute the same framework against a lab, a private network, or a production operator environment with the access level adjusted per phase.