Stratosphere-First Telecoms #3 — NTN Alignment: Making HAPS Native to 5G/6G

Executive Brief

• Goal: Make HAPS “just another RAT” the core can steer - ground ↔ stratosphere ↔ satellite - based on performance, cost, priority, or emergency state.

• How: Standard interfaces to EPC/5GC; policy engines (PCF) for steering; analytics (NWDAF) for closed-loop assurance; slices for public safety, health, utilities, and general users.

• Outcome: A multi-layer network that extends coverage, adds seasonal capacity, and keeps mission-critical services online when terrestrial links fail.

1) The Multi-Layer Blueprint

Layers:

• Terrestrial (macro/small cells) - primary high-capacity layer

• Stratosphere (HAPS) - wide-area, steerable coverage & pop-up capacity

• Satellite (LEO/MEO/GEO) - global reach, backhaul, and direct-to-device (emerging)

Principle: Policy > Physics. Devices and flows attach to the layer that best meets the current SLA, cost, and risk - not simply the one with the strongest signal.

2) Control Plane: Make HAPS a First-Class Citizen

• Identity & Registration: Advertise HAPS cells with distinct TAC/TAI and PLMN attributes for granular policy.

• Policy Control (PCF):

  • Normal mode: Prefer terrestrial; selectively steer priority UEs (public safety, health, utilities) to HAPS.

  • Degraded mode: Trigger pre-emption; elevate public-safety slices; throttle consumer classes.

• Analytics (NWDAF): Feed radio + platform telemetry (RSRP/SINR, attach time, beam utilization, backhaul health) to predict congestion and auto-shift traffic.

• Orchestration: Treat HAPS beams as addressable capacity objects—scale slices, move beams, or shift backhaul by intent.

3) User Plane & Slicing Across Layers

Define S-NSSAIs that follow the user/workload across ground, HAPS, and satellite:

1. Public Safety MCX - PTT/video/data, low latency target, pre-emption enabled

2. Health & Emergency Ops - hospital links, EOCs, med telemetry

3. Utility/SCADA & IoT - NB-IoT/RedCap profiles, battery-sensitive

4. General Public - best-effort consumer traffic

QoS continuity: Keep QCI→5QI mappings consistent end-to-end; preserve class through HAPS and any satellite backhaul.

4) Steering Logic: Who Goes Where, When

Inputs: Radio KPIs, slice priority, backhaul cost/health, energy state, and incident flags.

Examples:

• Everyday coverage: Handsets stick to terrestrial; rural/remote or construction corridors attach to HAPS.

• Event surge: HAPS adds pop-up beams; policy caps consumer throughput to protect venue ops/safety.

• Disaster: Pre-empt consumer classes; pin public-safety, health, and utility slices to HAPS; enable cell broadcast.

• Backhaul fault: Shift HAPS to satellite backhaul; rate-limit video; prioritize MCX voice/data and SOS/messaging.

5) Device & Ecosystem Readiness

• Handsets: Standard LTE/5G devices work with HAPS when bands/PHY align. For direct-to-device satellite services, expect staged capability (messaging → voice → data).

• IoT: NB-IoT/RedCap over HAPS supports remote sensors (energy, transport, agriculture) with long battery life.

• Public Safety Terminals: MCX-capable smartphones + LMR/TETRA radios bridged via interworking gateway.

• Apps: SOS/messaging, push-to-talk, live maps, and telemetry must gracefully degrade (lower bitrates, store-and-forward) under constrained links.

6) Sovereignty, Security & Compliance (Built-In)

• Data residency & lawful intercept: Core-facing gateways on sovereign soil; audited LI points; immutable logging.

• Zero-trust ground segment: mTLS, hardware roots of trust, signed firmware (SBOMs), secure update pipelines.

• Key custody: In-country KMS/HSM for SIM/eSIM/network keys; emergency credential workflows.

• Privacy by design: Role-based access to incident data; explicit retention policies for public-safety traffic.

7) Disaster Continuity: “Policy to the Rescue”

When terrestrial collapses, policy flips the network:

• Time-to-air (TTA): Launch HAPS; umbrella beam attaches; public-safety slice rises automatically.

• Public warning: Cell Broadcast pushes multilingual alerts; device profiles bias to voice/PTT + messaging.

• Critical operations: Hospitals, utilities, banking and command centres keep priority lanes; satellite backhaul engages if needed.

• Handover back to ground: As towers restore, policy returns consumer traffic; HAPS remains as guardrail until MTTR closes.

8) KPIs That Prove NTN Works

Availability & Speed

• Time-to-air (TTA) — launch → first attach

• Time-to-alert (T2A) — alert trigger → handset receipt

Mission-Critical

• PTT latency (95th), MCX call setup success, body-cam video success at constrained bitrates

Multi-Layer Efficiency

• Policy hit rate (correct layer chosen)

• Beam utilization and PRBs by slice

• Backhaul failover time and throughput under failover

Business Impact

• Sites restored with HAPS support

• Service credits avoided

• Population under coverage (dynamic estimate)

9) 12-Month Operator Roadmap

Quarter 1–2

• Reference architecture: EPC/5GC integration, PCF policies, NWDAF hooks, slice catalog.

• Lab validation: interop with LMR/TETRA, cell broadcast path, dual backhaul failover.

• Select two pilot areas: one coverage/capacity, one disaster drill.

Quarter 3–4

• Fly 60–90 day pilots: coverage first, then disaster simulation (fibre cut + power loss).

• Measure KPIs; tune steering and slices; publish runbooks for NOC/SOC/EOC.

• Contractualize: roaming-like commercial model for consumer add-ons; SLAs for public safety/utilities; enterprise resilience bundles.

10) Risks & Practical Mitigations

Risk	Impact	Mitigation
Airspace / weather windows	Delayed activation	Seasonal planning; dual launch sites; pre-cleared corridors
Spectrum / interference	Attach failures; coverage gaps	Early coordination; interference studies; dynamic power/tilt
Backhaul fragility	SLA breaches	Dual path (microwave + sat/LEO); QoS marking; auto-failover
Security incident	Regulatory & reputational damage	Zero-trust; HSM; SBOM; continuous monitoring; red-team drills
Vendor lock-in	Loss of leverage	Open interfaces; multi-orbit backhaul; contract exit ramps
Interop complexity	Slow incident response	Pre-pilot lab for MCX + LMR; tested E2E runbooks; regular exercises

Conclusion: Make the Network Layer-Agnostic

NTN alignment isn’t a science project - it’s good networking. When HAPS is treated as a native RAT, slices follow users across layers, policy steers traffic by intent, and resilience becomes routine. The winners will design once for coverage, capacity, and continuity, then let the core decide - ground, stratosphere, or space - in real time.