Papers
Topics
Authors
Recent
Search
2000 character limit reached

5G NR NTN Support Overview

Updated 1 April 2026
  • 5G NR NTN support is a standardized enhancement of 5G NR that integrates satellite, airborne, and terrestrial links to deliver ubiquitous broadband and precise positioning services.
  • The system adapts key PHY and MAC layers through waveform modifications, extended cyclic prefixes, and advanced Doppler estimation to maintain signal integrity across diverse orbits.
  • Implementation leverages SDR testbeds and AI-driven control to validate protocols in GEO, MEO, LEO, and HAPS/UAV configurations, paving the way for 5G-Advanced and 6G evolution.

5G NR NTN (Non-Terrestrial Networks) Support encompasses the standardized augmentation of 5G New Radio (NR) air interface and protocol stack to enable direct radio connectivity from terrestrial terminals to spaceborne (satellites), airborne (HAPS, UAV) and integrated terrestrial networks. This global-scale architecture, defined and iteratively extended by 3GPP Releases 17/18/19, underpins broadband, mission-critical, and positioning services—subject to unique Doppler, path loss, beam management, and protocol-timing constraints—in support of ubiquitous service continuity and advanced PNT layers for 5G-Advanced and 6G (Dureppagari et al., 2024).

1. System Architecture and NTN Configurations

5G NR NTN architectures are classified by payload processing and platform type. Transparent payloads (“bent-pipe”) relay radio-frequency (RF) signals from ground gateways to end users over satellites without on-board processing, while regenerative payloads integrate gNB-DU/CU and user-plane functions on the satellite, supporting ISLs and onboard resource management (Lin et al., 2021, Guidotti et al., 2022, Figaro et al., 8 Jan 2026). Target platforms span:

  • GEO: Provides continental-scale coverage, RTT ~250 ms, typically used for broadcast and delay-tolerant connectivity.
  • MEO: Medium-latency, intermediate Doppler; serves regional/cross-continental footprints.
  • LEO: Low-latency (~3–10 ms one-way), high Doppler (>±7 kHz at S-band), suitable for latency-sensitive eMBB, IoT, and advanced PNT (Sedin et al., 2020, Dureppagari et al., 2024).
  • HAPS/UAV: Regional and on-demand, negligible Doppler, suited to rapid deployment and localized service restoration.

User Equipment (UE) interfaces are FR1 (L-band/S-band for handheld, C-band for backhaul), and FR2 (Ku/Ka for VSAT and fixed/vehicular broadband), with FDD duplexing preferred to mitigate round-trip propagation delay (Lin et al., 2021, Guidotti et al., 2022).

2. Physical Layer Enhancements: Numerologies, Beams, and Doppler Management

NTN imposes stringent requirements on the NR PHY layer due to slant-range propagation, elevated Doppler shifts, and large cell radii.

  • Waveform & Numerology: CP-OFDM (downlink) and DFT-s-OFDM (uplink) are retained, with subcarrier spacing Δf ∈ {15,30,60,120} kHz adapted for satellite bands and Doppler spreads (Lin et al., 2021, Guidotti et al., 2022, Dureppagari et al., 2024). Extended cyclic prefixes absorb differential delays in NTN footprints >100 km.
  • Positioning Beams and PRS: Periodic, wide “positioning beams” (orthogonal to service beams) are TDM-multiplexed to transmit PRS at maximum coverage, facilitating PRS reception under higher delay/Doppler uncertainties and differentiating positioning and communication resources (Dureppagari et al., 2024, Gonzalez-Garrido et al., 2024).
  • Doppler and Timing Advance:
  • Random Access: PRACH formats with longer preamble and cyclic prefix, and extended RACH windows, are specified for large initial-access delays (Lin et al., 2021, Guidotti et al., 2022).

3. Positioning, Navigation, and Timing (PNT) Over NTN

LEO-based NR NTN supports high-precision PNT as a complement (and potential alternative) to GNSS (Dureppagari et al., 2024, Gonzalez-Garrido et al., 2024):

  • PRS Design: PRS is mapped to dedicated NR resource grid slots (e.g., 1–14 OFDM symbols in every 1 ms slot, periodicity 40 ms–1.28 s), with QPSK Gold sequence comb-mapping, wide spectral guard bands to accommodate Doppler, and frequency-division multiplexing (combSize) for orthogonal satellite PRSs (Gonzalez-Garrido et al., 2024, Dureppagari et al., 2024).
  • UE Processing:
    • Coarse acquisition exploits 2D delay/Doppler search over buffered PRS windows (W=400 ms typical), peak-detection threshold η as per target P_FA.
    • Sliding-window NLS batch estimation (10 epochs over 400 ms) supports sub-10 m accuracy for 97% of UEs (B=5 MHz, 14-symbol PRS, two Rx ports) (Dureppagari et al., 2024).
  • Performance:
    • Median TOA error: 30 ns (9 m) at B=5 MHz, one port; reduced by 20–25% with 2-port combining.
    • Link-budget gains: LEO@600 km provides ≈20 dB better FSPL than MEO, higher bandwidth gives ≈7 dB additional processing gain (Dureppagari et al., 2024, Gonzalez-Garrido et al., 2024).
  • Deployment Principles:
  • Joint Communication and PNT: Overlay solutions enable concurrent navigation data and 5G communication (e.g., DSSS overlay on 5G OFDM), allowing minimal receiver modification and robust decoding under LEO Doppler provided SIR and SINR thresholds are met (Edjekouane et al., 20 Oct 2025).

4. MAC, RLC, and Higher-Layer Protocol Extensions

Full-stack NR-NTN operation requires adaptation of radio resource management, scheduling, and network protocols:

  • HARQ and Scheduling:
  • Scheduling Policies:
    • Semi-persistent schedulers, TTI bundling, and extended protocol timers address high-latency command/ack windows and support periodic or broadcast/multicast services (Guidotti et al., 2022, Lin et al., 2021).
    • Spot beam and SDMA scheduling mitigate resource contention in large footprints.
  • Mobility Management:
    • Beam-centric Earth-fixed cell anchoring separates user-mobility from satellite motion; RRC supports conditional and predictive handover based on satellite ephemeris (Guidotti et al., 2022, Rossato et al., 21 Jan 2026).
    • Paging and registration areas dynamically adapt to satellite motion and coverage footprints.
  • QoS and Slicing:
    • NTN-compliant QoS identifier (NQI) maps multiple 5QI flows into slice aggregates, supporting end-to-end resource optimization over the satellite backhaul despite heterogenous latency/capacity requirements. Slice-based routing via MILP maximizes flow satisfaction and maintains tractable latency gaps under large-scale NTN operation (Abe et al., 10 Feb 2026).

Accurate system design for 5G NR NTN depends on precise link modeling and extensive calibration (Sandri et al., 2023, Figaro et al., 8 Jan 2026):

  • Channel Model:
    • Composite path loss (FSPL, lognormal shadowing, clutter) supplemented with atmospheric absorption (ITU-R P.676), scintillation (ionospheric, tropospheric), and rain fading (ITU-R P.838 at >10 GHz) per 3GPP TR 38.811/38.821 (Sandri et al., 2023).
    • Satellite and user mobility implemented in ECEF coordinates; beam patterns use circular-aperture or UPA models.
  • Simulation Results:
    • Link budgets (see Table below) agree within 1 dB with 3GPP calibration across LEO/GEO, S/Ka bands.
    • System-level simulations (ns3-NTN) demonstrate realistic throughput, PDR, and latency for a range of altitudes, bands, and load scenarios (Figaro et al., 8 Jan 2026).
    • Key findings: GEO Ka-band outperforms LEO S-band in throughput but incurs higher latency; guard periods and TA mismatches are primary sources of performance loss if not compensated.
Case Altitude Band FSPL (dB) SNR (dB) Throughput (Mbps)
LEO, S-band, 30 MHz 600 km 2 GHz 159 6.6 76
LEO, Ka-band, 400 MHz 600 km 20 GHz 179 8.5 300
GEO, S-band, 30 MHz 35,786km 2 GHz 190.6 0 21
GEO, Ka-band, 400 MHz 35,786km 20 GHz 210.6 11.6 450

6. Implementation: SDR-Based Prototyping and AI-Augmented NTN Control

Experimental validation has advanced through fully 3GPP-compliant SDR-based platforms and AI-native open RAN controls:

  • SDR Testbeds: Hardware–software co-design using SDR frontends and optimized GPP hosts (e.g., Amarisoft stack) realizes true GEO links; software adaptations include large frequency shift compensation, extended TA, and HARQ timer scaling, validated with real-world satellite round-trip delay and throughput (Hou et al., 25 Sep 2025).
  • AI-Open-RAN for NTN: Open RAN and AI-RAN architectures (AIO-RAN-NTN) integrate standard O-RAN splits and 3GPP interfaces with real-time AI-driven KPI prediction (e.g., LSTM-based SINR forecasting), allowing proactive scheduling and beam adaptation to counteract mobility and channel variance—even with pedestrian user motion (Do, 14 Nov 2025).
  • Performance: Prototyped GEO SDR systems achieve spectral efficiency of ~1 bps/Hz, RTT >1 s, BLER <1% at 12 dB SNR; AI-RAN controls recover throughput lost to mobility, cutting packet error rates from 8% to 2% in laboratory NTN emulation (Do, 14 Nov 2025, Hou et al., 25 Sep 2025).

7. Evolution Roadmap, Open Challenges, and Future Directions

5G NR NTN support is evolving rapidly through 3GPP releases:

References

Topic to Video (Beta)

No one has generated a video about this topic yet.

Sign Up to Generate All Videos Subscribe on YouTube

Whiteboard

No one has generated a whiteboard explanation for this topic yet.

Sign Up to Generate

Follow Topic

Get notified by email when new papers are published related to 5G NR NTN Support.

Sign Up to Follow Topic by Email