Low Earth Orbit Mega-Constellations, Civilian-Military Inseparability and The Limits of Targetability under IHL

by:

Din Mashkovich

CyJurII Scholar

on 8 August 2026

Introduction

The legal and strategic debates surrounding the weaponization of outer space frequently center on the limits of international humanitarian law (IHL), specifically how the rule of distinction and the proportionality test apply to dual-use satellite networks. The legal scholarship often leaves the underlying technical mechanisms of this analysis unexamined or underexamined. To move the conversation from abstract legal theory to operational reality, this post explores how the precise engineering and topological design of modern Low Earth Orbit (LEO) constellations lock this legal trap into place. By analyzing the physical routing layers of proliferated LEO (pLEO) backbones, the timing dependencies of terrestrial infrastructure, and the physical impossibility of segregating military and civilian space assets, the structural inevitability of their indispensability becomes clear.

The Principle of Distinction and Dynamic Routing in LEO Meshes

Traditional geostationary (GEO) satellites operate as static, isolated relays in high orbits.[1] In contrast, modern pLEO mega-constellations—such as Starlink, which consists of approximately 10,413 satellites in orbit as of June 2026[2]—function as highly dynamic, fully fledged internet protocol backbones.[3] Orbiting at altitudes below 2,000 kilometers, these satellites move at velocities of approximately 7.5 kilometers per second, requiring a continuously reconfiguring mesh network to maintain global coverage.[4] The physical link layer of this space backbone relies on optical inter-satellite links (OISLs) powered by laser communications terminals.[5] These terminals operate primarily in the infrared spectrum at 1550 nanometers, utilizing specialized solar rejection windows to block background solar radiation and band-pass filters to suppress signal-degrading noise.[6] Directing narrow laser beams between platforms moving at hypersonic relative speeds requires a multi-step pointing, acquisition, and tracking (PAT) sequence.[7] Coarse gimbals first point the laser toward the predicted coordinates of the receiving satellite, followed by fine-sensor acquisition and continuous tracking via fast-steering mirrors.[8]

Because these physical links are constantly appearing and disappearing based on orbital geometry, pLEO backbones cannot utilize reactive terrestrial routing protocols.[9] Instead, the network is modeled as a time-varying topological graph.[10] Ground-plane Software-Defined Networking (SDN) controllers discretize the network timeline into static epochs (snapshots) of approximately 10 seconds.[11] The satellites receive precomputed forwarding tables from the ground and perform simple, low-power table lookups to route packets.[12]

This dynamic mesh architecture is explicitly designed for distributed resilience. If an orbital node is physically or non-kinetically disabled, neighboring satellite routers instantly execute a millisecond-level local fast reroute.[13] This extreme survivability was highlighted in March 2026 when Starcloud integrated Starlink Mini Lasers directly into its orbital data center constellation, enabling space-based AI training across a resilient, ground-independent laser mesh.[14] Consequently, an adversary attempting to achieve a “definite military advantage” under IHL (Article 52(2) of Additional Protocol I[15]) cannot simply neutralize a single satellite. Because the network is self-healing, the physical destruction of a single node produces virtually no operational effect. The attacker is technically forced to disrupt the network systemically—such as by broad cyber-interference or wide-area electromagnetic jamming—which transforms a localized tactical engagement into a global, transboundary service degradation.

Timing Synchronization as a Critical Failure Point for Civilian Survival

While the civilian connectivity and communications are indeed important factors, they are not indispensable for civilian survivability. The true, highly sensitive single point of failure in the orbital domain is timing and synchronization. Terrestrial critical sectors—such as mobile transport networks and electrical grids—rely on satellite-based timing signals to align operations across vast distances.[16] Traditional Medium Earth Orbit (MEO) GNSS signals like GPS are fragile and easily jammed, leading operators to integrate LEO-based Positioning, Navigation, and Timing (PNT) systems, such as Satelles’ Satellite Time and Location (STL) service.[17] STL broadcasts encrypted signals from LEO that are 1,000 times (30 dB) stronger than GPS, allowing them to penetrate deep indoors and safeguard critical infrastructure.[18]

However, due to strict size, weight, power, and cost (SWaP-C) constraints, LEO satellites cannot carry the massive atomic clocks found on MEO platforms. Instead, they carry compact Chip-Scale Atomic Clocks (CSACs) or Oven-Controlled Crystal Oscillators (OCXOs), which drift rapidly and must be continuously synchronized against UTC-traceable ground stations or neighboring satellites via OISLs.[19] When these timing signals are disrupted, the cascading physical consequences on civilian survival are near-instantaneous. The table below illustrates the precise technical dependencies of terrestrial networks on orbital timing synchronization:[20]

These microsecond-level tolerances demonstrate that orbital timing is not merely a digital convenience; it is the physical backbone of modern urban survival. A systemic attack on a LEO network’s synchronization layer directly threatens civilian sustenance by shutting down electrical grids and emergency communications, meeting the technical threshold for an attack on objects indispensable to civilian survival under Article 54 of Additional Protocol I.[21]

The Myth of Separation: Starshield and Spectrum Contamination

To escape the legal constraints of this indispensability issue, states have attempted to separate military and civilian space functions, most notably through SpaceX’s Starshield program.[22] Operating under a classified $1.8 billion government contract since 2021, Starshield constructs purpose-built LEO satellites designed to support national security operations, including Earth observation and secure communications.[23]

However, from an engineering perspective, physical separation remains a structural illusion. Starshield does not deploy a separate physical infrastructure; it is explicitly designed to leverage the commercial Starlink bus, launch capability, and OISL terminals.[24] Starlink’s laser terminal is designed to integrate directly onto partner satellites built by traditional defense contractors like Lockheed Martin, Northrop Grumman, or York Space Systems, merging military sensors directly into the commercial LEO routing backbone.[25] On the ground, military forces utilize standard commercial terminals—such as the U.S. Army’s UAT-222 terminal with “five-terabyte global plans”—to access this shared space backbone, completely blurring the line between civilian hardware and military command-and-control.[26] Furthermore, this physical entanglement leads to spectrum contamination. In October 2025, reports revealed that the Starshield network had begun transmitting downlink communications using spectrum bands internationally reserved for uplinks, violating International Telecommunication Union (ITU) standards and risking severe signal degradation for adjacent civilian and scientific satellites.[27]

This technical reality carries profound legal consequences. Under Article IX of the Outer Space Treaty, states must conduct their space activities with “due regard to the corresponding interests of all other States Parties” and avoid “harmful interference.”[28] By overlapping military communication signals with civilian spectrum bands on shared orbital buses, states make it physically and spectrally impossible for an adversary to target the military capability without disrupting the shared commercial LEO substrate. Any electronic or cyber counter-operation directed at neutralizing Starshield’s military utility will inevitably spill over into civilian networks, triggering the widespread civilian collateral harm that defines the indispensability trap.

Precautionary Design as a Legal Duty

Because physical separation in orbit is technically and financially unfeasible—costing upwards of $10 billion to launch a large, redundant LEO constellation [29]—the focus of space jurisprudence must shift from retrospective proportionality balancing to proactive design precautions. Under IHL, states have an active duty to take precautions against the effects of attacks (Article 58 of Additional Protocol I[30]) by minimizing the exposure of civilian populations and objects under their control to military harm.

In terrestrial telecommunications, the Open Radio Access Network (O-RAN) architecture achieves flexibility by decoupling monolithic base stations into distinct logical units (RU, DU, and CU).[31] Applying similar logical separation principles to LEO networks—such as implementing Software-Defined Wide Area Network (SD-WAN) overlays, logical slicing, and zero-trust routing policies—would allow operators to prioritize and safeguard critical civilian functions (like timing and emergency transit) even during an active cyber or electronic conflict in orbit.

Under the precautionary framework of IHL,[32] states that entangle their operational military workflows with commercial pLEO backbones without integrating these logical protection layers are in violation of their duty to minimize civilian exposure.

Ultimately, the technical architecture of LEO mega-constellations reveals that the indispensability trap is not a temporary legal anomaly, but a permanent feature of modern orbital design. As society’s physical survival becomes increasingly synchronized to the LEO, the legal system must recognize that targeting these integrated networks is no longer a localized military option, but a systemic threat to civilian survivability.

References

[1] See MR Bhavani Shankar and others, ‘Feasibility study of full-duplex relaying in satellite networks’ (2015 IEEE 16th International Workshop on Signal Processing Advances in Wireless Communications (SPAWC), 2015); J Li and others, ‘Resource allocation and interference coordination strategies in heterogeneous dual-layer satellite networks’ (2025) 25(4) Sensors 1005.

[2] See Tereza Pultarova, ‘Starlink satellites: Facts, tracking and impact on astronomy’ (Space.com, 1 June 2026) https://www.space.com/starlink-satellites-facts-tracking-impact-astronomy accessed 11 July 2026.

[3] See H Zhang and others, ‘Source routing for LEO mega-constellations based on bloom filter’ (2025) 24(11) IEEE Transactions on Mobile Computing 12487; ‘Space, the internet’s next frontier’ (The CyberWire: Signals and Space, May 2026) https://thecyberwire.com/newsletters/signals-and-space/10/1 accessed 11 July 2026.

[4] See M Pengnoo, R Maneekut and P Kaewplung, ‘Performance analysis of dynamically routed O-ISLs under PAT constraints in LEO constellations’ (2026) IEEE Access https://doi.org/10.1109/ACCESS.2026.3703432 accessed 11 July 2026; T Sandholm and others, ‘Lightspeed data compute for the space era’ (2026) arXiv:2601.17589 https://arxiv.org/abs/2601.17589 accessed 11 July 2026; A Valentine, I Wakeman and G Parisis, ‘OrbCC: high-throughput and low-latency data transport for LEO satellite networks’ (2025) arXiv:2508.19067 https://arxiv.org/abs/2508.19067 accessed 11 July 2026.

[5] See Wanja de Sombre and others, ‘SKYLINK: Scalable and Resilient Link Management in LEO Satellite Network’ (arXiv:2509.08455v1, 10 September 2025) https://arxiv.org/html/2509.08455v1 accessed 14 July 2026; Mustafa Cardakli, ‘Challenges and Opportunities in Free Space Optical Satellite Communication’ (2026) 44(3) Journal of Lightwave Technology 903; ‘The Fiber-Optic Backbone of Space: Deconstructing Optical Inter-Satellite Links (OISLs)’ (AmiNext, 23 October 2025) https://www.aminext.blog/en/post/optical-inter-satellite-links-laser-communications-space-network-1 accessed 14 July 2026.

[6] ibid. See also Yeji Kim and others, ‘GPS-Based Relative Navigation for Laser Crosslink Alignment in the VISION CubeSat Mission’ (2025) 12(10) Aerospace 928.

[7] ibid.

[8] ibid.

[9] See Zeqi Lai and others, ‘Achieving Resilient and Performance-Guaranteed Routing in Space-Terrestrial Integrated Networks’ (IEEE INFOCOM 2023 - IEEE Conference on Computer Communications, Hoboken, May 2023) 1; J Liu, ‘Intelligent Routing Optimization via GCN-Transformer Hybrid Encoder and Reinforcement Learning in Space-Air-Ground Integrated Networks’ (2025) 15(1) Electronics 14; Li and others, ‘Problems and Requirements of Addressing in Integrated Space-Terrestrial Network’ (Internet-Draft draft-li-istn-addressing-requirement-00, Internet Engineering Task Force 2021) https://www.ietf.org/archive/id/draft-li-istn-addressing-requirement-00.html accessed 14 July 2026.

[10] See Camilo José Rojas Milla, ‘Design, Emulation, and Control of Edge Computing Systems for the Space Cloud in LEO Satellite Networks’ (PhD thesis, University of Genoa 2026); Zhaodi Li and others, ‘Topology Complexity Analysis of Space-Air-Ground Integrated Transportation Communication Networks’ (Preprint, ResearchGate, February 2026); ‘Multi-Attribute Consistency Segment Resilient Routing for LEO Satellite Mega Constellations’ (2025) 24(10) IEEE Transactions on Mobile Computing 10823; Li Zeng and others, ‘Service Function Chain Routing in LEO Networks Using Shortest-Path Delay Statistical Stability’ (arXiv:2603.04361v1, 4 March 2026) https://arxiv.org/html/2603.04361v1 accessed 14 July 2026.

[11] See ‘Toward Scalable SDN for LEO Mega-Constellations: A Graph Learning Approach’ (arXiv:2604.27478v1, 30 April 2026) https://arxiv.org/html/2604.27478v1 accessed 14 July 2026; ‘Perfect Hash-Based Routing Lookup for LEO Constellation Backbone Network’ (ResearchGate, March 2023) https://www.researchgate.net/publication/369000773_Perfect_Hash-Based_Routing_Lookup_for_LEO_Constellation_Backbone_Network accessed 14 July 2026; Siqi Yang and others, ‘CRT: Collision-Tolerant Residence Time for Deterministic Transmission in LEO Satellite Networks’ (arXiv:2605.03382v1, 5 May 2026) https://arxiv.org/html/2605.03382v1 accessed 14 July 2026; Li and others, ‘Problems and Requirements of Addressing in Integrated Space-Terrestrial Network’ (Internet-Draft draft-li-istn-addressing-requirement-00, Internet Engineering Task Force 2021) https://www.ietf.org/archive/id/draft-li-istn-addressing-requirement-00.html accessed 14 July 2026; Li Zeng and others, ‘Service Function Chain Routing in LEO Networks Using Shortest-Path Delay Statistical Stability’ (arXiv:2603.04361v1, 4 March 2026) https://arxiv.org/html/2603.04361v1 accessed 14 July 2026; ‘Optimization Strategy to Solve Transmission Interruption Caused by Satellite-Ground Link Switching’ (ResearchGate, February 2020) https://www.researchgate.net/publication/339245678_Optimization_Strategy_to_Solve_Transmission_Interruption_Caused_by_Satellite-Ground_Link_Switching accessed 14 July 2026.

[12] See Fangzhou Meng and others, ‘Mitigating DDoS Attacks in LEO Satellite Networks Through Bottleneck Minimize Routing’ (2025) 14(12) Electronics 2376.

[13] ibid.

[14] See ‘Starcloud to Integrate SpaceX’s Starlink Mini Lasers Into Its Orbital Data Center Constellation’ (Business Wire, 26 May 2026) https://www.businesswire.com/news/home/20260526670395/en/Starcloud-to-Integrate-SpaceXs-Starlink-Mini-Lasers-Into-Its-Orbital-Data-Center-Constellation accessed 14 July 2026; Jason Rainbow, ‘Starcloud Orders Starlink Lasers for Orbital Data Center Network’ (SpaceNews, 26 May 2026) https://spacenews.com/starcloud-orders-starlink-lasers-for-orbital-data-center-network/ accessed 14 July 2026.

[15] Protocol Additional to the Geneva Conventions of 12 August 1949, and relating to the Protection of Victims of International Armed Conflicts (Protocol I) (adopted 8 June 1977, entered into force 7 December 1978) 1125 UNTS 3 (Additional Protocol I) art 52(2).

[16] See ‘Critical Infrastructure’ (infiniDome) https://infinidome.com/commercial-list/critical-infrastructure/ accessed 14 July 2026; compare ‘Securing Mission-Critical Infrastructure: Why Space-Based Solutions Work’ (Terra Global Solutions, 2 March 2026) https://terraglobalsolutions.com/en/securing-mission-critical-infrastructure-why-space-based-solutions-work/ accessed 14 July 2026.

[17] See Christina Riley, ‘High-Performance and Resilient PNT (Position, Navigation & Timing)’ (Satelles, Inc, 9 June 2023) http://www.satelles.com accessed 14 July 2026; Daniel Burch, ‘How is our network synchronization solution with Iridium® STL technology boosting PNT resilience for L3Harris?’ (Adtran Blog, 11 June 2024) https://www.blog.adtran.com/en/network-synchronization-solution-with-iridium-stl-technology-boosting-pnt-resilience-for-l3harris accessed 14 July 2026; Satelles, Inc, ‘Profile of Responsible Use of Positioning, Navigation, and Timing Services: Response from Satelles, Inc’ (NIST RFI on PNT, Docket ID NIST-2020-0002, 13 July 2020).

[18] See Peter B Johnson, Andrew N Novick and Michael A Lombardi, ‘Measuring the Timing Accuracy of Satellite Time and Location (STL) Receivers’ (2023) Proceedings of the 2023 Precise Time and Time Interval Systems and Applications Meeting 207 https://doi.org/10.33012/2023.18694 accessed 14 July 2026; Satelles, Inc, ‘Public Wireless Supply Chain Innovation Fund Implementation’ (Written Submission to NTIA, DOC/NTIA Docket No 221202-0260, 27 January 2023).

[19] See Manuele Dassié and Gabriele Giorgi, ‘Relativistic Modelling for Accurate Time Transfer via Optical Inter-Satellite Links’ (2021) 100(1) Aerotecnica Missili & Spazio https://www.researchgate.net/publication/353777087_Relativistic_Modelling_for_Accurate_Time_Transfer_via_Optical_Inter-Satellite_Links accessed 14 July 2026; Lei Mu and others, ‘High-Precision Time Synchronization and Autonomous Maintenance for LEO Satellite Constellations Based on High-Stability Crystal Oscillators’ (2026) 26(6) Sensors 1839.

[20] Institute of Electrical and Electronics Engineers, IEEE Standard for Synchrophasor Measurements for Power Systems (IEEE Std C37.118.1-2011, 2011).

[21] Additional Protocol I, art 54.

[22] SpaceX, ‘Starshield’ (SpaceX) https://www.spacex.com/starshield accessed 14 July 2026.

[23] See Ty Roush, ‘SpaceX Developing Spy Satellite Network for the U.S., Report Says’ (Forbes Australia, 16 March 2024) https://www.forbes.com.au/news/billionaires/spacex-developing-spy-satellite-network-for-the-u-s-report-says/ accessed 14 July 2026.

[24] SpaceX (n 22).

[25] ibid. See also Mikayla Easley, ‘SDA Demos Laser Link Between 2 Vendors for Future SATCOM, Missile Tracking Network’ (DefenseScoop, 9 January 2025) https://defensescoop.com/2025/01/09/spacex-york-space-systems-sda-pwsa-tranche-0-laser-link-demonstration/ accessed 14 July 2026.

[26] See ‘SpaceX’s Starshield Enhances Military Communication with Faster, Safer and More Efficient Technology’ (Defence Industry Europe, 10 August 2025) https://defence-industry.eu/spacexs-starshield-enhances-military-communication-with-faster-safer-and-more-efficient-technology/ accessed 14 July 2026; Xavier Chavez, ‘Army Reserve Soldiers use SpaceX’s Starshield technology for faster, more convenient military communication’ (U.S. Army Reserve, 4 August 2025) https://www.usar.army.mil/news/images/igphoto/2003778327/ accessed 14 July 2026.

[27] See Evrim Ağacı, ‘SpaceX Starshield Satellites Found Violating Radio Rules’ (Evrim Ağacı) https://evrimagaci.org/gpt/spacex-starshield-satellites-found-violating-radio-rules-511094 accessed 14 July 2026, quoting Kevin Gifford: ‘there’s no concrete evidence that other satellites have suffered major disruptions...’.

[28] Treaty on Principles Governing the Activities of States in the Exploration and Use of Outer Space, including the Moon and Other Celestial Bodies (adopted 27 January 1967, entered into force 10 October 1967) 610 UNTS 205 (Outer Space Treaty) art IX.

[29] See, eg, Jonathan Liebenau, ‘Historical Reflections & an Economic Approach to LEOs as Infrastructure’ in Global Governance of Low Earth Orbit Satellites (2025) 34 https://doi.org/10.18778/8331-719-9.05 accessed 14 July 2026; House of Lords Select Committee on UK Engagement with Space, ‘Uncorrected Oral Evidence: Evidence Session No. 8’ (31 March 2025) https://www.parliamentlive.tv accessed 14 July 2026.

[30] Additional Protocol I, art 58.

[31] See Supermicro, ‘What is Open RAN?’ (Supermicro Glossary) https://www.supermicro.com/en/glossary/open-ran accessed 14 July 2026; Hewlett Packard Enterprise, ‘What is Open RAN?’ (HPE) https://www.hpe.com/us/en/what-is/open-ran.html accessed 14 July 2026; Rimma Iontel, ‘Decomposition, virtualization, and open interfaces: An introduction to Cloud RAN’ (Red Hat Blog, 16 March 2023) https://www.redhat.com/en/blog/cloud-ran-intro accessed 14 July 2026.

[32] Additional Protocol I, art 57.