Optical inter-satellite links (OISL - lasercom crosslinks)

Executive summary

Optical inter-satellite links (OISLs), also called lasercom crosslinks, use free-space optical communication between two spacecraft. The fundamental attractiveness of OISLs relative to radio-frequency (RF) crosslinks is the enormous increase in carrier frequency: optical carriers at ~193 THz (1,550 nm) support very high antenna gains from small apertures and wide modulation bandwidths (Khalighi & Uysal, 2014, DOI: 10.1109/COMST.2014.2329501, access 2026-05-07, doi.org ). The trade-off is extreme pointing difficulty: a spacecraft must aim a beam at another moving spacecraft over hundreds or thousands of kilometers, compensate for relative motion, and maintain acquisition and tracking under platform jitter.

This note finds:

• The physical-layer standard basis for lasercom is the CCSDS Blue Book 141.0-B-1, Optical Communications Physical Layer, issue 1, August 2019 (CCSDS 141.0-B-1, 2019-08, access 2026-05-07, public.ccsds.org ).
• At least one granted patent covers adaptive-optical free-space optical communication hardware (US Patent 9,800,332 B2, “Optical communications system,” granted 2017-10-24, access 2026-05-07, patents.google.com ). This is an IP record, not a performance proof.
• A peer-reviewed survey covers modulation, detection, and channel impairments for free-space optical systems (Khalighi & Uysal, DOI: 10.1109/COMST.2014.2329501, 2014).
• Adoption evidence exists for Amazon’s Project Kuiper OISL test at 100 Gbps over a reported 621 miles (~999 km) between prototype satellites in low Earth orbit (Amazon news release, 2023-12-14, access 2026-05-07, aboutamazon.com ). That performance claim is self-reported and therefore tagged [SINGLE-SOURCE] unless independent test data emerges.
• NASA’s Laser Communications Relay Demonstration (LCRD) provides a government-led technology demonstration context (NASA mission page, access 2026-05-07, nasa.gov ; this URL is a mission landing page and should be treated as a NASA/Tier-2 adoption source, not a peer-reviewed measurement).
• The European Space Agency’s SILEX program demonstrated early optical inter-orbit links (ESA, access 2026-05-07, esa.int ).

A quantitative link-budget example is provided in Section 4. The result is not a guaranteed system performance figure; it is a physics-based worked example using assumed inputs.

Technology architecture

2.1 Basic system model

An OISL terminal has three main functions: transmit a modulated laser beam, receive a beam from the opposite terminal, and maintain line-of-sight alignment through acquisition, tracking, and pointing (ATP). The transmitter typically uses a seed laser, modulator, optical amplifier if available, and a telescope or collimating lens. The receiver uses a telescope, optical filter, detector, and demodulator. On a moving spacecraft, the terminal must include a fine-steering mirror, fast steering mirror, and often a beacon-transmit/receive path for initial acquisition.

The peer-reviewed survey by Khalighi and Uysal establishes the general communication-theory model for intensity-modulation/direct-detection and coherent free-space optical systems, including noise, scintillation, and pointing-error effects (Khalighi & Uysal, 2014, DOI: 10.1109/COMST.2014.2329501, access 2026-05-07, doi.org ). Since space-to-space links avoid atmospheric turbulence, scintillation and aerosol scattering are not the primary channel impairment; acquisition and pointing stability become the dominant engineering problems.

2.2 Wavelengths and modulation

Common lasercom wavelengths are 850 nm, 1,064 nm, and 1,550 nm. The 1,550 nm C-band is attractive because it is compatible with terrestrial fiber-optic component supply chains, erbium-doped fiber amplifiers, and mature coherent receivers. The CCSDS optical physical-layer standard recognizes wavelength regions for space optical communications, though its detailed modulations differ for near-Earth versus deep-space links (CCSDS 141.0-B-1, 2019-08, access 2026-05-07, public.ccsds.org ).

Modulation choices include:

• On-off keying (OOK) with direct detection: simple, but less photon-efficient.
• Pulse-position modulation (PPM): more photon-efficient, especially for deep-space links; high-peak-power laser pulses are used.
• Phase-shift keying (PSK) or differential phase-shift keying (DPSK) with coherent detection: higher receiver sensitivity and bandwidth efficiency, often preferred for high-rate near-Earth and inter-satellite links.

The literature survey covers these modulation trade-offs in detail (Khalighi & Uysal, 2014, DOI: 10.1109/COMST.2014.2329501, access 2026-05-07, doi.org ).

2.3 Acquisition, tracking, and pointing

The patent record shows active development in adaptive optical elements for free-space optical communication. US Patent 9,800,332 B2 describes an optical communications system with adaptive optical components; the claims record technological work on correcting or steering optical beams (US 9,800,332 B2, granted 2017-10-24, access 2026-05-07, patents.google.com ). Again, the patent is not evidence that a specific commercial OISL product meets a performance specification.

Pointing budgets for OISL are tight. For a 1,550 nm beam from a 10 cm transmit aperture, the diffraction-limited half-angle is approximately \( \lambda / (\pi D) \approx 1.55 \times 10^{-6} / (\pi \times 0.1) \approx 4.93 \times 10^{-6} \) radians, or about 5 μrad. At a 3,000 km range, that corresponds to a spot radius on the order of hundreds of meters, so the receiver aperture must be pointed within a small fraction of a beamwidth. This is why fine steering mirrors and beacon-based tracking dominate OISL terminal design.

2.4 Standards and regulatory framing

The CCSDS 141.0-B-1 Blue Book is the main international physical-layer recommendation for optical space communications (CCSDS 141.0-B-1, 2019-08, access 2026-05-07, public.ccsds.org ). It is a standards-body document and therefore a Tier 1 source for protocol and physical-layer framing. However, commercial LEO constellations such as Amazon Kuiper have announced data rates above those traditionally emphasized in the CCSDS deep-space PPM regime (Amazon, 2023-12-14, access 2026-05-07, aboutamazon.com ). This creates a standards-coverage tension discussed in Section 6.

Source review

3.1 Peer-reviewed paper

• Khalighi, M. A., & Uysal, M., “Survey on Free Space Optical Communication: A Communication Theory Perspective,” IEEE Communications Surveys & Tutorials, vol. 16, no. 4, pp. 2231–2258, 2014.

DOI: 10.1109/COMST.2014.2329501

URL: doi.org

Accessed: 2026-05-07.

Status: Peer-reviewed (DOI-verified by issuing publisher).

This source supports the general FSO channel model, modulation comparisons, pointing-error analysis, and receiver sensitivity trade-offs. It is from an academic institution/publisher lineage independent of the commercial vendors.

3.2 Granted patent

• US Patent 9,800,332 B2, “Optical communications system,” granted 2017-10-24 to Facebook, Inc.

URL: patents.google.com

Accessed: 2026-05-07.

Status: Granted US patent (B2 publication), not an application.

This patent is used only as evidence of recorded adaptive-optical claims in the free-space optical communication field. It is not used to prove that any commercial OISL terminal meets a latency, throughput, or reliability specification.

3.3 Standards-body document

• CCSDS 141.0-B-1, “Optical Communications Physical Layer,” Recommendation for Space Data System Standards, Issue 1, August 2019.

URL: public.ccsds.org

Accessed: 2026-05-07.

Status: Standards-body document (Tier 1).

Worked link-budget calculation

The following is a quantitative example for a notional LEO-to-LEO OISL. It uses the standard Friis free-space path-loss equation. This is a physics-derived approximation; it does not include atmospheric loss because both terminals are in space.

4.1 Formula

Received power in dBm:

\[

P_r = P_t + G_t + G_r - L_{fs} - L_p - L_s

\]

Where:

• \(P_t\) = transmit power, dBm
• \(G_t, G_r\) = transmit and receive antenna gains, dBi
• \(L_{fs}\) = free-space path loss, dB
• \(L_p\) = pointing/tracking loss, dB
• \(L_s\) = system/optical/coupling loss, dB

Antenna gain for a circular aperture with diameter \(D\) and wavelength \(\lambda\):

\[

G = 10 \log_{10} \left[ \left( \frac{\pi D}{\lambda} \right)^2 \right]

\]

Free-space path loss:

\[

L_{fs} = 20 \log_{10} \left( \frac{4 \pi R}{\lambda} \right)

\]

4.2 Inputs

4.3 Calculation

\[

G_t = 10 \log_{10} \left[ \left( \frac{\pi \times 0.10}{1.55 \times 10^{-6}} \right)^2 \right]

\]

\[

\pi \times 0.10 / 1.55 \times 10^{-6} = 2.026 \times 10^5

\]

\[

(2.026 \times 10^5)^2 = 4.105 \times 10^{10}

\]

\[

G_t = 10 \log_{10}(4.105 \times 10^{10}) = 106.1\ \text{dBi}

\]

For the 0.20 m receive aperture:

\[

\pi \times 0.20 / 1.55 \times 10^{-6} = 4.052 \times 10^5

\]

\[

(4.052 \times 10^5)^2 = 1.642 \times 10^{11}

\]

\[

G_r = 10 \log_{10}(1.642 \times 10^{11}) = 112.1\ \text{dBi}

\]

\[

L_{fs} = 20 \log_{10} \left( \frac{4 \pi \times 3.0 \times 10^6}{1.55 \times 10^{-6}} \right)

\]

\[

\frac{4 \pi \times 3.0 \times 10^6}{1.55 \times 10^{-6}} = 2.433 \times 10^{13}

\]

\[

L_{fs} = 20 \log_{10}(2.433 \times 10^{13}) = 267.7\ \text{dB}

\]

\[

P_r = 30 + 106.1 + 112.1 - 267.7 - 1 - 3 = -23.5\ \text{dBm}

\]

4.4 Sensitivity and margin

Assume a receiver sensitivity of -30.0 dBm for a 10 Gbps coherent DPSK receiver at a target bit-error rate of \(10^{-9}\). This sensitivity figure is assumed for the worked example and is not independently verified in this note; it should carry the label needs primary source validation if used externally.

\[

\text{Margin} = P_r - P_{sens} = -23.5 - (-30.0) = 6.5\ \text{dB}

\]

4.5 Sensitivity check

The 6.5 dB margin implies that the link could tolerate up to 6.5 dB of additional loss before closing at the assumed sensitivity. In range terms:

\[

20 \log_{10}(R_{max}/3,000\ \text{km}) = 6.5

\]

\[

R_{max} = 3,000 \times 10^{6.5/20} \text{ km} = 3,000 \times 2.11 = 6,330\ \text{km}

\]

If transmit power were reduced to 100 mW (20 dBm), received power would fall to -33.5 dBm and the link would fail the assumed -30 dBm sensitivity by 3.5 dB. This illustrates the system-level trade-off between laser power, aperture size, range, and receiver sensitivity.

Important caveat: This is a link budget, not an operational test. Real OISL terminals include acquisition beacon overhead, detector noise, coding gain, and dynamic range effects not captured here.

Parameter table

Company disclosures vs. marketing claims

6.1 Amazon Project Kuiper

Amazon’s December 2023 announcement states that its prototype Kuiper satellites successfully completed an optical inter-satellite link test at 100 Gbps over a distance of 621 miles (~999 km) between two satellites in low Earth orbit (Amazon, 2023-12-14, accessed 2026-05-07, aboutamazon.com ). That is a company technical-doc/adoption-context source, not an independent measurement.

Amazon’s SEC Form 10-K, accessible from the EDGAR search page for CIK 0001018724, describes Project Kuiper as a long-term initiative requiring substantial capital investment and subject to risks including launch availability, regulatory approvals, and technology execution (Amazon 10-K search page, accessed 2026-05-07, sec.gov ). The 10-K does not repeat the 100 Gbps claim as a guaranteed operational performance figure. Investors should therefore treat the 100 Gbps figure as a self-reported prototype milestone, not a filed contractual commitment.

6.2 Mynaric

Mynaric AG, which sells laser communication terminals, discusses optical terminal products in investor materials (Mynaric investor relations page, accessed 2026-05-07, mynaric.com ). Its SEC filings, accessible through the EDGAR search page for CIK 0001657895, include risk factors related to production ramp, customer qualification, supplier dependencies, and operating losses (Mynaric EDGAR search page, accessed 2026-05-07, sec.gov ). If a vendor’s marketing says “space-proven” while the filing includes material uncertainty about production capacity, that is a disclosure tension, not a settled fact. Specific margins, unit prices, and yield rates are needs primary source validation.

6.3 Contradictions and source-bound claims

• Standards gap: CCSDS 141.0-B-1 emphasizes photon-efficient PPM for deep-space and near-Earth optical links (CCSDS, 2019-08). The Amazon Kuiper prototype’s 100 Gbps LEO-to-LEO link (Amazon, 2023-12-14) is likely outside the traditional high-photon-efficiency PPM regime covered by the Blue Book. Therefore, the commercial LEO OISL data-rate claim should not be read as “CCSDS-compliant” in the narrow sense. This is a legitimate standards-vs-commercial tension, not a fraud allegation.
• Peer-reviewed timeline: The Khalighi & Uysal survey predates large-scale LEO optical crosslink constellations and focuses heavily on atmospheric/turbulence-impaired links (Khalighi & Uysal, 2014). It should not be used to claim that modern LEO OISL terminals have been peer-reviewed at scale.
• Marketing vs. patent record: US Patent 9,800,332 B2 records adaptive-optical claims (US 9,800,332 B2, 2017-10-24). A patent claim is not evidence that a fielded OISL terminal meets a given throughput or reliability. Company press releases are Tier 2 adoption sources; they do not substitute for peer-reviewed measurement.

Adoption and deployment landscape

• NASA LCRD (Laser Communications Relay Demonstration): NASA has demonstrated laser communications relay technology through LCRD, launched in December 2021 (NASA mission page, accessed 2026-05-07, nasa.gov ). This is a US government demonstration and is useful for adoption context, but the mission page is not a peer-reviewed performance database.
• ESA SILEX: SILEX is an early European optical inter-orbit link demonstration that established the feasibility of space-to-space lasercom (ESA page, accessed 2026-05-07, esa.int ).
• Amazon Project Kuiper: Amazon has announced successful prototype OISL testing (Amazon, 2023-12-14, accessed 2026-05-07, aboutamazon.com ). This is self-reported.
• Telesat Lightspeed: Telesat’s Lightspeed constellation is architecturally described as using optical inter-satellite links in company investor materials (Telesat investor relations, accessed 2026-05-07, telesat.com ). Specific supplier terms and technical performance are needs primary source validation.
• SpaceX Starlink: SpaceX has publicly discussed laser inter-satellite links in Starlink satellites, but detailed performance data are not available in a Tier 1 source in this note. SpaceX is not US-listed, so it is excluded from the ticker list below.

Unit economics / business model

Who monetizes OISL

OISL monetization occurs at multiple layers:

1. Terminal hardware suppliers sell laser communication terminals, usually as flight hardware with engineering support. Public examples include Mynaric, which markets lasercom terminals to commercial and government customers (Mynaric investor relations page, accessed 2026-05-07, mynaric.com ). Hardware suppliers may also sell optical head assemblies, controllers, and acquisition/tracking subsystems.

1. Satellite prime contractors and defense integrators monetize OISL through government programs such as Space Development Agency (SDA) transport-layer satellites. This note does not have a stable Tier 1 source for exact SDA contract dollar values in this drafting session; those figures should be marked needs primary source validation until retrieved from SDA or SEC filings.

1. Constellation operators monetize the resulting connectivity: consumer broadband, enterprise/government communications, and capacity leases. Amazon’s Project Kuiper is a long-term initiative whose Form 10-K discloses substantial capital investment and execution risk (Amazon 10-K search page, accessed 2026-05-07, sec.gov ). OISL itself is not typically billed as a separate consumer product; it is an enabling subsystem that reduces ground-station dependence and improves network coverage.

Margin and value capture

No verified per-unit OISL terminal prices or gross margins are available in this note. Mynaric’s SEC filings include risk factors related to production ramp and operating performance (Mynaric EDGAR search page, accessed 2026-05-07, sec.gov ). Therefore, any claim about specific terminal margins would be unverified and should be marked needs primary source validation.

The value capture logic is straightforward: the largest value likely accrues to constellation operators that use OISL to reduce terrestrial gateway costs and improve low-latency connectivity, while terminal suppliers capture hardware revenue subject to qualification risk and manufacturing scale. This qualitative statement is supported by the Amazon 10-K’s description of Kuiper capital intensity and by Mynaric’s filing-risk language.

Disconfirming Evidence

The thesis that OISL/lasercom crosslinks are commercially transformative would be weakened or invalidated by the following:

1. Independent test data contradicting self-reported performance: If a government or academic test program publishes peer-reviewed measurements showing that LEO OISL terminals cannot sustain the 100 Gbps-class rates or the pointing/tracking reliability claimed in vendor marketing, the core performance thesis weakens.
2. Filed disclosure contractions: If Amazon, Telesat, or terminal suppliers file 10-K/20-F disclosures indicating that OISL terminals are failing in orbit, causing constellation write-downs, or delaying service, then marketed adoption would be contradicted by filed financials.
3. Cost per terminal remains prohibitive: If manufacturing yields, qualification costs, or launch insurance prevent OISL terminal production at constellation scale, operators may revert to RF crosslinks or ground-only architectures.
4. Standards and interoperability do not keep pace: If CCSDS and ITU do not update for high-rate LEO ISL waveforms and acquisition protocols, vendor lock-in may reduce buyer willingness to adopt cross-vendor OISLs.
5. Single-source performance claims are never independently reproduced: Until the 100 Gbps Amazon prototype result is corroborated by an independent institution, it remains a single-source favorable data point and cannot support a durable moat/feasibility conclusion.