---
id: "lesson-03-05"
track: "propulsion-navigation"
slug: "communications-across-years"
title: "Communication across years"
summary: "Design for light-time, disruption, scarce data, secure custody, and eventual independence."
minutes: 22
level: "Applied"
preparedBy: "GShips Project"
lastEditedAt: "2026-07-26"
conflicts: "Maintainer intends to explore a commercial venture based on some GShips work; no entity, funding, customer, sponsor, or partner relationship currently exists."
reviewRequiredDomains: "RF and optical communications, information theory, delay-tolerant networking, cybersecurity and cryptography, digital preservation, autonomy, institutional governance"
claimIds: "claim-01-05, claim-12-05, claim-13-02, claim-13-08, claim-13-10"
---
# Communication across years
> **Evidence boundary:** Deep-space radio networks, optical-communications demonstrations, and delay/disruption-tolerant networking are operational or demonstrated within the Solar System. No link has carried operational traffic across interstellar distance, and no institution has maintained an authenticated, interpretable conversation across generations. Data-rate records at planetary distance must not be extrapolated to light-years without a complete link, pointing, power, time, and maintenance budget.
## Plain-language summary
At interstellar distance, communication becomes correspondence. A message to a system `4.25 light-years` away takes at least `4.25 years` to arrive; the earliest physical response takes `8.5 years`. Earth cannot steer a ship in real time, resolve emergencies, or remain the final authority.
Distance also makes the signal faint. A credible link specifies power, wavelength, apertures, pointing, noise, coding, data rate, contact duration, and margins. “Use a laser” is not a budget.
Architecture is local-first. Each community must carry the knowledge, governance, trust anchors, and repair capacity for independence. Communications can exchange science, culture, warnings, software, and memory; they cannot substitute for local competence or consent.
## Light-time is a governance fact
For range `R`, minimum one-way delay is:
`t = R / c`
where `c` is the speed of light. Because a light-year is the distance light travels in one year, a target at `4.25 light-years` has a minimum one-way delay of `4.25 years`.
This changes institutions:
- emergencies are resolved locally;
- requests and answers can cross administrations and generations;
- contracts cannot assume prompt acknowledgment;
- Earth-based experts cannot remain required operators;
- command authority must be constitutionally local, not merely delegated during outages;
- messages need durable context, provenance, units, software versions, and expiration rules.
A sender should write for a receiver whose language, tools, and personnel may have changed. “Install this update” is inadequate. A safe transfer includes inspectable source, build instructions, tests, threat and hardware assumptions, provenance, and a locally authorized decision. Under GShips’ zero-dependency principle, core tools should rebuild from archived first-party source and documented platform primitives.
## Why signals fade
For an ideal line-of-sight radio link, the Friis relation is:
`P_r = P_t G_t G_r (λ / 4πR)²`
where `P_r` is received power, `P_t` transmitted power, `G_t` and `G_r` antenna gains, `λ` wavelength, and `R` range. Optical links use a different detailed treatment, but aperture, wavelength, diffraction, pointing, and background still govern how many photons reach the receiver.
Holding frequency, apertures, and all other terms fixed, increasing distance from `1 AU` to `4.25 light-years` multiplies range by about `268,700`. The additional free-space path loss is:
`20 log₁₀(268,700) ≈ 108.6 dB`
That means received power is about `72 billion` times smaller in this simplified comparison. Larger apertures, narrower beams, more power, longer integration, and better coding can recover performance, but add costs and failure modes.
At `1550 nm`, one photon has energy:
`E_γ = hc / λ ≈ 1.28 × 10⁻¹⁹ J`
Detectors have inefficiency and dark counts; celestial light adds background; pointing controls illumination; Doppler and clock error affect acquisition; and error correction adds overhead. A proposal should publish detected photons per information bit, not just beam power.
## Data rate is an integrated choice
The Shannon capacity of an ideal additive white Gaussian-noise channel is:
`C = B log₂(1 + S/N)`
where `C` is theoretical channel capacity, `B` bandwidth, and `S/N` signal-to-noise ratio. Real links operate below this bound because coding is finite, channels vary, pointing is imperfect, hardware has losses, and operational margin is required.
A reviewable link budget includes:
- transmitter output and wall-plug power;
- apertures, efficiency, beam pattern, and wavelength;
- pointing knowledge, control, and jitter;
- propagation and atmospheric loss and occultations;
- detector noise and celestial background;
- modulation, coding, and frame overhead;
- acquisition, duty cycle, weather, maintenance, and redundancy;
- useful data rate at a stated error probability and margin.
“Peak rate” is not sustained useful delivery. The mission question is how many authenticated, corrected, interpretable bits arrive per year after outages and maintenance.
## What current demonstrations establish
NASA’s Deep Space Network provides command, telemetry, navigation, and science services across Solar System missions. Its handbook documents real interfaces and performance. It does not demonstrate an interstellar link.
NASA’s Deep Space Optical Communications experiment flew with Psyche and demonstrated high-rate optical downlinks at progressively larger planetary distances. JPL reports a peak `267 Mbps` at about `31 million km` and operation at a record distance of about `494 million km`. It returned more than `13.6 terabits` of data during the demonstration. These are major results in acquisition, pointing, coding, and ground reception. They do not establish the power, aperture, pointing, or rate of a gram probe or habitat across light-years.
These results should calibrate models and the next experiment, not be quoted without distance, terminals, weather, geometry, and date.
## Delay/disruption-tolerant networking
Internet protocols often assume relatively short round trips and a contemporaneous end-to-end path. Delay/Disruption-Tolerant Networking instead stores data in “bundles” and forwards it when a scheduled or opportunistic next contact becomes available. Bundle Protocol Version 7, standardized in RFC 9171, defines this overlay architecture. Bundle Protocol Security, RFC 9172, defines integrity and confidentiality blocks for bundles; RFC 9173 defines default security contexts.
NASA reports bounded DTN demonstrations and operational space deployments. Benefits for disaster regions, remote science, rural networks, critical infrastructure, or sensor networks remain application-specific and require comparative field evidence.
DTN does **not** defeat light-time, create a path, supply energy, aim an antenna, preserve custodians, or make old cryptography safe. Store-and-forward still needs durable storage, contact plans, routing and forwarding policy, congestion control, replay handling, clocks, and priority rules. BPSec does not supply key establishment, exchange, revocation, security policy, or protection from a compromised implementation; those remain separate system responsibilities.
Priority is political: warnings, personal messages, science, archives, telemetry, and software may compete. Allocation rules should be published, appealable, and resistant to status capture.
## Security across generations
A century-scale trust system cannot assume that today’s algorithms, keys, certificate authorities, vendors, or nation-states remain valid. It needs:
- local root-of-trust governance and multiple authorized custodians;
- algorithm agility and migration procedures;
- key rotation, revocation, recovery, and succession;
- authenticated time or explicit operation when trusted time is uncertain;
- replay protection despite long delays and duplicated bundles;
- separation of receiving, reviewing, simulating, and commanding;
- quarantine for data and software from any source, including Earth;
- signed, content-addressed archives with independent checks;
- recovery exercises after deliberate loss of keys and directories.
A delayed message may be authentic but unsafe because assumptions, vulnerabilities, or system state changed. Authentication establishes origin and integrity, not current relevance, truth, authority, or consent.
LLMs can summarize, translate, recover context, or triage logs, but create prompt-injection, provenance, hallucination, and reproducibility risks. External text is untrusted data. An LLM should not execute messages, rotate keys, allocate emergency bandwidth, or sign commands. Critical decoders, verification, routing, and archives need inspectable offline implementations and tests.
## Archives are part of the link
Receiving bits is not preserving meaning. The CCSDS Open Archival Information System model emphasizes the representation information a designated community needs to understand preserved data.
For intergenerational correspondence, retain:
- raw received symbols or frames where feasible;
- decoded payload plus checks and correction history;
- protocol and file-format specifications;
- character encodings, units, schemas, and vocabularies;
- software source, compiler or interpreter specifications, and test vectors;
- provenance, signatures, confidence, and access restrictions;
- emulators or migration tools for obsolete media and hardware;
- multiple physical copies with routine fixity and restore tests.
The archive should be recoverable without original librarians, vendors, cloud models, or one medium, while preserving restrictions on personal or hazardous material.
## Eventual independence
An interstellar community cannot be a remotely managed outpost. Communications policy should distinguish:
- **advice**, which recipients may evaluate;
- **shared observations**, which retain uncertainty and method;
- **requests**, which can be refused;
- **cultural correspondence**, governed by privacy and consent;
- **safety alerts**, which demand rapid local review but not blind execution;
- **commands**, which should generally have no standing interstellar authority over a self-governing population.
This is not abandonment. It is honest design under physics. Earth and ship can remain related while recognizing that each must act without timely rescue and that people born aboard did not consent to permanent remote subordination.
## An Earth-first test program
A credible ladder would operate real delayed networks rather than merely simulate latency:
1. deploy DTN between remote communities and scientific stations with intermittent links;
2. measure useful delivery, energy, storage wear, congestion, and recovery—not just packet throughput;
3. run multi-year archives with format migration and independent restore teams;
4. inject clock loss, compromised keys, corrupt bundles, revoked authorities, and missing vendors;
5. test optical and radio terminals at increasing range with full published link budgets;
6. allocate scarce bandwidth through transparent human governance exercises;
7. require critical software to rebuild offline from first-party source and test vectors;
8. demonstrate that local operators remain safe when Earth is silent.
These tests can produce evidence relevant to resilient terrestrial communications and Solar System missions regardless of the interstellar conclusion. Any claimed terrestrial benefit still needs a named setting, comparator, users, outcome measures, and sustained field evidence.
## Evidence ledger
- **L03-05-A — Deep-space radio communications are operational throughout the Solar System, and optical deep-space links have been demonstrated at planetary distances.** Basis: observed. Readiness: operational for radio; demonstrated and scaling for optical. Confidence: strong.
- **L03-05-B — Free-space loss and light-time make an interstellar link categorically different from a planetary-distance record unless power, aperture, pointing, rate, and margin are reclosed.** Basis: physical model. Readiness: no interstellar operational link. Confidence: strong.
- **L03-05-C — Bundle Protocol is a normative store-carry-forward specification, BPSec is a normative integrity-and-confidentiality specification, and bounded space deployments exist; neither standard establishes end-to-end delivery or terrestrial benefit.** Basis: normative standards plus observed bounded deployment. Readiness: operational in bounded deployments. Confidence: strong for scope and supported for deployment transfer.
- **L03-05-D — Century-scale communications require local cryptographic migration, archive recovery, and governance that have not been demonstrated as an integrated institution.** Basis: systems assessment. Readiness: components operational; integrated duration unverified. Confidence: supported.
- **L03-05-E — An interstellar community must be operationally and politically capable of acting without timely Earth command.** Basis: physical constraint and normative governance conclusion. Readiness: institution not demonstrated. Confidence: strong for delay; supported and contestable for governance design.
Linked corpus claims: `claim-01-05`, `claim-12-05`, `claim-13-02`, `claim-13-08`, and `claim-13-10`. See the claim registry for each record's current evidence grade and independent-review state.
## Assumptions and limits
The path-loss comparison holds frequency, apertures, pointing, and other terms fixed; it is not a designed link. The photon calculation does not include detector or coding performance. The Alpha Centauri distance is rounded. Published mission records and standards can change. Cybersecurity discussion is architectural and omits implementation-specific threats. OAIS is a conceptual archival model, not a turnkey preservation system. Political independence is a normative conclusion open to structured review, not a claim that cultural ties should end.
## What would change this conclusion?
Technical readiness would rise after end-to-end links demonstrated acquisition, authenticated useful delivery, outage recovery, and maintainable terminals at increasingly representative range and power, with independently reproduced budgets. Institutional confidence would rise after multi-decade archive and key-migration exercises succeeded without original experts, vendors, cloud services, or one storage medium. It would fall if pointing, weather, component aging, storage wear, cryptographic migration, or archive interpretation produced unrecoverable gaps. Higher peak optical rate at Solar System distance would update one subsystem, not erase light-time or the need for local authority.
## Sources and locators
- [S01 — NASA Jet Propulsion Laboratory, “Deep Space Network Telecommunications Link Design Handbook”](https://deepspace.jpl.nasa.gov/dsndocs/810-005/). Locator: current modules for telecommunications link design, ground-system performance, frequency and interface constraints; current issue page dated 2026-05-22; accessed 2026-07-25.
- [S02 — NASA Jet Propulsion Laboratory, “Deep Space Optical Communications (DSOC)”](https://www.jpl.nasa.gov/missions/deep-space-optical-communications-dsoc/). Locator: technology-demonstration overview, peak data rate, cumulative data, and maximum-distance results; accessed 2026-07-25.
- [S03 — NASA, “Delay/Disruption Tolerant Networking”](https://www.nasa.gov/communicating-with-missions/delay-disruption-tolerant-networking/). Locator: operational rationale, store-and-forward bundles, mission uses, and High-Rate DTN demonstration; accessed 2026-07-25.
- [S04 — Internet Engineering Task Force, RFC 9171, “Bundle Protocol Version 7”](https://www.rfc-editor.org/rfc/rfc9171.html). Locator: Sections 1–5, DTN architecture, bundle format, and processing; January 2022.
- [S05 — Internet Engineering Task Force, RFC 9172, “Bundle Protocol Security (BPSec)”](https://www.rfc-editor.org/rfc/rfc9172.html). Locator: Sections 1.1–1.2, 3, 6–9 on supported services, scope, security blocks, key-management boundary, policy, threats, and security contexts; January 2022.
- [S05a — Internet Engineering Task Force, RFC 9173, “Default Security Contexts for Bundle Protocol Security”](https://www.rfc-editor.org/rfc/rfc9173.html). Locator: Sections 1–2 and 4–6 on default integrity and confidentiality contexts and their interoperability scope; January 2022.
- [S06 — CCSDS, “Reference Model for an Open Archival Information System (OAIS)”](https://ccsds.org/searchpubs/entry/3054/). Locator: CCSDS 650.0-M-3 publication record and linked reference model on information packages, representation information, and preservation functions; December 2024; accessed 2026-07-25.
- [S07 — NASA, “Delay/Disruption Tolerant Networking Overview”](https://www.nasa.gov/reference/delay-disruption-tolerant-networking-overview/). Locator: architecture overview and distinction from conventional end-to-end Internet assumptions; accessed 2026-07-25.
## Editorial record
- Prepared by: GShips Project
- Last edited: 2026-07-25
- Last independently reviewed: Not yet reviewed; no review date
- Required review: RF and optical communications; information theory; delay-tolerant networking; cybersecurity and cryptography; digital preservation; autonomy; institutional governance
- Conflicts: Maintainer intends to explore a commercial venture based on some GShips work; no entity, funding, customer, sponsor, or partner relationship currently exists
- Corrections: [Suggest a correction](https://gships.dammonburden.com/corrections)