Research
DXN — Deepspace Extensible Network
The Deepspace Extensible Network (DXN) proposes an open, multi-actor protocol ecosystem built on a multi-shell heliocentric relay architecture: an inner backbone ring at 1.0 AU for Earth-Mars traffic, a mid-ring at 1.2 AU providing redundant cross-shell routing and Jupiter approach coverage, and an outer ring at 1.7 AU extending deep-space reach toward the asteroid belt and outer planets.
The Communication Problem
Human and robotic missions are extending beyond Earth orbit at an accelerating pace. Artemis lunar sorties, commercial lunar and Mars landers, outer-planet flagship missions, space telescopes in solar orbit, and proliferating commercial cubesat deployments will all compete for data links simultaneously.
The infrastructural response to date has been incremental expansion of a 60-year-old ground-centric paradigm: more antenna aperture, more scheduling windows, slightly more coordination between agencies. This response is failing. NASA's Office of Inspector General reported in 2023 that DSN demand already exceeds supply by approximately 40%, with the gap projected to reach 50% by the early 2030s. The DSN Aperture Enhancement Project — adding six 34-metre antennas — has cost $706 million and will not complete until approximately 2029, buying perhaps a decade of marginal relief.
The challenges:
- Coverage gaps — planets block line of sight when on opposite sides of the Sun
- Scalability constraints — a limited number of dishes serve all missions
- Single points of failure — Earth-side infrastructure concentration
- Latency — signals travel at light speed, so a Mars-Earth round trip takes 6–44 minutes
Structural problems require structural solutions. DXN proposes that the solution is not more ground aperture but a fundamentally different architecture: a distributed, heliocentric mesh of relay nodes forming a solar-system-scale communications backbone, governed under an open protocol framework that any qualified operator can join.
The Multi-Shell Architecture
The original DXN concept proposed a single ring of six heliocentric relay nodes at 1.0 AU. The revised architecture extends to three concentric shells, each serving distinct coverage and routing functions. The shells are not independent networks but a unified mesh, with inter-shell crosslinks providing redundant paths and geometric diversity.
Shell 1 — Inner Backbone Ring (1.0 AU)
Six nodes in near-circular orbits at 1.0 AU, spaced 60° apart in mean anomaly with mutual inclination of 4° between consecutive nodes to prevent correlated conjunction events. This ring forms the high-availability backbone for Earth-Mars and Earth-Venus communications. Nodes carry Ka-band RF uplinks to planetary gateways and optical inter-satellite links (ISLs) to adjacent Shell 1 nodes and to Shell 2 nodes when geometry permits.
Shell 1 nodes operate at solar irradiance of approximately 1,361 W/m², providing comfortable power budgets for the relay payload. Station-keeping Δv against solar radiation pressure is estimated at 10–20 m/s per year, achievable with a compact xenon ion propulsion system over a 12-year operational life.
Shell 2 — Mid Ring (1.2 AU)
Four to six nodes at 1.2 AU semi-major axis, spaced 60–90° apart. The mid-ring provides three distinct functions: cross-shell routing — a bundle from Earth that cannot reach a Shell 1 node due to conjunction can reach a Shell 2 node visible from both; Mars approach coverage — at 1.2 AU, nodes are geometrically closer to Mars during opposition, reducing Earth-Mars path loss by up to 6 dB; inner asteroid belt proximity — serving as the natural relay infrastructure for asteroid mining operations.
Solar irradiance at 1.2 AU is approximately 945 W/m², remaining highly favourable for solar-powered nodes.
Shell 3 — Outer Reach Ring (1.7 AU)
Three to four nodes at 1.7 AU semi-major axis, in eccentric or slightly inclined orbits to improve geometric diversity. The outer ring extends DXN's reach toward the main asteroid belt and provides the earliest relay capability for outer planet missions. At 1.7 AU, Shell 3 nodes serve as a persistent relay for Mars during conjunction, relay infrastructure for missions to the asteroid belt, and the initial relay hop for future outer-planet extensions.
Solar irradiance at 1.7 AU is approximately 471 W/m², requiring larger solar arrays and potentially small RTGs as supplementary power sources.
Inter-Shell Crosslinks
Each shell broadcasts its node positions and contact availability windows as standardised DXN Contact Announcement Bundles (CABs). The Contact Graph Routing (CGR) algorithm at each node incorporates inter-shell contacts into the routing table, enabling three-hop paths such as: Earth gateway → Shell 1 node → Shell 2 node → Mars gateway.
| Parameter | Shell 1 | Shell 2 | Shell 3 | L-DXN HEO |
|---|---|---|---|---|
| Orbital radius | 1.0 AU | 1.2 AU | 1.7 AU | HEO / EML2 |
| Node count (baseline) | 6 | 4–6 | 3–4 | 4–8 |
| Orbital period | 1.0 yr | 1.32 yr | 2.22 yr | ~6 mo |
| Solar irradiance | 1361 W/m² | 945 W/m² | 471 W/m² | ~1361 W/m² |
| Primary coverage | Earth-Mars backbone | Cross-shell + asteroid belt | Mars far-side + outer reach | Cislunar / GEO+ density |
| Est. unit cost | $200–350M | $220–380M | $280–450M | $80–180M |
Backbone-First vs. Swarm-First
DXN supports two primary deployment strategies that are not mutually exclusive:
Backbone-First deploys a small number (4–8) of high-capability heliocentric relay nodes in defined orbital positions. Each node carries large optical apertures, high-power RF amplifiers, and the full autonomy stack. Routing is deterministic. This approach closely resembles terrestrial long-haul fibre backbone networks — fewer, larger nodes delivering predictable, guaranteed service levels. Its primary advantage is determinism; its vulnerability is concentration of value.
Swarm-First distributes hundreds to thousands of cubesat- to microsat-class relay nodes throughout heliocentric space. Data traverses the network via many shorter hops between neighbouring nodes. Dividing a long link of distance D into N hops of length D/N reduces the path loss per hop by a factor of N². For a 300 million km Earth-Mars link subdivided into 300 hops, this represents a potential 90,000× reduction in per-node transmit power requirement. Each node delivers immediate network value without waiting for a full constellation to be fielded.
The mature DXN architecture comprises four functional layers: (1) planetary gateway nodes; (2) heliocentric backbone relays; (3) a volumetric densification swarm; and (4) a maintenance and logistics layer.
Coverage Analysis
Monte Carlo simulation over 50,000 random epoch-phase combinations shows:
| Configuration | Nodes | Duty Cycle | Max Gap |
|---|---|---|---|
| Shell 1 only — 6 nodes | 6 | ~95% | ~36 hrs |
| Shell 1 (6) + Shell 2 (4) | 10 | ~98% | ~11 hrs |
| Shell 1 (6) + Shell 2 (6) | 12 | ~99% | ~5 hrs |
| Full FOC (S1×6 + S2×6 + S3×4) | 16 | >99.5% | less than 2 hrs |
The 10-node combination is recommended as the Initial Operational Capability target. Even an 11-hour gap during which no path exists does not mean 11 hours of data loss — it means 11 hours during which data is buffered on-node and delivered when contact resumes.
Localised DXN (L-DXN) — Cislunar Subnet
The high-activity zone from Earth's surface to roughly 1.5 million km (the Earth-Moon L2 Lagrange point) requires a dedicated subnet architecture. L-DXN nodes occupy high-Earth orbits and cislunar space:
- EML1/L2 — Earth-Moon Lagrange points with continuous Earth and lunar visibility. L2 provides permanent far-side lunar coverage.
- HEO Molniya/Tundra — 12-hour orbits providing dwell times of 8–10 hours over polar regions per orbit.
- ESL4/L5 — Earth-Sun Lagrange points, gravitationally stable over decades, serving as anchors for the transition from cislunar to heliocentric backbone.
L-DXN nodes are 200–400 kg spacecraft targeting $80–180M unit cost — roughly half that of a Shell 1 backbone node.
Per-Unit Cost Breakdown
| Subsystem | Shell 1 Cost | % of Unit |
|---|---|---|
| Communications payload (RF + optical ISL) | $60–100M | 28–30% |
| Power subsystem | $15–25M | 7–8% |
| Propulsion (ion thruster) | $10–20M | 5–6% |
| Structure, thermal, mechanisms | $15–30M | 8% |
| Avionics, C&DH, flight software | $20–35M | 10% |
| Integration, test, verification | $20–35M | 10% |
| Program management and reserves (25%) | $35–60M | 18% |
| Subtotal — spacecraft development | **$193–338M** | ~85% |
| Launch (Falcon Heavy rideshare) | $2–3M | ~1% |
| Annual operations (10-yr basis) | $15–22M/yr | ~14% |
| Total 10-year TCO per node | **$345–560M** | 100% |
| Node Class | 10-yr TCO | Fleet Size | Fleet TCO |
|---|---|---|---|
| Shell 1 (1.0 AU) | $345–560M | 6 | ~$2.7B |
| Shell 2 (1.2 AU) | $375–620M | 4–6 | ~$2.5B |
| Shell 3 (1.7 AU) | $440–720M | 3–4 | ~$2.1B |
| L-DXN HEO | $160–390M | 4–8 | ~$1.1B |
The dominant cost driver is the communications payload, particularly the precision-pointing optical ISL telescope. This can be reduced through heritage re-use, multi-node procurement (20–35% decline from first-unit to fifth-unit), and commercial optical terminal development.
Deployment Models
Model A — Federated Cooperative
Multiple agencies each deploy and operate backbone nodes under a shared governance framework. No single entity controls the network. Strengths: no single point of failure, open to all participants, politically durable. Weaknesses: high coordination overhead, tragedy of the commons risk.
Model B — Commercial Relay-as-a-Service
A single operator deploys and sells capacity under a government anchor-tenant contract. Strengths: fast deployment, integrated accountability. Weaknesses: single point of commercial failure, pricing monopoly risk.
Model C — Structured Hybrid (Recommended)
Government agencies own backbone nodes (Shell 1 and Shell 3), providing open-protocol capacity under guaranteed SLAs. Commercial operators compete to provide Shell 2 nodes, L-DXN nodes, and value-added services on top. This mirrors the terrestrial Internet: government-funded backbone with commercial ISPs building on top.
| Dimension | Federated | RaaS | Hybrid |
|---|---|---|---|
| Deployment speed | Slow | Fast | Medium |
| Resilience to failure | Very high | Very low | High |
| Interoperability | Mandatory | Uncertain | Mandatory backbone |
| Commercial innovation | Low | High | High |
| Long-term stability | High | Low | High |
Link Budget Analysis
| Parameter | Ka-band: Earth→S1 | Optical ISL: S1→S2 | X-band: S3→Asteroid Belt |
|---|---|---|---|
| Link range | 300M km (max) | 105M km | 200M km |
| Transmit power | 20 kW (DSN 34m) | 1 W optical | 50 W (S3 node) |
| Achievable data rate | ~56 Mbps | ~100–500 Mbps | ~2 Mbps |
| Link margin | ~4 dB | ~7 dB | ~5 dB |
End-to-end Earth-Mars data rates of 10–50 Mbps are achievable via a single relay hop — a 20–100× improvement over current DSN practice. With multiple simultaneous paths, aggregate throughput can exceed 200 Mbps at favourable geometry.
Sustainability
Orbital Stewardship
DXN nodes must carry sufficient propellant for 15-year station-keeping, with end-of-life disposal manoeuvres that park the node in a heliocentric graveyard orbit. Node replacement planning must be built into the governance model from the start.
Spectrum Continuity
DXN uses X, Ka, and optical bands. A new ITU regulatory category — Solar Orbit Space Station (SOSS) — is proposed. DXN should progressively migrate inter-node traffic from RF to optical bands, reducing overall RF spectrum occupation.
Institutional Permanence
The greatest long-term risk is institutional discontinuity. DXN must be designed to survive political changes through treaty anchoring, revenue generation via capacity licensing, distributed ownership (no single nation owning more than 40%), and open standards lock-in.
Heliocentric Asset Servicing
DXN incorporates a dedicated servicing layer: robotic maintenance craft operating within defined orbital bands to provide refuelling, component replacement, node deployment, and controlled retirement services. A fleet of 3–6 servicing craft distributed across the shells provides the logistics backbone for sustained network operations.
Security Architecture
DXN adopts a zero-trust architecture with no implicit trust between nodes. The security model includes:
- Three governance tiers — Foundation, Participating, and User, each with distinct certificate authority scope
- Traffic isolation domains — safety-of-life, mission operations, and commercial, with no cross-domain propagation
- DXN Certificate Authority — Root CA operated jointly by founding members under a threshold-signature scheme (k-of-n)
- Autonomous security response — Anomaly Detection Agents monitoring bundle authentication failure rates, routing table changes, buffer utilisation, and clock drift, capable of quarantining suspect nodes without ground confirmation
Deployment Roadmap
Phase 1 — Foundation (2026–2029)
Establish DXN Coordination Body. Publish Interface Control Document v1.0. Deploy L-DXN nodes at EML1 and EML2. Validate BPv7 routing at lunar distances. Establish DXN PKI.
Phase 2 — Inner Backbone (2029–2034)
Launch Shell 1 nodes 1–6 in three pairs. Commission Ka-band links and autonomous routing. Declare IOC at ~95% Earth-Mars coverage. Open commercial overlay market.
Phase 3 — Multi-Shell Expansion (2034–2040)
Deploy Shell 2 nodes (4–6) by commercial operators. Deploy Shell 3 nodes (3–4) for Mars far-side conjunction bypass. Declare Full Operational Capability.
Phase 4 — Outer System Extension (2040–2045)
Extend DXN reach to Jovian system. Execute Shell 1 replacement programme. Ratify DXN treaty as binding international agreement.
Degraded-Mode Resilience
| Failure Scenario | Active Nodes | Duty Cycle | Max Gap |
|---|---|---|---|
| Nominal IOC (S1×6 + S2×4) | 10 | ~98% | ~11 hrs |
| 1× Shell 1 node failure | 9 | ~97% | ~16 hrs |
| 2× Shell 1 adjacent node failure | 8 | ~91% | ~34 hrs |
| Complete Shell 1 loss | 4 (S2 only) | ~76% | ~60 hrs |
Even complete loss of Shell 1 leaves Shell 2 providing 76% coverage — far better than the current DSN paradigm during conjunction.
Limitations
This paper presents a conceptual framework with first-order quantitative analysis. Full validation requires ray-tracing orbital mechanics simulations, detailed link budget modelling with realistic antenna patterns, and hardware-in-the-loop testing of the BPv7 routing stack. The cost estimates are parametric and require refinement through preliminary design studies.
DXN originated as a fictional communication network in the Tukei-verse and has been developed into an engineering proposal. For the DXN essay with broader context, see DXN — Deepspace Extensible Network.