Open simulation and systems analysis for a fault-tolerant Mars orbital relay constellation.
Future Mars surface operations—human or robotic—will require continuous, low-latency, and resilient communications coverage. Today’s Mars relay architecture depends on a small number of aging orbiters performing multiple roles simultaneously, creating inherent single-point-of-failure risks and constraining operational scale.
Sentinel Network explores a different approach: a dense, fault-tolerant orbital relay layer designed explicitly as infrastructure rather than as a set of multi-mission flagship orbiters.
In Sentinel, each satellite functions as a communications relay, providing line-of-sight, store-and-forward connectivity between Mars surface assets, orbital users, and Earth relay paths.
This repository contains open simulations and systems-level analyses evaluating the feasibility, performance, and failure tolerance of such a network.
Mars communications today face several structural limitations:
- Reliance on a small number of critical relay assets
- Increasing data demand from surface operations
- Limited fault tolerance and graceful degradation
- Growing latency constraints for autonomous systems
As surface activity scales, data transport, not sensing or computation, becomes the bottleneck.
Sentinel Network is a conceptual Mars orbital relay constellation designed around the following principles:
- Fault tolerance by design
No single satellite is mission-critical; resilience is achieved at the network level. - Graceful degradation
Network performance degrades predictably under partial or catastrophic loss. - Infrastructure-first architecture
A payload-agnostic relay layer supporting multiple missions and users. - Scalable deployment
Constellation sizing and orbital regimes selected for long operational lifetimes and volumetric efficiency.
- Constellation size: 168 satellites
- Function: Communications and navigation relay
- Coverage: Global Mars surface coverage
- Performance focus: Revisit time, availability, and resilience
- Deployment philosophy: Bulk deployment compatible with high-capacity launch systems, but not dependent on any specific launch vehicle
- Operational lifetime: Long-duration orbital regimes (multi-decade class)
Quantitative performance results and failure-mode analyses are documented in the simulations and supporting documentation.
Sentinel-Network/
├── README.md — Project overview and context
├── LICENSE — MIT License
├── simulations/ — Coverage, revisit, and loss-tolerance modeling
├── docs/ — Architecture summary and modeling scope
│ ├── architecture_summary.md
│ └── modeling_scope.md
├── figures/ — Generated plots and constellation visuals
└── data/ — Constants and reference parameters
Fault tolerance is achieved at the constellation level through redundancy, spatial diversity, and orbital geometry rather than through complex, internally fault-tolerant satellites.
Satellite loss—whether isolated, clustered, or catastrophic—is treated as an expected operating condition. Network performance degrades in a predictable and quantifiable manner as nodes are lost, rather than experiencing abrupt failure modes.
This project intentionally does not attempt to:
- Replace high-data-rate science orbiters or flagship missions
- Specify detailed RF link budgets, modulation schemes, or hardware designs
- Optimize for mission-specific payloads or proprietary architectures
- Address surface terminal or ground network implementation details
The focus is on first-order system behavior and architectural feasibility.
Sentinel Network is intended to complement existing Mars orbiters by providing a dedicated, resilient communications infrastructure layer optimized for availability and continuity.
Science orbiters and specialized missions would continue to perform high-value sensing and data generation roles, while Sentinel provides a persistent relay backbone capable of supporting scaled surface operations and multi-user demand.
Active and evolving.
- Core simulations: complete
- Documentation: in progress
- Higher-fidelity modeling: planned
- External review and iteration: encouraged
This repository represents an open systems study, not a finalized design or flight program.
This project is released under the MIT License.
Reuse, modification, and extension are encouraged with attribution.
This work is an independent technical exploration.
It is not affiliated with, endorsed by, or representative of any space agency or commercial entity.