A structured research environment for radio astronomy, AI-driven signal processing, and advanced IT automation.
Table of Contents
The Proxmox Astronomy Lab is a high-performance research environment designed for Hydrogen Line Radio Astronomy, AI-driven SDR signal processing, and secure, remote collaboration. The lab combines Proxmox, Kubernetes, AI/ML, and modern IT automation to create a reproducible, scalable research platform built on enterprise-grade practices and open-source technologies.
At its core, the lab is a fusion of IT engineering and citizen science—designed not just for scientific discovery, but also as a structured and documented template for others to replicate. Everything here is designed with reproducibility, automation, and high-performance computing in mind.
The lab is built around three interconnected areas that form its foundation.
| Component | Description |
|---|---|
| 📡 Hydrogen Line Radio Astronomy | AI-enhanced SDR processing for HVCs, LSBJs, and SNR discovery. A research-grade SDR observation station built for precision Doppler shift analysis and long-term hydrogen line tracking. |
| 🔐 Secure Remote Research & IT Lab | External researchers & IT professionals can securely access the lab for radio astronomy data analysis, AI workloads, and IT infrastructure testing, with policy-driven access controls. |
| 📖 Reproducible Documentation & AI Integration | Comprehensive GitHub-based documentation and public-facing research notes to help others replicate a high-performance, AI-driven research environment. AI-powered Retrieval-Augmented Generation (RAG) allows for contextual querying of documentation, policies, and workflows. |
This repository is organized into the following main directories and files:
| Directory/File | Purpose |
|---|---|
| 📁 assets | Images, diagrams, and visual resources for documentation |
| 📁 docker | Docker configurations, Portainer stacks, and container definitions |
| 📁 docs | Primary documentation organized by category |
| 📁 entra-hybrid-cloud | Entra ID integration and hybrid identity management |
| 📁 infrastructure | Core infrastructure components and configuration |
| 📁 itil | IT service management documentation and processes |
| 📁 k8s-rancher-rke2 | Kubernetes configuration and workloads |
| 📁 lab-services | Documentation for lab services and applications |
| 📁 monitoring | Monitoring stack configuration and dashboards |
| 📁 observatory-and-projects | Radio astronomy projects and research |
| 📁 wiki | Knowledge base articles and guides |
| 📄 phase-1.md | Core Infrastructure Foundation documentation |
| 📄 phase-2.md | Structured Services & Research Validation |
| 📄 phase-3.md | Application Deployment & Research Infrastructure |
| 📄 phase-4.md | Research Workflows & Public Data Integration |
| 📄 ROADMAP.md | Project roadmap and phase planning |
The docs directory contains detailed documentation organized by function:
| Docs Subdirectory | Content |
|---|---|
| 📁 Applications | Documentation for all applications and services |
| 📁 Compliance-Security | Security frameworks, policies, and CIS controls |
| 📁 Control-Plane | Core infrastructure management services |
| 📁 Documentation-Standards | Templates and style guides for documentation |
| 📁 Entra-Hybrid-Cloud | Microsoft Entra ID and Azure integration |
| 📁 Infrastructure | Hardware, networking, and virtualization details |
| 📁 ITIL-Processes | IT service management procedures |
| 📁 Research-Projects | Scientific research methodologies and datasets |
Radio astronomy has traditionally required large, expensive facilities—but modern Software-Defined Radio (SDR) technology, AI, and advanced computing are changing that. This lab is designed to push the limits of what's possible in a home-based, research-grade Hydrogen Line observation setup.
The primary scientific focus is on three key research domains that leverage hydrogen line observations.
HVCs are massive interstellar clouds moving at speeds different from normal galactic rotation. Studying them can reveal insights into the formation of galaxies and the cosmic web. This lab aims to track the movement of these clouds over time using AI-enhanced Doppler shift analysis.
LSBJs are some of the most elusive objects in the universe—they have very little visible light but contain significant hydrogen gas. This project seeks to detect their presence using Hydrogen Line emissions, helping to map faint galactic structures.
A major challenge in amateur radio astronomy is weak signals buried in noise. This lab uses AI-enhanced noise reduction techniques to improve the clarity and reliability of Hydrogen Line observations.
The radio astronomy equipment chain is designed for optimal hydrogen line detection and analysis.
| Component | Hardware | Purpose |
|---|---|---|
| 📡 Antenna | Nooelec Hydrogen Line Parabolic (20 dBi) | Captures Hydrogen Line emissions at 1.42 GHz. |
| 📡 Pre-LNA Filter | BP-2 Band-Pass Filter | Filters out unwanted RF interference before amplification. |
| 📡 LNA (Amplifier) | 1420 MHz Cavity LNA (34 dB Gain) | Boosts weak Hydrogen Line signals. |
| 📡 SDR (Receiver) | Airspy R2 | High dynamic range SDR for precise Hydrogen Line analysis. |
| 📡 Clock Source | GPS-Disciplined Oscillator (GPSDO) | Provides precise timing for Doppler shift calculations. |
| 📡 Edge Processing | N100 Mini PC | First-stage SDR signal processing before lab transfer. |
The table above details each component in the signal chain from antenna to initial processing.
Data from the SDR hardware is processed through multiple stages for analysis and storage.
- SDR captures real-time 1420 MHz Hydrogen Line data
- Data is processed using AI noise filtering and spectral enhancement
- Results are stored in PostgreSQL & TimescaleDB for time-series tracking
The observatory has several planned enhancements to improve capability and precision.
- Motorized tracking mount for targeted observations
- Upgraded 1.2m dish for higher gain and better resolution
- Integration with OpenSpace & AI-enhanced spectral classification
One of the unique aspects of this lab is that it's not just for personal research—it also functions as a secure research platform that allows external users to collaborate, test, and run workloads remotely.
The lab supports multiple external user workflows for research collaboration.
- Run SDR processing workloads remotely
- Test AI-driven workflows in Kubernetes
- Work with structured datasets & time-series astronomy data
The lab implements several security measures to protect research data while enabling collaboration.
| Security Feature | Implementation |
|---|---|
| Tailscale with Entra SCIM | Secure remote access controlled by Entra ID policies and ACLs by group |
| Role-Based Access Control (RBAC) | Limits access to different lab functions |
| Virtualized Research Workspaces | Kasm Workspaces for browser-based, secure research |
| CISv8 Compliance | Ubuntu Server 24.04 LTS @ CISv8 L2, Windows Server 2025 Standard @ CISv9 L1 compliance |
| Wazuh SEIM/XDR | Daily CIS scans via CIS-CAT Lite w/controls mapped to NIST/ISO27001 |
The security features above ensure that external collaborators can access resources safely while maintaining data integrity.
The Proxmox Astronomy Lab is not just about research—it's about documenting everything in a way that others can follow and replicate. Our documentation covers everything from infrastructure and security to research methodologies and data processing workflows.
The knowledge base is organized to support both human navigation and AI-powered retrieval.
- GitHub Repository: All infrastructure, scripts, and workflows are public and version-controlled.
- Step-by-step guides: From infrastructure deployment to SDR processing.
- Security & Compliance: Complete CISv8 implementation with mappings to NIST and ISO 27001.
- ITIL Processes: Structured change management, incident response, and service catalogs.
- Research Methodologies: Detailed procedures for Hydrogen Line observation and analysis.
One of the biggest challenges in complex projects is finding relevant documentation quickly. This lab uses AI-powered document retrieval to allow natural language queries on:
- Research papers & Hydrogen Line methodology
- Infrastructure configurations & compliance policies
- Custom workflows for AI-powered SDR processing
HPC Proxmox GPU Node w/RTXA4000 GPU, 5950X and 128GB of RAM
The result is a context-aware, AI-enhanced research assistant that helps users navigate and understand the lab's resources efficiently.
The lab is built on a high-performance infrastructure stack optimized for research workloads.
| Component | Specifications |
|---|---|
| Compute Nodes | 3 × Ryzen 5700U (64GB RAM, 2TB NVMe) |
| High-Performance Node | Ryzen 5950X (128GB RAM, 4TB NVMe, RTX A4000) |
| Storage | Proxmox Node05, ZFS 8×8TB HDD RAID10 + NVMe partition as a SLOG |
| Network | Dual 10G SFP for high-speed data transfer |
The table above outlines the key hardware components that make up the lab's infrastructure backbone.
The storage architecture is optimized for performance and scalability for research workloads.
✅ All AI/ML & K8s workloads run on local NVMe storage for maximum performance.
✅ Network storage (NFS & S3) provides fast, scalable research data access.
The lab is being implemented in structured phases, with clear milestones and documentation for each.
| Phase | Focus | Status | Key Deliverables |
|---|---|---|---|
| Phase 1 | Core Infrastructure Foundation | ✅ Complete | Proxmox cluster, network segmentation, security baseline |
| Phase 2 | Structured Services & Research Validation | ✅ Complete | ITSM framework, monitoring stack, initial SDR validation |
| Phase 3 | Application Deployment & Research Infrastructure | 🚧 In Progress | Kubernetes workloads, AI pipelines, SDR data flow |
| Phase 4 | Research Workflows & Public Data Integration | ⏳ Upcoming | Real-time processing, public datasets, research dashboards |
See the ROADMAP.md for detailed information on each phase and implementation timeline.
This project is a transparent, living process where we document our successes and our mistakes. We follow real-world ITIL project management principles, but this is also a learning experience. We show our work warts and all for transparency. Mistakes and course corrections are part of the process, and that's intentional.
🔹 Security policies and best practices should not be blindly lifted from this repo. Every lab has unique needs, and configurations here are tailored to our environment. Always review and adapt security measures accordingly.
The Proxmox Astronomy Lab integrates AI/ML-enhanced signal processing, automation, and research workflows, but with a strong commitment to ethical AI practices. AI is a tool to enhance scientific discovery, not to replace rigorous analysis or responsible decision-making.
- Transparency - AI/ML models used for SDR processing, RAG knowledge retrieval, and automation are documented, explainable, and auditable.
- Data Integrity – Hydrogen Line radio astronomy data is processed with AI for enhancement, not manipulation. Scientific accuracy remains paramount.
- Privacy & Security – No user data, queries, or access logs are shared or monetized. All AI processing is local, not cloud-based.
- Open Science & Reproducibility – AI-powered signal enrichment and automation pipelines are open-source, so others can verify and improve them.
AI in scientific computing should be aiding research, not obscuring truth. The Proxmox Astronomy Lab adheres to ethical AI guidelines that prioritize transparency, accuracy, and reproducibility over automation for automation's sake.
To clone and explore the lab's documentation and infrastructure:
git clone https://github.com/yourusername/proxmox-astronomy-lab.git
cd proxmox-astronomy-labThe following documentation areas provide essential entry points to understanding the project:
- Infrastructure Overview - Lab hardware and architecture
- ITIL Simulation Approach - How enterprise IT practices are implemented
- Observatory Projects - Scientific research focus areas
- CISv8 Compliance Framework - Security implementation
Check out the complete Documentation Structure for a comprehensive guide to all resources.
This is an open-source research project. If you're interested in AI-powered radio astronomy, high-performance research infrastructure, or IT automation, feel free to contribute, test, or collaborate.
🛰 Follow the project on GitHub
📖 Read the full documentation in the /docs folder
This lab builds on the work of many open-source projects and communities, particularly those in radio astronomy, SDR processing, and scientific computing. Special thanks to:
- The GNU Radio community
- RTL-SDR and Airspy developers
- Proxmox and Kubernetes communities
- Wazuh, Prometheus, and Grafana projects
This project is licensed under the MIT License - see the LICENSE file for details.