Close
0%
0%

Signal Inteligence Stack

Leveraging SDR and Machine Learning for automated RF situational awareness

Similar projects worth following
I wanted to listen to Fire/Police dispatch radio in my area but found that mainaining attention on it for any significant awareness was cumbersome. The intention of this project is to first leverage SDR to capture audio then use an NVIDIA Jetson Nano running Whisper to transcribe. Once transcribed, a main program will alert me to keywords. Alternatively, I would like to host a small LLM with RAG capabilities so I can just ask it for summaries or specific insight as it references the transcribed audio.

The SIGINT MiniLab is a compact, modular, and mostly autonomous signal intelligence system designed to monitor, decode, and summarize local RF activity. Housed in a custom-built 8U server rack with a combination of 3D printed and commercial components, the system leverages SDR technology, edge AI inference, and centralized data orchestration to provide real-time situational awareness.

Core modules include:

  • Signal Capture: Raspberry Pi 4 units paired with RTL-SDR dongles for identifying and capturing voice and data transmissions across a range of frequencies.

  • Intelligence Inference: An NVIDIA Jetson Orin Nano Developer Kit for speech-to-text transcription, tone and intention analysis, and contextual tagging of intercepted signals.

  • Central Orchestration: A Dell 7050 Micro PC acting as the central controller, user interface host, and future server for mission-critical apps like ATAK.

This project aims to create a fully deployable system capable of monitoring police/fire dispatch, public safety networks, weather broadcasts, and more — all without requiring constant user supervision. Future goals include AI-driven event summarization, real-time map overlays, and off-grid operational capabilities.

  • 1 × Jetson Orin Nano 8Gb Developer Kit https://www.nvidia.com/en-us/autonomous-machines/embedded-systems/jetson-orin/nano-super-developer-kit/
  • 1 × Dell 7050 Micro Computer https://www.dell.com/support/kbdoc/en-us/000124030/dell-optiplex-7050-system-guide
  • 1 × TP-Link SG-108E https://www.tp-link.com/us/home-networking/8-port-switch/tl-sg108e/#specifications
  • 4 × Gatorworks 8U Rails https://gatorco.com/product/rack-rails-8u-grw-rackrail-08u/
  • 1 × RTL-SDR V4 https://www.rtl-sdr.com/

View all 7 components

  • Physical Layout

    Brett Erickson05/27/2025 at 02:04 0 comments

    As of right now, I have most of the physical components I need. The only exception is the rack mounts for the Jetson, the router, and the antennas. I also need to redesign the RPi panel to also house the associated SDRs.

View project log

Enjoy this project?

Share

Discussions

Ryan Benasutti wrote 11/05/2025 at 03:20 point

Nice project, I built something very similar but I found whisper to have poor accuracy on noisy comms. Some fine tuning is probably required. Did you end up sticking with whisper?

  Are you sure? yes | no

Brett Erickson wrote 7 days ago point

Thank you! Project is still in the works. It's been on the back burner for a couple months and I've only done the hardware. I'm interested in your experience with whisper. I might have to look into ways of filtering the audio or something for preprocessing. Using ML/AI is new to me, but I'm excited to see how I can play around with it.

  Are you sure? yes | no

Similar Projects

Does this project spark your interest?

Become a member to follow this project and never miss any updates