GeigerBIP39Generator

A python 3 script for generating BIP39 Bitcoin mnemonic codes with random data coming from a Geiger counter.

Index

1. Disclaimer
2. Requirements
3. How to use it

Disclaimer

I am not responsible for any coins lost due to the misuse of this script, nor due to it's use on a compromised computer. The script should be run on a computer that is permanently disconnected from the internet.

Requirements

• This script needs Python3. I'm running version 3.8.10, but other versions will likely work.

• Additionally, you will need this modules: scipy numpy sounddevice hashlib datetime os
  Some of them should come with your Python instalation.

  You can try to install them with pip:
  python3 -m pip install scipy
python3 -m pip install numpy
python3 -m pip install sounddevice

  This ones should come with your instalation:
  python3 -m pip install hashlib
python3 -m pip install datetime
python3 -m pip install os

• Finally, you will need a Geiger counter that can be connected to your computer as a microphone (i.e. that has an aux connection). I use a GQ GMC-300E Plus, but other GQ models or brands should work (although you might need to change the response time of your tube, more about this below).

How to use it

1. Clone this repository to your computer:
   git clone https://github.com/danieldemercado/GeigerBIP39Generator

   Navigate to the cloned repository:
   cd GeigerBIP39Generator

2. Connect your Geiger counter to the computer as a microphone, using a male to male aux cable. Make sure that the Geiger counter input is selected as your default microphone in your computer settings, and that the counts are being detected.

3. If you are using a different Geiger counter than mine, you will probably need to change the response time of your tube. Open GeigerBIP39Generator.py with your favourite editor and locate the funcion get_ticks:

   def get_ticks(wav_file):                    # This function reads the .wav file with the Geiger-Muller counter data and returns the times of the counts
   
       global max_volume
   
       a = read(wav_file)                      # Read the .wav file
   
       volume = np.array(np.abs(a[1]),dtype=float)         # Sound volume of the data
       time = np.arange(len(a[1]))/a[0]                    # Time for each data point
   
       if max_volume == None:                              # Gets the max volume in the data. Different computers might register different max volumes.
           max_volume = np.amax(volume)
   
       skip_time = 0.000270                                # The time (in seconds) that the Geiger tube needs to return to normal plus a little more (For GMC-300E plus ~ 270 μs) 
       skip_points = int(skip_time * a[0])                 
   
       ticks = []                                          # Empty list to save the tick's time
   
       n = 0                                               # While loop for processing the .wav
       while n < len(volume):
   
           if volume[n] > 0.25*max_volume:                               # If a tick is detected
               ticks.append(time[n])                              # Save its time
   
               n += skip_points                                   # Skip the data until de tube returns to normal
   
           n += 3                                          # Resolution of the data analysis. This must be a time smaller than the skip time (i.e.: n/a[0] << skip_time)
   
       return ticks

   Change skip_time = 0.000270 (in seconds) to the response time of your tube plus a little more (for the GQ GMC-300E Plus, the response time is ~240 μs, so 270 μs works fine). You must find the response time of your tube manually, by recording some data and analizing the time width of the counts.

   If the response time of your tube is much smaller (bigger) than 270 μs, you will probably also need to increase (reduce) the data analysis resolution by reducing (increasing) the integer number in n += 3.

4. Run the script with the command: python3 GeigerBIP39Generator.py

5. The script will ask you to choose a mnemonic code words length. Choose the one you like.

6. Wait untill the generation process finishes, this might take up to 20 minutes if you are only using background radiation. When this is done, the BIP39 words should be printed to the terminal.

You can check how it works in this video.