TrulyRandom

A powerful .NET library for generating and testing high-quality truly random data using consumer hardware (generation is available only for Windows). Includes easy-to-use implementation of the NIST SP 800-22 test library. No external dependencies.

Nuget

The library allows developers to retrieve low-entropy data from microphones, cameras and user input (mouse and keyboard), process this low-entropy data into high-entropy using different randomness extractors, test it using a library of tests defined in NIST SP 800-22, store and retrieve it and finally use it to get random data of various types:

//Using camera as an entropy source
VideoSource source = new VideoSource();

//Creating extractors
DeflateExtractor deflateExtractor = new DeflateExtractor();
ShuffleExtractor shuffleExtractor = new ShuffleExtractor();
HashExtractor hashExtractor = new HashExtractor();

//Creating tester
Tester tester = new Tester();

//Creating buffer
Buffer buffer = new Buffer();

//Linking all the modules together
deflateExtractor.AddSource(source);
shuffleExtractor.AddSource(deflateExtractor);
hashExtractor.AddSource(shuffleExtractor);
tester.AddSource(hashExtractor);
buffer.AddSource(tester);

//Starting the chain
source.Start();
deflateExtractor.Start();
shuffleExtractor.Start();
hashExtractor.Start();
tester.Start();
buffer.Start();

//Waiting for data to collect
while (buffer.BytesInBuffer < 1000)
{
    Thread.Sleep(100);
}

//Using the data
int result = buffer.DataSource.GetInt();

Library can be simply used to test user data with test functions defined in NIST SP 800-22:

if (TrulyRandom.NistTests.Frequency(userData).PValues[0] > 0.01)
{
    System.Console.WriteLine("Success!");
}

And finally it provides a number of additional features:

1. Sample application to ease the startup;
2. Creation of the detailed test reports;
3. Long-term evaluation of the generator quality;
4. Automatic dumping of the exccess random data to the disk and retrieval from it when required;
5. Automatic test parameter selection in accordance to NIST SP 800-22;
6. Extractor seeding;
7. Dynamic compression adjustment depending on the output buffer state;
8. Module statistics: throughput in bytes per second, enthropy level, success rate, etc.;
9. Device autoselection for video and audio sources;
10. Still image detection;
11. Generation of different types of random data from the collected entropy;
12. Additional utilites like random array shuffle.

Please, if you find this library useful (or useless) or you have any other suggestions or comments - don't hesitate to contact me.