Running NIST STS 2.1.2

Topic: NIST STS 2.1.2

Date: [MM/DD/YYYY]

Source: [NIST SP800-22, Software Documentation, Personal Testing]

Key Points

NIST STS is used to assess the randomness of binary sequences.
Includes multiple tests defined in the NIST SP800-22 publication.
Requires a properly formatted bitstream as input.
Must be built using make and gcc on a Unix-like system.
Test results are stored in the experiments/ directory.

Examples & Details

✅ Prerequisites

Linux/macOS/WSL/Cygwin with:
- make
- gcc
Terminal and file access

📦 Download and Extract

                    unzip sts-2.1.2.zip
                    cd sts-2.1.2

🛠️ Build the Suite

                  make

🧪 Input File Format

Pure 0s and 1s, no headers, no whitespace.

Example:

                    01010011010101...

🏁 Run the Suite

              ./assess 1000000

Interactive options:

Input file
Select tests
Run tests
View results
Exit

Follow steps: 0 → 1 → 2 → 3 → 0

📁 Output

Stored in:

          experiments/AlgorithmTesting/
          ├── data.epsilon
          ├── results.txt
          ├── stats.txt

Each test result includes logs and summaries.

🧪 Common Test Numbers

Test #	Name
1	Frequency (Monobit)
2	Block Frequency
3	Runs
4	Longest Run of Ones
5	Rank
6	Discrete Fourier Transform
7	Non-overlapping Template

Questions & Clarifications

How are thresholds for p-values determined across different tests?
What is the statistical basis for deciding pass/fail?
Can custom test suites be integrated?

Action Items / Next Steps

Automate test run with subprocess/Python
Generate visual summaries of p-value distributions
Compare test results across multiple RNG implementations
Read SP800-22 full documentation

Summary

NIST STS 2.1.2 is a powerful suite for validating the statistical randomness of bitstreams. After compiling the suite, a clean binary file of known length can be tested via an interactive CLI. Results are saved for inspection. Automation is possible using scripting, and understanding individual test behavior is key for interpreting results.