Determine the influence of IPv6 target generation algorithms on address selection and randomness-test outcomes

Determine how IPv6 target generation algorithms (TGAs) influence scanners’ address selection within prefixes—specifically the selection of subnet bits and Interface Identifier bits—and ascertain how this influence affects the pass/fail outcomes of the applied NIST randomness tests on those address components.

Background

The study assesses randomness in targeted addresses using selected NIST tests (frequency, runs, spectral, and cumulative sums) applied separately to the subnet and Interface Identifier (IID) components for sessions with at least 100 packets. Results suggest scanners favor structured subnet selection while choosing IIDs more randomly.

The authors note that different target generation algorithms (TGAs) could shape observed address-selection behavior and thereby influence NIST test outcomes, but the specific impact of TGAs is not established within the present analysis.

References

We do not know how TGAs influence the selection of addresses and how they influence the outcome of this analysis. This will be part of our future work.

— A Detailed Measurement View on IPv6 Scanners and Their Adaption to BGP Signals  (2506.20383 - Egloff et al., 25 Jun 2025) in Appendix (Testing for randomness: NIST Test Suite)