Randomness test of LightMAC-based Pseudorandom Number Generator (PRNG) with NIST SP 800-22 Revision 1A test

Susanti, Bety Hayat, Fadiel, Wahyu Achmad, Farida, Yeni and Briliyant, Obrina ORCID: https://orcid.org/0000-0002-1054-8112 2026. Randomness test of LightMAC-based Pseudorandom Number Generator (PRNG) with NIST SP 800-22 Revision 1A test. CAUCHY Jurnal Matematika Murni dan Aplikasi 11 (2) , pp. 2116-2137. 10.18860/cauchy.v11i2.43424

PDF - Published Version Available under License Creative Commons Attribution Share Alike. Download (4MB)

Abstract

Random numbers are essential for ensuring the security and unpredictability of cryptographic systems; however, resource-constrained devices still lack effective lightweight random number generators. This gap leaves encryption, authentication, and key generation vulnerable, exposing critical infrastructure to adversarial exploitation. This study presents the development and statistical evaluation of a Pseudo-Random Number Generator (PRNG) based on the LightMAC construction instantiated with the PRESENT lightweight block cipher. The proposed PRNG operates by iteratively processing a fixed seed message through LightMAC-PRESENT with a varying counter. The randomness of the generated bit sequences was rigorously analyzed using the NIST SP 800-22 Revision 1A test suite. The evaluation encompassed multiple implementation configurations, including two seed message lengths ($|x|$) of 64 bits and 128 bits, two key lengths for the underlying PRESENT cipher of 80 bits and 128 bits, and three counter lengths of 8 bits, 16 bits, and 32 bits. The results demonstrate that the LightMAC-PRESENT-based PRNG satisfies the NIST statistical requirements under specific parameter configurations. For a 64-bit seed message, a 128-bit PRESENT key combined with an 8-bit or 16-bit counter successfully passes all tests. For a 128-bit seed message, the generator achieves compliance with an 80-bit key and a 128-bit key. The optimal configuration identified is a 128-bit seed message, an 80-bit key, and a 16-bit counter, which passes all tests across all key-pattern variations. These findings provide critical insights into the optimal parameters for secure, efficient, lightweight deterministic random bit generation in resource-constrained environments.

Item Type:	Article
Date Type:	Publication
Status:	Published
Schools:	Schools > Computational & Mathematical Sciences Schools > Computer Science & Informatics
ISSN:	2086-0382
Date of First Compliant Deposit:	8 October 2026
Date of Acceptance:	14 September 2026
Last Modified:	08 Oct 2026 15:15
URI:	https://orca.cardiff.ac.uk/id/eprint/190107

Actions (repository staff only)

Edit Item