Enhanced Pseudorandom Number Generation using Mersenne-LFSR Fusion

Pseudorandom number generators (PRNGs) are fundamental tools in various applications, including cryptography, simulation and statistical sampling. Traditional PRNGs often face challenges related to periodicity, randomness quality, and computational efficiency. This paper presents the Mersenne LFSR fusion, an innovative approach that combines the Linear Feedback Shift Register (LFSR) and Mersenne Twister algorithms. By integrating the fast and simple structure of the LFSR with the high-quality randomness and long period of the Mersenne Twister, this method significantly improves PRNG performance. Experimental analysis shows that the Mersenne-LFSR fusion has superior randomness properties and longer period compared to traditional methods. The robustness of the fusion counter was validated through a scrambler-descrambler application, which effectively encoded and decoded data without introducing patterns. These enhancements make the proposed model a strong contender for applications that demand high-quality pseudorandom number sequences. Despite its superior performance, the Mersenne LFSR fusion has a limitation of higher power consumption, which may be a consideration in power-constrained environments. Nevertheless, it offers significant improvements in security and reliability, exceeding standard benchmarks for modern computing needs.