Design and Synthesis of LFSR based Random Number Generator

Digital random number generators play a vital role in cryptography applications which is commonly implemented using Linear Feedback Shift Registers (LFSRs). One of the major disadvantages of the LFSR based Random Number Generator (RNG) is that they are easily predictable since the sequences produced are periodic. Therefore, an RNG model is proposed in this paper to increase the unpredictability of the LFSR based RNG by including a polynomial modulator which consists of a multiplexer, a counter and a comparator to select different primitive polynomials. The randomness of the proposed RNG models with varying bit lengths are verified using NIST 800-22 Test suite and their power consumptions are obtained. From the NIS T results, it is evident that the sequences generated by the proposed RNG are random and from the results, it can be inferred that with an increase in the number of bits the power consumption increases.