On the Period of the Linear Congruential and Power Generators Pär Kurlberg and Carl Pomerance
نویسندگان
چکیده
This sequence was first considered as a pseudorandom number generator by D. H. Lehmer. For the power generator we are given integers e, n > 1 and a seed u0 > 1, and we compute the sequence ui+1 = u e i (mod n) so that ui = u ei 0 (mod n). A popular case is e = 2, which is called the Blum–Blum–Shub (BBS) generator. Both of these generators are periodic sequences, and it is of interest to compute the periods. To be useful, a pseudorandom number generator should have a long period. In this paper we consider the problem of the period statistically as n varies, either over all integers, or over certain subsets of the integers that are used in practice, namely the set of primes and the set of “RSA moduli,” namely numbers which are the product of two primes of the same magnitude. If (e, n) = 1, then the sequence e (mod n) is purely periodic and its period is the least positive integer k with e ≡ 1 (mod n). We denote this order as ord(e, n). If (e, n) > 1, the sequence e (mod n) is still (ultimately) periodic, with the period given by ord(e, n), where n is the largest divisor of n that is coprime to e. (The aperiodic lead-in to such a sequence has length bounded by the binary logarithm of n.) In this paper we shall denote ord(e, n) by ord(e, n). The periods of both the linear congruential and power generators may be described in
منابع مشابه
On a Problem of Arnold: the Average Multiplicative Order of a given Integer Pär Kurlberg and Carl Pomerance
For g, n coprime integers, let `g(n) denote the multiplicative order of g modulo n. Motivated by a conjecture of Arnold, we study the average of `g(n) as n ≤ x ranges over integers coprime to g, and x tending to infinity. Assuming the Generalized Riemann Hypothesis, we show that this average is essentially as large as the average of the Carmichael lambda function. We also determine the asymptot...
متن کاملOn the Period of the Linear Congruential and Power Generators
This sequence was first considered as a pseudorandom number generator by D. H. Lehmer. For the power generator we are given integers e, n > 1 and a seed u = u0 > 1, and we compute the sequence ui+1 = u e i (mod n) so that ui = u ei (mod n). A popular case is e = 2, which is called the Blum–Blum–Shub (BBS) generator. Both of these generators are periodic sequences, and it is of interest to compu...
متن کاملLower Bounds on the Period of Some Pseudorandom Number Generators
We are interested in obtaining lower bounds on the periods of two standard pseudorandom number generators from number theory—the linear congruential generator, first introduced by D. H. Lehmer, and the so called power generator. For the former, given integers e, b, n (with e, n > 1) and a seed u = u0, we compute the sequence ui+1 = eui + b (mod n). For the power generator, given integers e, n >...
متن کاملOn the Performance of Birthday Spacings Tests with Certain Families of Random Number Generators
We examine how a statistical test based on discrete spacings between points, in 1 or more dimensions, detects the regularities in certain popular classes of random number generators. We provide a rule of thumb giving the minimal sample size for the test to reject the generator systematically, as a function of the generator's size (or period length), for generator families such as the linear con...
متن کاملParallel Linear Congruential Generators with Prime Moduli
Linear congruential generators (LCGs) remain the most popular method of pseudorandom number generation on digital computers. Ease of implementation has favored implementing LCGs with power-of-two moduli. However, prime modulus LCGs are superior in quality to power-of-two modulus LCGs, and the use of a Mersenne prime minimizes the computational cost of generation. When implemented for parallel c...
متن کامل