cp's OEIS Frontend

This is a front-end for the Online Encyclopedia of Integer Sequences, made by Christian Perfect. The idea is to provide OEIS entries in non-ancient HTML, and then to think about how they're presented visually. The source code is on GitHub.

A328701 Period in residues modulo n in iteration of x^2 + x + 1 starting at 0.

Original entry on oeis.org

1, 1, 2, 2, 1, 2, 3, 4, 2, 1, 2, 2, 3, 3, 2, 8, 1, 2, 2, 2, 6, 2, 4, 4, 4, 3, 2, 6, 7, 2, 7, 16, 2, 1, 3, 2, 3, 2, 6, 4, 7, 6, 5, 2, 2, 4, 5, 8, 3, 4, 2, 6, 1, 2, 2, 12, 2, 7, 11, 2, 4, 7, 6, 32, 3, 2, 2, 2, 4, 3, 10, 4, 18, 3, 4, 2, 6, 6, 3, 8, 2, 7, 2, 6, 1, 5, 14, 4, 1, 2, 3
Offset: 1

Views

Author

Jianing Song, Oct 26 2019

Keywords

Comments

a(n) is the period of {A002065 mod n}.
Let f(0) = 0, f(k+1) = (f(k)^2+f(k)+1) mod n, then a(n) is the smallest t such that f(i) = f(i+t) for all sufficiently large i.
Obviously a(n) <= A290731(n): f(1), f(2), ..., f(A290731(n)+1) are all of the form (s^2+s+1) mod n, so there must exists 1 <= i < j <= A290731(n)+1 such that f(i) = f(j), and a(n) <= j - i <= A290731(n). The equality seems to hold only for n = 3, 6 or n is a power of 2.

Examples

			In the following example, () denotes the cycles.
A002065(n) mod 4: 0, (1, 3), so a(4) = 2.
A002065(n) mod 7: 0, (1, 3, 6), so a(7) = 3.
A002065(n) mod 29: 0, (1, 3, 13, 9, 4, 21, 28), so a(29) = 7.
A002065(n) mod 61: (0, 1, 3, 13). {A002065(n) mod 61} enters into the cycle (0, 1, 3, 13) from the very beginning, so a(61) = 0.
A002065(n) mod 64: 0, (1, 3, 13, 55, 9, 27, 53, 47, 17, 51, 29, 39, 25, 11, 5, 31, 33, 35, 45, 23, 41, 59, 21, 15, 49, 19, 61, 7, 57, 43, 37, 63), so a(64) = 32.
		

Crossrefs

Cf. A002065, A328702 (indices to enter the cycles), A290731.

Programs

  • PARI
    a(n) = my(v=[0],k); for(i=2, n+1, k=(v[#v]^2+v[#v]+1)%n; v=concat(v, k); for(j=1, i-1, if(v[j]==k, return(i-j))))

Formula

a(n1*n2) = lcm(a(n1),a(n2)) if gcd(n1,n2) = 1.
It seems that for e > 0, a(3^e) = 2; a(5^e) = 1 if e = 1, 4*5^(e-2) otherwise; a(7^e) = 3; a(11^e) = 2 if e = 1, 10*11^(e-2) otherwise; a(13^e) = 3 if e = 1, 12*13^(e-2) otherwise ...
Proof that a(2^e) = 2^(e-1) by induction: we will show that {f(1), f(2), ..., f(2^(e-1))} is a reduced system modulo 2^e, where f is defined in the comment section. It is easy to see that this is true for e = 1, 2.
Suppose that {f(1), f(2), ..., f(2^(e-1))} is a reduced system modulo 2^e, e = 1, 2. For each 1 <= i <= 2^(e-1), f(2^(e-1)+i) - f(i) = Sum_{j=i..2^(e-1)+i-1} (f(j+1)-f(j)) = Sum_{j=i..2^(e-1)+i-1} (f(j)^2+1) = 2^(e-1) + Sum_{j=i..2^(e-1)+i-1} f(j)^2. Of course, {f(i), f(i+1), ..., f(2^(e-1)+i-1)} is also a reduced system modulo 2^e.
Note that if x == y (mod 2^e), then x^2 == y^2 (mod 2^(e+1)). So f(2^(e-1)+i) - f(i) == 2^(e-1) + (1^2+3^2+5^2+...+(2^e-1)^2) == 2^e (mod 2^(e+1)), 1 <= i <= 2^(e-1). This shows that {f(1), f(2), ..., f(2^(e-1)), f(2^(e-1)+1), f(2^(e-1)+2), ..., f(2^e)} is a reduced system modulo 2^(e+1). QED.