Maths Olympiad Prep

Track / Stage 8 / 155 of 180 #1855 of 1964

Problem 1855

IMO Shortlist mid-range; USAMO P2/P5
Number theory Difficulty 8.7 Prove it Team Selection Test · Turkey

For integers m3m \ge 3, nn and x1,x2,,xmx_1, x_2, \dots, x_m if xi+1xixixi1(modn)x_{i+1} - x_i \equiv x_i - x_{i-1} \pmod n for every 2im12 \le i \le m-1, we say that the mm-tuple (x1,,xm)(x_1, \dots, x_m) is an *arithmetic sequence in (mod n)*. Let p5p \ge 5 be a prime number and 1<a<p11 < a < p-1 be an integer. Let {a1,a2,,ak}\{a_1, a_2, \dots, a_k\} be the set of all possible remainders when positive powers of aa are divided by pp. Show that if a permutation of {a1,a2,,ak}\{a_1, a_2, \dots, a_k\} is an arithmetic sequence in (mod pp), then k=p1k = p-1.

This one wants a proof. Work it on paper, then read the official solution and mark yourself. Be honest about it: the record is only any use to you if it is.

Official solution

Let (b1,b2,,bk)(b_1, b_2, \dots, b_k) be a permutation of {a1,a2,,ak}\{a_1, a_2, \dots, a_k\} which is an arithmetic sequence in (mod pp). Then, for some integers cc and dd we have bic+id(modp)b_i \equiv c + i d \pmod p for every i=1,2,,ki = 1, 2, \dots, k.

It is easy to see that kk is the order of aa in (mod pp). Then, we have ak1(modp)a^k \equiv 1 \pmod p and
b1+b2++bka+a2++ak=ak1a10(modp). b_1 + b_2 + \dots + b_k \equiv a + a^2 + \dots + a^k = \frac{a^k - 1}{a-1} \equiv 0 \pmod{p}.
On the other hand, we have
b1+b2++bkk(2c+d(k+1))2(modp) b_1 + b_2 + \dots + b_k \equiv \frac{k(2c + d(k+1))}{2} \pmod{p}
and hence, we get
2c+d(k+1)0(modp). 2c + d(k + 1) \equiv 0 \pmod{p}.
Similarly, we have
b12+b22++bk2a2+a4++a2k=a2a2k1a210(modp). b_1^2 + b_2^2 + \dots + b_k^2 \equiv a^2 + a^4 + \dots + a^{2k} = a^2 \frac{a^{2k} - 1}{a^2 - 1} \equiv 0 \pmod{p}.
since 1<a<p11 < a < p-1. Moreover,
i=1kbi2=i=1k(c+id)2=kc2+cdk(k+1)+d2k(k+1)(2k+1)6 \sum_{i=1}^{k} b_i^2 = \sum_{i=1}^{k} (c + i d)^2 = k c^2 + c d k(k+1) + d^2 \frac{k(k+1)(2k+1)}{6}
Therefore, we get
kc2+cdk(k+1)+d2k(k+1)(2k+1)60(modp)(2) k c^2 + c d k(k+1) + d^2 \frac{k(k+1)(2k+1)}{6} \equiv 0 \pmod{p} \quad (2)
By (1), we can replace c=d(k+1)/2c = -d(k+1)/2 in (2) and get
d2k(k+1)24d2k(k+1)22+d2k(k+1)(2k+1)60(modp) \frac{d^2 k(k+1)^2}{4} - \frac{d^2 k(k+1)^2}{2} + \frac{d^2 k(k+1)(2k+1)}{6} \equiv 0 \pmod{p}
and hence,
d2k(k+1)2(k+12+2k+13)=d2k(k+1)(k1)120(modp). \frac{d^2 k(k+1)}{2} \left( -\frac{k+1}{2} + \frac{2k+1}{3} \right) = \frac{d^2 k(k+1)(k-1)}{12} \equiv 0 \pmod{p}.
Clearly, d0d \neq 0 (mod pp) and 2<kp12 < k \le p-1. Therefore, we see that k=p1k = p-1.

Source: MathNet, licensed CC-BY-4.0. Statement and solution reproduced as published; topic, difficulty and ordering added by this site.