牛客多校第九场 A (fwt-异或卷积)

Circulant Matrix

题目描述

Niuniu has recently learned how to use Gaussian elimination to solve systems of linear equations.
Given n and $a[i]$, where $n$ is a power of $2$, let’s consider an $n\times n$ matrix $A$.

The index of $A[i][j]$ and $a[i]$ are numbered from $0$.
The element $A[i][j]$ satisfies $A[i][j] = a[i\, xor\, j]$,
XOR

Let $p = 1000000007$.
Consider the equation
$A x = b \pmod p$
where $A$ is an $n \times n$ matrix, and $x$ and $b$ are both $n \times 1$ row vector.

Given $n, a[i], b[i]$, you need to solve the $x$.
For example, when $n = 4$, the equations look like
$A[0][0]\cdot x[0] + A[0][1]\cdot x[1] + A[0][2]\cdot x[2] + A[0][3]\cdot x[3] = b[0] \pmod p\\ A[1][0]\cdot x[0] + A[1][1]\cdot x[1] + A[1][2]\cdot x[2] + A[1][3]\cdot x[3] = b[1] \pmod p\\ A[2][0]\cdot x[0] + A[2][1]\cdot x[1] + A[2][2]\cdot x[2] + A[2][3]\cdot x[3] = b[2] \pmod p\\ A[3][0]\cdot x[0] + A[3][1]\cdot x[1] + A[3][2]\cdot x[2] + A[3][3]\cdot x[3] = b[3] \pmod p$
and the matrix $A$ can be decided by the array $a$.

It is guaranteed that there is a unique solution $x$ for these equations.

Input

The first line contains an integer, which is $n$.
The second line contains $n$ integers, which are the array $a$.
The third line contains $n$ integers, which are the array $b$.

$1 <= n <= 262144$
$p = 1000000007$
$0 <= a[i] < p$
$0 <= b[i] < p$

Output

The output should contains $n$ lines.
The $i$-th(index from $0$) line should contain $x[i]$.
$x[i]$ is an integer, and should satisfy $0 <= x[i] < p$.

问题分析

首先，我们通过观察题目得到

$$A\times x = b \pmod p$$
将矩阵 $A$展开
$$\left[ \begin{array}{ccc} a_0&a_1&a_2&a_3\\\\ a_1&a_0&a_3&a_2\\\\ a_2&a_3&a_0&a_1\\\\ a_3&a_2&a_1&a_0 \end{array} \right]$$
我们可以发现这个矩阵左上角和右下角相同，左下角和右上角相同，矩阵再扩大亦是如此。
并且可以看出矩阵的全部信息只和第一行有关，恰好第一行就是数组 $a$，所以实际上这是一个$fwt$(异或卷积)。
所以我们可以将问题转换成：给出向量 $a$和$b$， $fwt(a)\cdot fwt(x) = fwt(b)$，求原向量 $x$。

那么我们就可以做一次$fwt(a)$，做一次 $fwt(b)$，然后 $x[i] = \frac{b[i]}{a[i]}$，再做一次逆 $fwt(x)$就好了啊。

代码

#include <cstdio>
typedef long long LL;
const int N = 262150, p = 1e9+7;
LL a[N], b[N], x[N], n, inv2;
void fwt(LL *a) {
    for (int i = 1; 1 << i <= n; i++) {
        for (int j = 0; j < n; j += 1 << i) {
            for (int k = 0; k < 1 << (i - 1); k++) {
                LL x = a[j + k];
                LL y = a[j + k + (1 << (i - 1))];
                a[j + k] = (x + y) % p;
                a[j + k + (1 << (i - 1))] = (x + p - y) % p;
            }
        }
    }
}

void fwt(LL *a,int t) {
    for (int i = 1; i < n; i <<= 1) {
        for (int j = 0; j < n; j += (i << 1))
            for (int k = j; k < i + j; k++) {
                LL x = a[k], y = a[k + i];
                a[k] = (x + y) % p, a[k + i] = (x - y + p) % p;
                if (t)a[k] = a[k] * inv2 % p, a[k + i] = a[k + i] * inv2 % p;
            }
    }

}

LL Pow(LL a,LL n) {
    LL ans = 1;
    while (n) {
        if (n & 1) ans = a * ans % p;
        a = a * a % p;
        n >>= 1;
    }
    return ans % p;
}

int main() {
    scanf("%lld", &n);
    inv2 = Pow(2, p - 2);
    for (int i = 0; i < n; ++i)
        scanf("%lld", a + i);
    for (int i = 0; i < n; ++i)
        scanf("%lld", b + i);
    fwt(a);
    fwt(b);
    for (int i = 0; i < n; ++i) {
        x[i] = b[i] * Pow(a[i], p - 2) % p;
    }
    fwt(x, 1);//逆fwt
    for (int i = 0; i < n; ++i)
        printf("%lld\n", x[i]);
//    实际上我们如果要算逆fwt也可以这样求
//    for (int i = 0; i < n; ++i)
//        printf("%lld\n", x[i] * Pow(n, p - 2) % p);
    return 0;
}