looking for some solutions? You are welcome.

SOLVED: Given the output C of the function H(A, B) = (A ^ B) ^ ((A & B) << 1), how to write a function that outputs A given C and B?

lyrically wicked:

Assuming that a and b are unsigned integers, consider the following function:

function H(a, b) {
    var c = (a ^ b) ^ ((a & b) << 1);
    return c
};

The description of this function can be found in NORX specification (the links are available at this page).

I need to implement the reverse function of H (in Javascript). That is, given c and b, the function is required to output a.

The specification provides an algorithm for the reverse function (assuming that v[i] denotes an i-th bit of an integer v):

a[0] = c[0] XOR b[0];
a[1] = (c[1] XOR b[1]) XOR (a[0] AND b[0]);
...
a[i] = (c[i] XOR b[i]) XOR (a[i-1] AND b[i-1]);

But I don't know how to implement it.

I tried the following function (note that the arguments are the elements of UintNArrays, so we assume that N is either 8 or 16 or 32):

function revH(c, b) {
    var x = (c ^ b) >>> 0; var a = x;

    for (var i = 1; i < N; ++i) {
        x0 = (c ^ b) ^ ( (a >>> i) & (b >>> i )); 
        a = (a & ~(0x1 << i)) | ((x >>> (N-i)) << i);
    };
    return a
};

But this function is not correct (it does not output correct results). How to implement the reverse function of H?



Posted in S.E.F
via StackOverflow & StackExchange Atomic Web Robots
Share:

No comments:

Recent