But if you want to strictly prove p!=np, you should rely on some kind of random number generation algorithm. or maybe i haven't fully understood it.
He random number generator part has already been discussed earlier. The one-way function he presented is a good step towards a proof, it can possibly be used as a basis for a formal theorem of p!=np but I don't see any such theorem in the paper.
I can't wait to hear about the use cases for this encryption scheme.
As in "Theorem 3". we construct an encryption algorithm, as the plaintext-ciphertext is known, the problem of guessing the key can be transformed to another encryption algorithm, when only the ciphertext is known, the problem of guessing the key, which satisfies the property of one-way function intuitively.
"Theorem 4" theoretically proves that the above-mentioned problem of guessing the key satisfies the property of one-way function through the calculation of conditional probability. this also proves that the one-way function exists.
The existence of one-way functions can directly prove p!=np, which is a question of common sense.
The encryption algorithm in this paper can be used in all symmetric encryption scheme that require high security.