# Knapsack

EN | ZH

## Backpack problem¶

First, let's introduce the backpack problem. Suppose a backpack can weigh W. Now there are n items with weights of $a_1, a_2,..., a_n$. We want to ask which items can fit the backpack. Filled up and each item can only be loaded once. This is actually solving such a problem.

$$x_1a_1 + x_2a_2 +, ..., + x_na_n = W$$

All of these $x_i$ can only be 0 and 1. Obviously we have to enumerate all the combinations of n items to solve this problem, and the complexity is $2^n$, which is the beauty of backpack encryption.

When encrypting, if we want to encrypt the plaintext as x, then we can represent it as an n-bit binary number and then multiply it by $a_i$ to get the encrypted result.

But what should I do when decrypting? We did make it difficult for others to decrypt the ciphertext, but we really have no way to decrypt the ciphertext.

But when $a_i$ is super-incremental, we have a solution. The so-called super-increment means that the sequence satisfies the following conditions.

$$a_i>\sum_{k=1}^{i-1}a_k$$

That is, the ith number is greater than the sum of all the previous numbers.

Why can you decrypt it if you meet such a condition? This is because if the encrypted result is greater than $a_n$, the preceding coefficient must be 1. On the contrary, the equation cannot be established anyway. Therefore, we can get the corresponding plaintext immediately.

However, this has another problem. Since $a_i$ is public, if the attacker intercepts the ciphertext, it is easy to crack such a password. In order to make up for this problem, an encryption algorithm such as Merkle–Hellman appears. We can use the initial backpack set as the private key, the transformed backpack set as the public key, and then slightly change the encryption process.

Although the super-increment sequence is mentioned here, it is not said how it is generated.

## Merkle–Hellman¶

### Public private key generation¶

#### Generating a private key¶

The private key is our initial backpack set. Here we use the super-increment sequence, how to generate it? We can assume that $a_1=1$, then $a_2$ is greater than 1, and similarly can generate subsequent values in turn.

#### Generating a public key¶

In the process of generating a public key, the operation of modular multiplication is mainly used.

First, we generate the modulus m of the modular multiplication, here we want to make sure

$$m>\sum_{i=1}^{i=n}a_i$$

Second, we choose the multiplier w of the modular multiplication as the private key and ensure

$$gcd(w,m)=1$$

After that, we can generate the public key by the following formula.

$$b_i \equiv w a_i \bmod m$$

And this new backpack set $b_i$ and m as the public key.

### encryption and decryption¶

#### Encryption¶

Suppose we want to encrypt the plaintext as v, each bit is $v_i$, then the result of our encryption is

$$Sum_ {i = 1} ^ {n} i = b_iv_i m way$$

#### Decryption¶

For the decryption side, we can first ask for the inverse of $m^{-1}$ for m.

Then we can multiply the obtained ciphertext by $w^{-1}$ to get the plaintext, because

$$Sum_ {i = 1} ^ {w} i = n ^ {- 1} b_iv_i way m = sum_ {i = 1} ^ {n} i = a_iv_i m way$$

here has

$$b_i \equiv w a_i \bmod m$$

The encrypted message for each block is less than m, so the result is naturally plaintext.

### 破¶

The system was deciphered two years after the proposed encryption system. The basic idea of deciphering is that we do not necessarily need to find the correct multiplier w (ie trapdoor information), just find the arbitrary modulus m&#39; and The multiplier w&#39; can be used to generate a super-incrementing backpack vector by using w&#39; to multiply the public backpack vector B.

### Examples¶

Here we take Archaic in 2014 ASIS Cyber Security Contest Quals as an example, [topic link] (https://github.com/ctfs/write-ups-2014/tree/b02bcbb2737907dd0aa39c5d4df1d1e270958f54/asis-ctf-quals-2014/archaic ).

First look at the source program

secret = 'CENSORED'

msg_bit = bin(int(secret.encode('hex'), 16))[2:]


First we get all the bits of secret.

Second, use the following function to get the keypair, including the public and private keys.

keyPair = makeKey (curtain (msg_bit))


Carefully analyze the makekey function as follows

def makeKey(n):

privKey = [random.randint (1, 4 ** n)]
s = privKey 
for i in range(1, n):

privKey.append(random.randint(s + 1, 4**(n + i)))

s + = privKey [i]
q = random.randint (privKey [n-1] + 1, 2 * privKey [n-1])
r = random.randint(1, q)

while gmpy2.gcd(r, q) != 1:

r = random.randint(1, q)

pubKey = [ r*w % q for w in privKey ]

return privKey, q, r, pubKey


It can be seen that prikey is a super-incremental sequence, and the obtained q is larger than the sum of all the numbers in prikey. In addition, we get r, which is exactly the same as q, which indicates that the encryption is a backpack encryption.

Sure enough, the encryption function is to multiply each bit of the message by the corresponding public key and sum.

def encrypt(msg, pubKey):

msg_bit = msg

n = only (pubKey)
cipher = 0

i = 0

for bit in msg_bit:

cipher += int(bit)*pubKey[i]

i += 1

return bin(cipher)[2:]


For the cracked script we use the script on [GitHub] (https://github.com/ctfs/write-ups-2014/tree/b02bcbb2737907dd0aa39c5d4df1d1e270958f54/asis-ctf-quals-2014/archaic). Make some simple modifications.

import binascii
# open the public key and strip the spaces so we have a decent array

fileKey = open (&quot;pub.Key&quot;, &#39;rb&#39;)
pubKey = fileKey.read().replace(' ', '').replace('L', '').strip('[]').split(',')

nbit = only (pubKey)
# open the encoded message

fileEnc = open (&quot;enc.txt&quot;, &#39;rb&#39;)

print "start"

# create a large matrix of 0's (dimensions are public key length +1)

A = Matrix(ZZ, nbit + 1, nbit + 1)

# fill in the identity matrix

for i in xrange(nbit):

A[i, i] = 1

# replace the bottom row with your public key

for i in xrange(nbit):

A[i, nbit] = pubKey[i]

# last element is the encoded message

A[nbit, nbit] = -int(encoded)

res = A.LLL()

for i in range(0, nbit + 1):

# print solution

M = res.row(i).list()

flag = True

for m in M:

if m != 0 and m != 1:

flag = False

break

if flag:

print i, M

M = ''.join(str(j) for j in M)

# remove the last bit

M = M[:-1]

M = hex(int(M, 2))[2:-1]

print M


Decoded after output

295 [1, 0, 0, 0, 0, 0, 1, 0, 1, 0, 1, 0, 0, 1, 1, 0, 1, 0, 0, 1, 0, 0, 1, 0, 1, 0, 1, 0, 0, 1, 1, 0, 1, 0, 1, 1, 1, 1, 1, 0, 0, 1, 1, 1, 0, 0, 1, 0, 1, 1, 0, 0, 0, 1, 0, 0, 1, 1, 0, 0, 1, 0, 0, 0, 0, 1, 1, 0, 0, 1, 1, 0, 1, 1, 0, 0, 1, 0, 0, 0, 0, 1, 1, 0, 1, 0, 1, 0, 1, 1, 0, 0, 1, 1, 0, 0, 1, 1, 0, 0, 1, 0, 0, 0, 0, 1, 1, 0, 0, 1, 0, 0, 0, 1, 1, 0, 1, 0, 0, 0, 0, 1, 1, 0, 0, 1, 0, 0, 0, 1, 1, 0, 0, 1, 0, 0, 0, 1, 1, 0, 1, 1, 0, 0, 0, 1, 1, 1, 0, 0, 0, 0, 0, 1, 1, 0, 0, 1, 0, 0, 1, 1, 0, 0, 0, 1, 1, 0, 0, 1, 1, 0, 0, 0, 1, 0, 0, 1, 1, 1, 0, 0, 1, 0, 0, 1, 1, 0, 1, 0, 1, 0, 0, 1, 1, 0, 1, 1, 0, 0, 0, 1, 1, 1, 0, 0, 0, 0, 0, 1, 1, 1, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1, 0, 1, 1, 0, 0, 1, 1, 0, 0, 0, 0, 1, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1, 0, 1, 1, 1, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1, 0, 1, 1, 0, 0, 0, 1, 1, 0, 0, 0, 1, 0, 1, 1, 0, 0, 0, 1, 1, 0, 1, 1, 0, 0, 1, 0, 1, 0]

415349535f3962643364356664323432323638326331393536383830366130373036316365
&gt;&gt;&gt; import binascii
>>> binascii.unhexlify('415349535f3962643364356664323432323638326331393536383830366130373036316365')

&#39;ASIS_9bd3d5fd2422682c19568806a07061ce&#39;


It should be noted that the matrix of res obtained by the LLL attack we only contains the 01 value is the result we want, because when we encrypt the plaintext, it will be decomposed into binary bit strings. In addition, we need to remove the last number of the corresponding row.

flag 是 ASIS_9bd3d5fd2422682c19568806a07061ce

### Title¶

• 2017 national classic