Vincent
Guest
Posts: n/a

 02-19-2004
Dear experts,

I am a college student and I was asked a question on
Cipher (network security). I am frustrated as I could not

Scenario:
Consider the subsitution cipher for English text which
consists of A,B,...,Y,Z only (a total of 26 letters). The
encryption rule is to substitute a letter by another
letter which is different from itself. For instance,
subsitute A by W (but not A), B by H (but not B),...etc.
The actual subsitution rule is governed by a key. Once
the key is chosen, the subsitution rule is fixed and can
represented as follows:

A-W; B-H; C-J; D-K; E-Y; .... ;X-B; Y-U; Z-L

Questions:
1) Based on the above cipher system, determin the total
number of different keys
2) If an attacker uses a brute force attack to decrypt a
particular message, and he try 1,000,000 keys in a
second. WHat is the average time that he can decrypt the
message?
3) Is it possible to decrypt the ciphertext "WXEUV" by
this brute force attack? Why?
4) Is it possible to decrypt a ciphertext which consists
of 100,000 letters? WHy?

Please give me some ideas if possible.
Thanks a lot for your help.

Jay Walters
Guest
Posts: n/a

 02-19-2004
This feels like a game of boggle

How to crack: write a recursive function that iterates
through the possibly. For a person to physically look at
each variant would take forever. However writing another
function to perform substring checks against a dictionary
to tag possible successes would cut this down.

It's 8am ... I'm tired. So I'm shooting in the dark here.

A real example would be better, do you have spaces in
your text, or are all the words run together?

number of possibilities / number of keys tried per second
= total number of seconds

fixed key size of 26 ... 26 different letters something
like : (26*25*24* ... *3*2*1) = total number of
possibilities minus the math for a!=a, b!=b

Interms of question 4 and 5 ... I smell a trick here.

Decrypting a single word would be very hard because as
you run through all of the cipher key possibilities,
you'd basically create every 5 letter word in the
dictionary. So yes, you could decrypt it, you just
wouldn't know which word is right

The more words to decrypt, the greater chance you'll get
a match.

Bottom line: It would still take some time, but it's very
achievable.

>-----Original Message-----
>Dear experts,
>
>I am a college student and I was asked a question on
>Cipher (network security). I am frustrated as I could

not
>
>Scenario:
>Consider the subsitution cipher for English text which
>consists of A,B,...,Y,Z only (a total of 26 letters).

The
>encryption rule is to substitute a letter by another
>letter which is different from itself. For instance,
>subsitute A by W (but not A), B by H (but not B),...etc.
>The actual subsitution rule is governed by a key. Once
>the key is chosen, the subsitution rule is fixed and can
>represented as follows:
>
>A-W; B-H; C-J; D-K; E-Y; .... ;X-B; Y-U; Z-L
>
>Questions:
>1) Based on the above cipher system, determin the total
>number of different keys
>2) If an attacker uses a brute force attack to decrypt a
>particular message, and he try 1,000,000 keys in a
>second. WHat is the average time that he can decrypt the
>message?
>3) Is it possible to decrypt the ciphertext "WXEUV" by
>this brute force attack? Why?
>4) Is it possible to decrypt a ciphertext which consists
>of 100,000 letters? WHy?
>
>Please give me some ideas if possible.
>Thanks a lot for your help.
>.
>

Vincent
Guest
Posts: n/a

 02-20-2004
Hello Jay,

lot. Thanks again!

Vincent

>-----Original Message-----
>This feels like a game of boggle
>
>How to crack: write a recursive function that iterates
>through the possibly. For a person to physically look at
>each variant would take forever. However writing another
>function to perform substring checks against a

dictionary
>to tag possible successes would cut this down.
>
>It's 8am ... I'm tired. So I'm shooting in the dark here.
>
>A real example would be better, do you have spaces in
>your text, or are all the words run together?
>
>number of possibilities / number of keys tried per

second
>= total number of seconds
>
>fixed key size of 26 ... 26 different letters something
>like : (26*25*24* ... *3*2*1) = total number of
>possibilities minus the math for a!=a, b!=b
>
>Interms of question 4 and 5 ... I smell a trick here.
>
>Decrypting a single word would be very hard because as
>you run through all of the cipher key possibilities,
>you'd basically create every 5 letter word in the
>dictionary. So yes, you could decrypt it, you just
>wouldn't know which word is right
>
>The more words to decrypt, the greater chance you'll get
>a match.
>
>Bottom line: It would still take some time, but it's

very
>achievable.
>
>
>
>
>
>
>
>
>
>
>
>>-----Original Message-----
>>Dear experts,
>>
>>I am a college student and I was asked a question on
>>Cipher (network security). I am frustrated as I could

>not
>>
>>Scenario:
>>Consider the subsitution cipher for English text which
>>consists of A,B,...,Y,Z only (a total of 26 letters).

>The
>>encryption rule is to substitute a letter by another
>>letter which is different from itself. For instance,
>>subsitute A by W (but not A), B by H (but not

B),...etc.
>>The actual subsitution rule is governed by a key. Once
>>the key is chosen, the subsitution rule is fixed and

can
>>represented as follows:
>>
>>A-W; B-H; C-J; D-K; E-Y; .... ;X-B; Y-U; Z-L
>>
>>Questions:
>>1) Based on the above cipher system, determin the total
>>number of different keys
>>2) If an attacker uses a brute force attack to decrypt

a
>>particular message, and he try 1,000,000 keys in a
>>second. WHat is the average time that he can decrypt

the
>>message?
>>3) Is it possible to decrypt the ciphertext "WXEUV" by
>>this brute force attack? Why?
>>4) Is it possible to decrypt a ciphertext which

consists
>>of 100,000 letters? WHy?
>>
>>Please give me some ideas if possible.
>>Thanks a lot for your help.
>>.
>>

>.
>