Ethical Hacking

Prof. Indranil Sengupta

Lecture – 28
Private - Key Cryptography (Part II)

Continuing with our discussion on Private-Key Cryptography, in this lecture we shall be

(Refer Slide Time: 00:35)

Now, as I said now here we shall be revisiting some of the practical algorithms, which
are used in real scenarios. The algorithms that we will be talking about our Data
Encryption Standard or DES, triple DES which is an extension of that and the most
recently one, recent one which is most widely used nowadays is the Advanced
Encryption Standard or AES, ok.
(Refer Slide Time: 01:01)

Talking about the practical algorithms, there are numerous algorithms in fact. If you look
at the literature, in the literature there is lot of research going on in the development of
new algorithms. There are good features in these algorithms, there are weaknesses, there
are drawbacks, but the cryptographers or the persons who are developing these
algorithms they are continuously trying to come up with better and better algorithms. So,
data encryption standard was one of the algorithms, which was proposed earlier in the
80s, where the block size was 64 bits.

So, it encrypted 64 bits of data at a time and the key size was 2, was 56 bits. This DES
was used for quite a significant amount of time, but because the key size is not that large,
2 to the power 56 in the present day context is not a very large number. Using the fastest
computer with brute force technique, you can generate all these keys and you can mount
a brute force attack. So, there are other algorithms which have been proposed, this idea is
one of the algorithms which have been explored here also the block size is 64 bit, but the
key size has been enhanced to 128 bits.

But as I said the most widely used symmetric key or private key algorithm today is
called advanced encryption standard or AES. This is also referred to as Rijndael
cryptosystem taking a cue from the names of the inventors. Here the block size is 128,
but for the key size we have a choice. You can have 128, 192 or 256 bits depending on
the level of security that you want in a particular application; larger the key size, higher
will be the security.

(Refer Slide Time: 03:19)

So, let us briefly look at the data encryption standard first. So, as it said at one point in
time this was most widely used. This is also sometimes referred to as Data Encryption
Algorithm or DEA. As I said it is a block cipher which handles block size of 64 bits and
the key size is 56 bits. So, if you have a longer plaintext, then it will be split into blocks
of 64, but in general the last block can be less than 64. So, there will be some additional
bits added to it, called pad to make this also 64 and accordingly using this algorithm the
ciphertext will be generated, right. This is how it works.
(Refer Slide Time: 04:23)

Now, I am not going to the detail explanation of how these are designed. So, I am
showing you the overall schematic how this DES is internally and what are the kind of
steps that are being carried out to encrypt a given plaintext. So, as I said, this is a block
cipher with 64 bit of plain text so that 64 bit number is coming here at the top. And, on
the other side I have the 56 bit key that is coming here. Now, you see in DES there are,
there is a concept of a round. There are 16 rounds. In 16 rounds you are carrying out
similar kind of calculations, but 16 times, one after the other.

So, what is the kind of calculation? First these 64 bits that is coming, you do an initial
transposition, jumble up the order of the bits. This is called initial permutation. Then this
initially permuted value will go through the 16 rounds. Then you do a 32 bit swap,
because you have 64 bit number. You take two 32 bit chunks and interchange their order.
The last you bring it to first. The first to bring it to last and then again whatever
permutation you used here, you use a reverse input permutation here, the reverse
permutation here and whatever comes out that will be your 64 bit ciphertext.

But another thing you see each of these rounds is dependent on another factor which is
coming from the right side. These are the so called round keys. The value of the round
keys are different, not the same. Although the value of K is same, PC is a permuted.
Choice some kind of key permutation is going on and first permutation then LCS is left
circular, circular shift, this key is undergoing a circular shift left and using some
combination function you are generating the round keys.

So, round keys are changing with the round number. So, you see it is a quite complicated
process through which I am generating the ciphertext from the plaintext. Now each of
these rounds I am not going into the exact functionality, but broadly speaking what
happens in these rounds, are like this.

(Refer Slide Time: 07:01)

You see, the 64 bit plaintext that is coming and is going through each of the rounds, if
you divide it to be into 2 parts, you call it a left part and a right part. So, at the ith
iteration, the left part is copied with ith part. So, in the after around let us call it and
. So, when you generate the output of this round, this will be your and this will be
your . So, whatever this was here, this will get copied to , = , but in this you
have a function.

This left part bit by bit exclusive-OR, some non-linear function. This is a complex
function of not only , but also the round key that is coming from the right side. So,
this is computed in a complicated way and any crypto, cryptosystem which have this
kind of a, you can say structure there are many cryptosystems based on this structure,
this is called in general Fiestel structure, ok. Now, as I said, DES as such is good, but the
only concern is the short key length of 56 bits. For critical applications we need keys of
larger sizes.
(Refer Slide Time: 08:47)

So, the first attempt which was done or you may say explored is to use multiple runs of
the DES algorithm to provide greater security. So, here you have an algorithm which is
referred to as triple DES. As the name implies triple means, you are using DES three
times for every encryption. So, what do you do? There will be three different keys, 56 bit
keys for DES and there will be three executions of the DES algorithm for encrypting
every 64 bit plaintext. For every 64 bit plaintext, to encrypt it into a 64 bit ciphertext you
will be requiring three executions or three runs of the DES algorithm. The drawback
obviously, is your method will become three times slower.

But the advantage is that you are using three 56 bit keys, 56 multiplied by 3 that is
essentially your strength of the key, you can say the way it works is that from the
plaintext you apply the first key carry out the DES encryption process. Well decryption I
have not shown. Decryption process is also quite similar, but the rounds are executed in
the reverse order. So, whatever comes, you run the decryption algorithm with key K2
and whatever comes you again run the encryption algorithm with key K3.

The point to note on, note is that any data if we apply the encryption algorithm followed
by the decryption algorithm you will get back the same thing. Similarly, if you apply the
decryption first followed by encryption then also you will be getting back the same thing.
So, this principle has been used in structuring this order of execution and as I said, 56
multiplied by three is 168. So, effective key size is 168.
(Refer Slide Time: 11:07)

Here pictorially it works like this. For encryption the plaintext is coming first. You do an
encryption with the first key , then decryption with , then encryption with . You
get the final ciphertext. For decryption you do it in the reverse order. In the last step you
did an encryption with , you will do a decryption with . In the middle you did a
decryption with , you do an encryption with . Here you did an encryption with ,
you do a decryption with .

So, in sequence in the reverse sequence, encryption, decryption will cancel out, cancel
out, cancel out and finally, you will be getting back the plaintext. This is how triple DES
works. Triple DES has been also quite widely used. In fact, in many of the money ATM
machines, triple DES was used till some time back. So, this was quite widely used.
(Refer Slide Time: 12:17)

Now, there was a need for a new standard which was understood and identified, because
DES was used for a pretty long time. There was no apparent weakness in the algorithm.
The only concern was the key size, was smaller and many theoretical attacks designed by
mathematicians were published which reduces the complexity of breaking DES. So, DES
was no longer considered to be very secure. Well of course, triple DES is an option, but
three executions of algorithm for every single encryption, is too much of an overhead,
ok. So, that also is some kind of a drawback.

Now, the standard organization NIST in US who actually standardizes the cryptographic
algorithms, they came up with a call for ciphers in 1997. They ask the mathematicians
and scientists to publish their algorithms. There will be like a competition. They will be
selecting the best of them as the next generation standard for encryption. So, in that way
there were 15 candidates which are finally accepted in 1998, were shortlisted and finally,
the Rijndael cryptosystem was selected as the advanced encryption standard, in short
AES.

Sometimes we refer to this algorithm as just the AES algorithm, ok. Today this is one of
the most widely used symmetric key algorithms that has been employed or deployed in
many applications.
(Refer Slide Time: 14:11)

So, some features of this AES cryptosystem. Now, this I have already mentioned earlier,
block length and the key length. Well in the, in the original Rijndael proposal which was
submitted, both the block length and the key length can be independently chosen to be
either 128, 192 or 256. But when NIST accepted the proposal as a standard, they added
some additional constraint. They said that well let the block length be only 128, let us not
have flexibility in the block length, but let the key can be either 128, 192 or 256.

So, although the original Rijndael standard can have variable block sizes, but AES will
have only 128 bit block sizes. The advantage of this method is that they are very easy to
implement both in hardware and also in software and there has been lot of attacks which
have been attempted theoretical and hardware based attacks, but still today this algorithm
is known to be resistant against all known attacks. So, it is a good and safe algorithm to
use, you can say that.
(Refer Slide Time: 15:41)

So, very briefly let us try to understand how AES works. The first thing is that depending
on the key size 128, 256 or 192 the number of rounds similar to DES can vary. Now this
table here shows that if key is 128 you need 10 rounds, 192 you need 12 rounds, 256 you
need 14 rounds. This table shows you that. The rounds are identical except the first and
last one where there is a small change that will show and each of the rounds there are
some basic operations which are carried out. What are the basic operations? First
operation is a substitution operation which is referred to as SubBytes.

Here every byte, you see block size is 128. So there will be 16 bytes. So, every byte is
substituted by some other byte. There is substitute. Something called substitution box
which is there inside. Using the SubBytes function, it replaces a byte by another byte. It
is basically a permutation function. Then there is something called shift rows which is a
transposition. The order is made different. the bits are not changed, but order is made
different of the rows.

Then there is a mixed column operation where different columns using some operator,
you combine the columns, mix them up in some certain way. So, this, there is a
complicated functionality here and finally, add round key. Similar to DES, round keys
are also generated here and whatever that 128 bit value is coming, you do a bit by bit
exclusive-OR with this round key to generate the 128 bit data for the next round.
So, this diagram if you see, it gives the overall schematic. You start with the 128 bit
plaintext which is coming. These are the rounds, but before starting with the first round
you do a pre round transformation. There is an initial transformation, after that there are
the rounds. The last round is slightly different. So, you see the first round is different in
the sense that there is something done just before that and similarly the last round is also
slightly different. And there is a key expansion module in the other side which takes as
input the key 128, 192 or 256 bits and it generates the round keys similar to DES. This is
how the structure of AES looks like.

(Refer Slide Time: 18:53)

Now, if you look into each of the rounds accepting the first and the last. So, the rounds,
these 4 functionalities occur in this order, substitution bytes, followed by shift rows,
followed by mixed column and finally, add round key where the round key is bit by bit
exclusive-OR with this number. And, another thing the 128 bit data that we are having
that you are encrypting and you are moving at each stage, this is actually represented as a
state vector.

There is a 4 by 4 matrix, there are 16 such states and each of them is an 8 bit quantity. 16
into 8 is 128. So, that is how this state is maintained and all these operators are defined
on that state matrix. That is how it is defined in AES.
(Refer Slide Time: 20:03)

And looking at the overall picture; so, here you see that in all the rounds, these operators
as I said are applied in subbyte, shiftrow, mix column and add round. In the first round
there is an additional add round key in the beginning and in the last round there is no
mixed column states, only sub byte shift row and add round. This is how it is done. You
see the structure as you can see, is fairly complicated, but these 4 basic functionalities I
talked about subbytes, shiftrow, mix column and add round key, these are defined in
such a way, they can be implemented very easily both using some hardware circuits.
Also using some instruction set of a computer they are efficient to implement, ok.

This was one of the objectives of these algorithms, how fast they can be implemented,
what is the maximum speed they can offer in various implementations, ok. So, with this
we come to the end of this lecture, where we have talked about some of the practical
private or symmetric key algorithms. Now, as I said whenever you are trying to secure a
network, you find that there are some vulnerabilities. There are a number of ways you try
to secure your infrastructure or your data or files, whatever, now this encryption is one of
the very commonly used tool or technique to provide you with the required level of
security.

You think of emails, you are sending mails. Some mails may be quite confidential, you
may want such confidential mails should be automatically encrypted so that no one
should be able to tap my messages, my mails. So, these encryption or decryption
techniques can be integrated with other applications in a suitable way so that they can
provide you with some levels of security as desired by some applications. In the next
lecture we shall be starting some discussion on public key cryptosystems and what are
the methods that can be used there.

Thank you.