386

Chapter 21:
Security
Learning objectives
By the end of this chapter you should be able to:
■ show understanding of how encryption works
■ show understanding of digital certification
■ show awareness of the Secure Socket Layer (SSL)/Transport Layer Security (TLS) protocols.
 Part 3: Chapter 21: Security

21.01 Encryption fundamentals

Encryption can be used as a routine procedure when storing data within a computing
system. However, the focus in this chapter is on the use of encryption when transmitting data
over a network.
There are three issues that will be considered in this chapter:
• is the encryption algorithm sufficiently robust to prevent the encrypted data being
decrypted by some unauthorised third-party?
• how is it possible to ensure that a secret key remains secret?
• how can the receiver of a communication be sure who sent the communication?
The use of encryption is illustrated in Figure 21.01.The process starts with original data
referred to as plaintext, whatever form it takes. The plaintext is encrypted by an encryption
algorithm which makes use of a key. The product of the encryption is ciphertext, which
is transmitted to the recipient. When the transmission is received it is decrypted using a
decryption algorithm and a key to produce the original plaintext.

Plaintext Encryption Ciphertext Decryption Plaintext

Key Key

Figure 21.01 Overview of encryption and decryption 387

KEY TERMS

Plaintext: data before encryption

Ciphertext: the result of applying an encryption algorithm to data

Security concerns
There are a number of security concerns relating to a transmission.
• Confidentiality: Only the intended recipient should be able to decrypt the ciphertext.
• Authenticity: The receiver must be certain who sent the ciphertext.
• Integrity: The ciphertext must not be modified during transmission.
• Non-repudiation: Neither sender nor receiver should be able to deny involvement in
the transmission.
• Availability: Nothing should happen to prevent the receiver from receiving the transmission.
This chapter will consider only confidentiality, authenticity and integrity.
The confidentiality concern arises because a message could be intercepted during
transmission and the contents read by an unauthorised person. The concern about integrity
reflects the fact that the transmission might be interfered with deliberately but also that
there might be accidental corruption of the data during transmission.

Encryption methods
The fundamental principle of encryption is that the encryption algorithm must not be
a secret: it must be in the public domain. In contrast, an encryption key must be secret.
 Cambridge International AS & A Level Computer Science

However, there are two alternative approaches. One is symmetric key encryption, and the
other is asymmetric key encryption also known as public key encryption.

KEY TERMS

Symmetric key encryption: one private key is held by both sender and receiver and is used for both
encryption and decryption
Asymmetric key encryption: there is a public key and a private key one of which is used for
encryption and the other for decryption

In symmetric key encryption there is just one key. This key is a secret shared by the sender
and the receiver of a message. The sender uses the encryption algorithm together with the
key to encrypt some plaintext. The receiver decrypts the ciphertext using the same key.
The issue with symmetric key encryption is delivery of the secret key. The sender needs the
key to encrypt but how can the key be securely delivered to the receiver to allow decryption?
In asymmetric key encryption two different keys are used, one for encryption and the other
one for decryption. Only one of these is a secret.
If asymmetric encryption is to be used the process is initiated by someone in possession of
two keys. One of these is a public key which is sent to anyone who is going to partake in an
encrypted communication. The other is a secret private key which is never sent to anyone.
Having a means of secure transmission of a secret key is no longer an issue.
The most likely scenario is that the holder of the two keys wishes to receive a transmission. In
388 this case a sender uses the public key to encrypt some plaintext and sends the ciphertext to
the receiver. The receiver is now the only person who can decrypt the message because the
private and public keys are a matched pair. The public key can be provided to any number of
different people allowing the receiver to receive a private message from any of the different
people. There are two points to note here.
• If two people require two-way communication, both communicators need a private key
and must send the matching public key to the other person.
• There are two requirements to ensure confidentiality should the transmission be
intercepted and the message extracted: the encryption algorithm must be complex and
the number of bits used to define the key must be large.

Question 21.01
One method used by an unauthorised person attempting to decrypt a message is called a
brute-force attack where all possible values for the key are tried. Calculate how long it would
take to try all possibilities for a 64-bit or 128-bit key, assuming each attempt took 1 second.
The above account does not completely answer the question of how encryption works.
The missing factor is an organisation to provide keys and to ensure their safe delivery to
individuals using them. This will be discussed in the next section.

21.02 Digital signatures and digital certificates

Using asymmetric encryption, the decryption–encryption works if the keys are used the
other way round. An individual can encrypt a message with a private key and send this to
a recipient who has the corresponding public key and who can then use this to decrypt
the received ciphertext. This approach would not be used if the content of a message was
 Part 3: Chapter 21: Security

confidential because anyone might be in possession of the public key. However, it could be
used if it was important to verify who the sender was. Only the sender has the private key
and the public key only works with that one specific private key. Therefore, if the recipient
finds that the decryption is successful, the message has in effect been received with a digital
signature identifying the sender.

Cryptographic
Message hash Digest
function

Sender’s Encryption Digital

private key signature

Figure 21.02 Sender using a one-way hash function to send a digital signature

There is a disadvantage in using this method of applying a digital signature: it is associated

with an encryption of the whole of a message. An alternative is for the sender to use a public
cryptographic one-way hash function which creates a number that is uniquely defined for
the particular message, called a ‘digest’. The process at the sender’s end of the transmission
is outlined in Figure 21.02. The private key is used to encrypt the digest. The encrypted
digest is the digital signature. The message can be transmitted as plaintext together with
the encrypted digest as a separate file. Because the digest is much smaller than the whole
message the encryption and the transmission are faster processes than if the whole message
were encrypted. 389

The processes that take place at the receiver end are outlined in Figure 21.03. The same
public one-way hash function is used to create a digest from the received message. Then the
encrypted version of the original digest is decrypted using the public key.
If the two digests are identical the receiver can be confident that the message is authentic
and has been transmitted unaltered.
Note that the digital signature is different each time this process is used. This is because the
digest is uniquely defined by the hash function being applied to that particular message.

Cryptographic
Message hash Digest
function

Digital
Decryption Digest
signature

Sender’s
public key

Figure 21.03 Receiver checking that the received transmission is authentic and unchanged
 Cambridge International AS & A Level Computer Science

However, the authenticity only confirms to the receiver that the message was sent from the
person who had sent them the public key. It does not consider the fact that someone might
create a public key and pretend to be someone else.
Therefore, a more strict way of ensuring authentication is needed. This can be provided by a
Certification Authority (CA) as part of a Public Key Infrastructure (PKI).

Certification
Person A Person A’s public key Authority
(the receiver) (CA)

Person A’s CA’s

Person A’s public key private
digital key
Person A’s
certificate digital certificate
Digital
Person A’s certificate
website creation
process

Figure 21.04 Processes involved in obtaining a digital certificate

Let’s consider a would-be receiver who has a public–private key pair. The receiver wants to
be able to receive secure messages from other individuals, and these individuals want to be
confident about the identity of the receiver. The public key must be made available in a way
390
that ensures authentication. The steps taken by the would-be receiver to obtain a digital
certificate to allow safe public key delivery are illustrated in Figure 21.04. The process can be
summarised as follows.
1 An individual (person A) who is a would-be receiver and has a public–private key pair
contacts a local CA.
2 The CA confirms the identity of person A.
3 Person A’s public key is given to the CA.
4 The CA creates a public-key certificate (a digital certificate) and writes person A’s public
key into this document.
5 The CA uses encryption with the CA’s private key to add a digital signature to this document.
6 The digital certificate is given to person A.
7 Person A posts the digital certificate on a website.
Figure 21.04 shows person A placing the digital certificate on that person’s website but another
option is to post it on a website designed specifically for keeping digital certificate data.
Anyone who wishes to extract the public key from the certificate has to use the CA’s public key.
For this overall process to work there is a need for standards to be defined regarding the
public key infrastructure and the production of the digital certificate. As ever, the name for
the standard, X.509, is not very memorable.

TIP
There are two similar processes that have been discussed. In one case someone with a private key
sends a public key to someone else. In another case the CA sends a digital certificate containing a
public key. Try not to confuse these two.
 Part 3: Chapter 21: Security

The following are a few notes to summarise the options available.

•• The starting position is someone who has a public–private key pair which are associated
with a specific asymmetric key encryption algorithm.
•• This person could just make the public key available to anyone who is going to be either a
sender or a receiver.
•• For optimum security the person instead sends the public key to a Certification Authority.
•• The Certification Authority creates a digital certificate which contains the public key with
proof of the ownership of the public key.
•• Anyone wishing to use the public key obtains it from this digital certificate.
•• A message encrypted with the public key could be sent to the owner of the private key.
•• A message encrypted with the private key could be sent to anyone having the public key.
•• The owner of the private key could use it to create a digital signature that could be used to
authenticate an email as was suggested in Chapter 9 (Section 9.03).

21.03 Symmetric key encryption methods

For many years the Data Encryption Standard (DES) was the normal choice for symmetric key
encryption. As the weakness of DES became a problem, Triple DES took its place. In 2001 the
Advanced Encryption Standard (AES) was introduced as a superior approach. For education
purposes only a simplified DES (S-DES) was introduced which allowed the sort of operations
performed in encryption to be better understood. The following is an overview of S-DES.
S-DES is an example of a block cipher which means encrypting blocks of bits. In S-DES 8-bit 391
blocks are encrypted. A 10-bit key is chosen. The first stage is to create two 8-bit keys from
the 10-bit key. The first step in this first stage is a permutation (reordering of digits) which can
be illustrated by the following.
Suppose that the 10-bit key is chosen to be 0101010101
when subjected to a permutation which can be represented by 3 5 2 7 4 10 1 9 8 6
it becomes 0010110011
The numbering of the bit positions is read from left to right so the new 10-bit version has the
old position 3 value followed by the old position 5 value and so on. The next step is to apply
a circular left shift to the first 5 bits and to the last 5 bits. This produces 0101000111. Finally,
the first of the two 8-bit keys to be used in the encryption is created using the permutation
6 3 7 4 8 5 10 9. In our example this key is 00011011. A slightly modified version of this is used
to create a second 8-bit key.
The second stage is the encryption, which is a five step process.
1 An initial permutation.
2 Application of a function using the permuted code and the first encryption key.
3 A switch of the first and last 4-bit parts.
4 A repeat application of the function but this time with the second encryption key.
5 A final permutation using the reverse of the initial permutation sequence.
The decryption by the receiver of a transmission uses the same generated 8-bit keys and
follows the reverse of the above process.
 Cambridge International AS & A Level Computer Science

The AES standard defines the block length as 128 bits. The user can choose to use 128, 192 or
256 bits for the key. The chances of the key being identified from the transmitted ciphertext
are small. The main concern is the safety of the method used to provide the key to both
sender and receiver.

Extended Question 21.01

Would you like to investigate S-DES further? You could attempt an encryption and
decryption.

21.04 Public key encryption methods

RSA (Rivest-Shamir-Adleman), the usual method for public key encryption, is named after
the three people who created it. The major features of the method are the key generation
algorithm and the encryption function.
The key generation can be summarised as follows.
1 Two very large prime numbers p and q are chosen and their product n is calculated.
2 The product (p-1)(q-1) is calculated.
3 A prime number e less than (p-1)(q-1) and not a factor of it is chosen (65537 is the usual choice).
4 Another number d is found which satisfies the condition that the product of d times e
when divided by (p-1)(q-1) gives a remainder of 1.
5 The public key becomes the pair (n,e).
6 The private key becomes the pair (n,d).
392
The security of the algorithm depends on the fact that finding factors of a very large number
is not feasible within any reasonable time scale. Computing n from p and q is straightforward
but deducing p and q given n is not!
The encryption works on numbers so a text to be encrypted must first have the characters
replaced by numbers according to a sensible coding scheme. If such a number x is to be
encrypted as y then y is calculated so that the following relationship holds:
y = xe mod n
A similar relationship involving d rather than e is used for decryption.
Public key encryption is inherently more secure than symmetric key encryption but the
algorithms are not as fast. It is quite common for public key encryption to be used to deliver
securely a key that can then be used for symmetric key encryption.

21.05 SSL and TLS

When we access a website, we have two concerns. The first is whether or not the website
is genuine. The second is whether we can transfer sensitive personal data to the website,
for example to buy a product offered for sale on the website. The Secure Socket Layer (SSL)
protocol was created to give assurance to a website user when a client–server application is
underway. As described in Chapter 17 (Section 17.04), the interface between an application
and TCP uses a port number. In the absence of a security protocol, TCP services an
application using the port number. The combination of an IP address and a port number is
called a ‘socket’. When the Secure Socket Layer protocol is implemented it functions as an
additional layer between TCP in the transport layer and the application layer. When the SSL
protocol is in place, the application protocol HTTP becomes HTTPS.
 Part 3: Chapter 21: Security

The following are some facts concerning SSL.

•• Although SSL is referred to as a protocol, it is in fact a protocol suite.
•• There is a Record Protocol that deals with the format for data transmission.
•• There is also a Handshake Protocol responsible for security.
•• The operation of SSL happens without any action from the user.
•• The starting point for SSL implementation is a connection between the client and the
server being established by TCP.
•• The client browser then invokes the Handshake Protocol from the SSP suite.
•• The Handshake Protocol requests from the server its SSL certificate which is a digital
certificate confirming its identity.
•• The server sends this SSL certificate plus its public key.
•• The browser uses this public key to encrypt a key which is to be used as a one-off session
key for symmetric key encryption to be used for the data transfer during the session.
•• There may also be a need at this time to agree which encryption algorithms are to be used.
SSL was originally a proprietary protocol. However, it was taken over by the Internet
Engineering Task Force (IETF) in order for a standardised version to be produced. This
progressed to version 3.0. When the IETF realised that an improved version was required it
decided that a new name was appropriate. Transport Layer Security (TLS) is an upgraded
version of SSL recommended for use because of some security concerns with the use of SSL.
Despite this SSL is still in general use.

Discussion Point: 393

The use of encryption has always been a controversial subject. There are two important
aspects to this. The first is whether powerful, unbreakable encryption algorithms should be
made available to the public. The second relates to the key escrow scheme, which allows
governments access to all secret keys. You might wish to consider how the content here has
relevance to some of the topics in Chapter 9 and Chapter 10.

21.06 Quantum cryptography

Quantum mechanics provides fundamental laws of physics applicable to the behaviour of
particles. The particles that transmit light are called photons. Photons demonstrate wave
behaviour, so that each photon appears to vibrate in a particular direction at right angles
to its direction of travel. The direction each photon vibrates in is called its polarisation,
and is represented in a diagram as a double-ended arrow. A photon can be created with a
specific polarisation to represent a value for a bit. If we allow four possibilities for the state of
polarisation there are two ways to represent a 1 and two ways to represent a 0. This can be
illustrated as follows:

=1 = 0 = 1 =0

This scheme can be used to enable a sender and receiver to create a ‘shared secret’ code
consisting of a number of bits. Table 21.01 illustrates the process.
 Cambridge International AS & A Level Computer Science

Bit values sent 1 0 1 1 0 0 0 1 0 1

Polarisation basis + × × + + × × + × +

Polarisation chosen by receiver × + + + × × + × × +

Bit values confirmed 1 0 0 1

Table 21.01 Creation of a ‘shared secret’ code

In Table 21.01 the first row indicates the bit sent, the second row shows the basis used for
this with + representing the one using vertical and horizontal and x representing the diagonal
pair. The sender chooses the bit pattern at random and also the polarisation basis for each
value at random. The third row shows the receiver’s choice for the polarisation basis for
each value. Again this is a random choice. Following the transmission the sender informs the
receiver about the polarisation basis used for each value. The receiver responds by saying
which ones were chosen to match. For these matches there is now a stored value for a bit. In
the example shown a ‘shared secret’ code 1001 has been created.
The above scheme has been incorporated in Quantum Key Distribution (QKD) systems.
Earlier in this chapter it was mentioned how a key might be transferred using asymmetric key
encryption for subsequent use in symmetric key encryption. QKD offers an alternative for the
key transfer. The key is still to be used in the normal way. The advantage of QKD is that the
transfer does not involve defined values just photons. Anyone trying to intercept the flow of
photons in an attempt to discover their polarisation will by the laws of quantum mechanics
394 destroy the photons. A photon cannot be detected and measured then sent on again. The
main drawback of quantum cryptography is that it cannot be implemented using standard
communication media. It requires a dedicated, special purpose ‘quantum channel’ between
sender and receiver. The costs of providing this are very high so routine use is unlikely.
There are ambitious hopes for the future of quantum computing but progress is slow. QKD is
one of the few examples where there has been significant success evidenced.

TASK 21.01
The concept of a ‘shared secret’ has been used in traditional encryption schemes.
An example is the Diffie–Hellman key agreement method. Investigate the principles
behind how this works.

Reflection Point:
Chapter 21: Security has some difficult concepts. Have you made sure that you understand
the overriding principles and definitions used?