Introduction

Cryptography is the science of making information secure in the presence of adversaries. It does so under the assumption that limitless resources are available to adversaries. Ciphers are algorithms used to encrypt or decrypt data so that if intercepted by an adversary, the data is meaningless to them without decryption, which requires a secret key.

Cryptography is primarily used to provide a confidentiality service. On its own, it cannot be considered a complete solution, rather it serves as a crucial building block within a more extensive security system to address a security problem. For example, securing a blockchain ecosystem requires many different cryptographic primitives, such as hash functions, symmetric key cryptography, digital signatures, and public key cryptography.

In addition to a confidentiality service, cryptography also provides other security services such as integrity, authentication (entity authentication and data origin authentication), and non-repudiation. Additionally, accountability is also provided, which is a requirement in many security systems.

Cryptography

A generic cryptographic model is shown in the following diagram:

Figure 3.1: A model of the generic encryption and decryption model

In the preceding diagram, P, E, C, and D represent plaintext, encryption, ciphertext, and decryption, respectively. This model utilizes concepts such as entities, senders, receivers, adversaries, keys, and channels. These concepts are explained as follows:

• Entity: Either a person or system that sends, receives, or performs operations on data
• Sender: This is an entity that transmits the data
• Receiver: This is an entity that takes delivery of the data
• Adversary: This is an entity that tries to circumvent the security service
• Key: A key is data that is used to encrypt the plaintext and also to decrypt the ciphertext
• Channel: A channel provides a medium of communication between entities

We mentioned the fact that cryptography provides various services. In the following sections, we'll discuss these services in greater detail.

Confidentiality

Confidentiality is the assurance that information is only available to authorized entities.

Integrity

Integrity is the assurance that information is modifiable only by authorized entities.

Authentication

Authentication provides assurance about the identity of an entity or the validity of a message.

There are two types of authentication mechanisms, namely, entity authentication and data origin authentication, which are discussed in the following sections.

Entity authentication

Entity authentication is the assurance that an entity is currently involved and active in a communication session. Traditionally, users are issued a username and password that is used to gain access to the various platforms with which they are working. This practice is known as single-factor authentication, as there is only one factor involved, namely, something you know; that is, the password and username.

This type of authentication is not very secure for a variety of reasons, for example, password leakage; therefore, additional factors are now commonly used to provide better security. The use of additional techniques for user identification is known as multi-factor authentication (or two-factor authentication if only two methods are used).

Various authentication methods are described here:

The first method uses something you have, such as a hardware token or a smart card. In this case, a user can use a hardware token in addition to login credentials to gain access to a system. This mechanism protects the user by requiring two factors of authentication. A user who has access to the hardware token and knows the login credentials will be able to access the system. Both factors should be available to gain access to the system, thus making this method a two-factor authentication mechanism. If the hardware token is inaccessible (for example, lost or stolen) or the login password is forgotten by the user, access to the system will be denied. The hardware token won't be of any use on its own unless the login password (something you know) is also known and used in conjunction with the hardware token.

The second method uses something you are, which uses biometric features to identify the user. With this method, a user's fingerprint, retina, iris, or hand geometry is used to provide an additional factor for authentication. This way, it can be ensured that the user was indeed present during the authentication process, as biometric features are unique to every individual. However, careful implementation is required to guarantee a high level of security, as some research has suggested that biometric systems can be circumvented under specific conditions.

Data origin authentication

Also known as message authentication, data origin authentication is an assurance that the source of the information is indeed verified. Data origin authentication guarantees data integrity because, if a source is corroborated, then the data must not have been altered. Various methods, such as Message Authentication Codes (MACs) and digital signatures, are most commonly used. These terms will be explained in detail later in the chapter.

Non-repudiation

Non-repudiation is the assurance that an entity cannot deny a previous commitment or action by providing incontrovertible evidence. This is a security service that offers definitive proof that a particular activity has occurred. This property is essential in debatable situations whereby an entity has denied the actions performed, for example, the placement of an order on an e-commerce system. This service produces cryptographic evidence in electronic transactions so that in case of disputes, it can be used as confirmation of an action.

Non-repudiation has been an active research area for many years. Disputes in electronic transactions are a common issue, and there is a need to address them to increase consumers' confidence levels in such services.

The non-repudiation protocol usually runs in a communications network, and it is used to provide evidence that an action has been taken by an entity (originator or recipient) on the network. In this context, two communications models can be used to transfer messages from originator A to recipient B:

• A message is sent directly from originator A to recipient B; and
• A message is sent to a delivery agent from originator A, which then delivers the message to recipient B

The primary requirements of a non-repudiation protocol are fairness, effectiveness, and timeliness. In many scenarios, there are multiple participants involved in a transaction, as opposed to only two parties. For example, in electronic trading systems, there can be many entities, including clearing agents, brokers, and traders, who can be involved in a single transaction. In this case, two-party non-repudiation protocols are not appropriate. To address this problem, multi-party non-repudiation (MPNR) protocols have been developed.

MPNR refers to non-repudiation protocols that run between multiple parties instead of the traditional two parties.

Accountability

Accountability is the assurance that actions affecting security can be traced back to the responsible party. This is usually provided by logging and audit mechanisms in systems where a detailed audit is required due to the nature of the business, for example, in electronic trading systems. Detailed logs are vital to trace an entity's actions, such as when a trade is placed in an audit record with the date and timestamp and the entity's identity is generated and saved in the log file. This log file can optionally be encrypted and be part of the database or a standalone ASCII text log file on a system.

In order to provide all of the services discussed in this section, different cryptographic primitives are used, which are presented in the next section.