Introduction to Cryptography

Editor's Notes

• #2: Welcome to this course on Cryptography and Network Security. I am Adri Jovin and I will be guiding you throughout this course. I am a Assistant Professor in the Department of Information Technology in Sri Ramakrishna Institute of Technology and I hold a PhD in Computer Science and Engineering. My research work is a blend of Artificial Intelligence, Cryptography, Network Security and obviously Distributed Systems. Let us now dive into the introduction to cryptography.
• #3: Anyone who is really interested in this course must first understand the necessity for cryptography and network security. The graph represented here is inspired by the graph which was released by CERT in 2001. This was further extended by certain researchers from the University of Adelaide till the year 2016. But the inference from this graph is quite common, the attack sophistication has increased over the period of time. It can also be observed that, with the raise of sophistication using tools, the knowledge required by an adversary over the resource which is required to be exploited is low. Hence it is very hard for any potential resource to counter an attack. It would be found that a lot of attacks are listed in the graph. I shall provide a outline of these attacks sometime in future. In this course the first four modules are designed to obtain a basic insight on various cryptographic techniques and the last module provides an outline of the issues and applications deployed in network security.
• #4: On going back into historical times, it could be observed that Cryptography had been an art of writing and solving codes. Cryptography found its role in engineering and especially in Computer Engineering in the late 1980s. It basically deals with mathematics, in specific Discrete Mathematics. There are historical evidences that cryptography had provoked its role during the regime of the Egyptian Civilization. Cryptography is commonly used for diplomatic communications, message transmission during war times. No one can forget the historical usage of "Enigma", a famous cryptographic technique, by the German Army during World War II. Apart from these applications, cryptography is used to protect the data carried by individuals and corporates.
• #5: These are a few keywords which need to be known by any one who deals with cryptography and network security. I am not going to elaborate on these terms rather, I am providng you with the spark to ignite the flame of cryptography in you. More terms could be found in RFC 2828 which provides the definition for all the terms used with respect to cryptography. This document could be found in the official website of IETF, the Internet Engineering Task Force. If you are really interested, you can go through the RFC 2828 to know more about these terms.
• #6: One of the basic part of security is the OSI security architecture which defines what attacks are, what are the various security mechanisms and what security services need to be satisfied by any service provider who offers a communication service. Let us get into the definitions. An attack is any action initiated by an adversary that compromises the security of an information owned by an organisation, and an individual as well. A security mechanism is a process that is designed to detect, prevent or recover from any security attack. A security service is one which enhances the security of the data processing systems and information transfers within and between organizations.
• #7: Two common terminologies used in the domain of security are threat and attack. Most people find it hard to differentiate between these terms. Let me try to help you differentiate between these terminologies. A threat is a possible danger which might exploit a vulnerability. It may be a bug in a system which may or may not be noticed by the developer at the time of release of a software. An intelligent adversary is capable of identifying this vulnerability and initiate an attack. An attack is the process in which an adversary assaults a system using the vulnerability in the system. The attacker uses it to evade the security services and violate the security policy of the system. For an example, most users may be unaware of open ports in a system. The open port acts as a threat. It may be used by an attacker to have access over some document in your system. This is termed the attack. Hope, I have clearly distinguished between a threat and an attack.
• #8: Coming to the attacks, attacks are basically classified into two types namely Passive attack and Active attack. Hearing the terms passive and active, some may think that the impact of passive attacks may be less compared to that of an active attack. Practically speaking, both passive and active attacks are comparatively dangerous and the level of damage created depends on the attitude of the adversary. Coming down to a level in specific, passive attacks attempt to learn or make use of the information from the system without affecting any of the system resources. Hence, it makes it very difficult to find out whether an attack has really happened or not. On the other hand acrive attacks directly have their hand over the resources or affect the operations relevant to the resource of interest which may result in modification of the data stream or crafting of a false data stream which may almost resemble the original data stream. Release of message content and traffic analysis are certain examples of passive attacks. Masquerading, replay, modification and denial of service are examples of active attacks. The illustrations of the examples mentioned here are discussed in the forecoming slides.
• #9: Let us start from the passive attacks. The first one is the 'Release of Message Contents'. It could be observed that the names of the sender is named Bob and the receiver is named Alice and adversary is named Darth. Some people consider this as a formal representation and some other do not. The characters are named, so that it makes it convenient for people to remember the parties involved in a communication rather than getting confused with charateristic notations like A, B or D. The first mention of these characters appeared in the context of cryptography in a article published by Rivest, Shamir and Adleman in 1978. Apart from the perspective of understanding neither Alice nor Bob has a significant contribution in cryptography. There are further more characters which you can find if you go through the wikipedia page of Alice and Bob. Now, let us come to the subject. Bob intends to send a message to Alice in a open communication channel. In the event of the message being sent, Darth taps into the channel and exposes the message to some other party or to the public. Many of you would have come across news channels which provide news on tapped messages between political leaders or in Wikileaks where you could find confidential government documents which are exposed to the public.
• #10: The traffic analysis exhibits a high degree of passiveness. When Bob communicates with Alice, Darth really do not tap the communication channel and expose any message rather he just observes the communication between the two parties. This may be a potential information for some other party who wants to understand the relationship between Alice and Bob.
• #11: Now, coming to the active attacks. The first example is the masquerading attack. This is similar to the physical forgery which we have experienced. Here Darth pretends to be Bob and send a message to Alice. We too might have received bogus e-mails stating that we are eligible for some offer or requesting some money, under the banner of a reputed company. These messages are crafted in such a way that it resembles an authenticated message from the company.
• #12: The second example quoted here is the "Replay" attack. In this scenario, Bob sends a message to Alice. Darth gets a copy of the message and send it again and again to Alice, sometimes pushes Alice to a state to conclude that the legitimate message send by the Bob is also a crafted one. The continuous replay may result in another state called "Denial of Service".
• #13: The next example for active attack is the modification of message. The message when getting transmitted from one media to another is subjected to a number of transformations. Hence it is vulnerable to modifications, even if a change is done in the bit level transmission. However, there are incidents observed where an adversary intentionally intervene and manipulate the message sent by Bob to Alice.
• #14: The last example is the Denial of Service. Here Bob is a legitimate user who has acces towards a server. Darth, by some means restrict the access of Bob to the server. One of the possibilities may be flooding the traffic which makes the router so busy in managing the flood rather than caring for the request of Bob. It is usually less possible to get access to a resource amidst a flooded network. Till now we have come across an overview of the attacks. Let us now move towards the services which need to be ensured by a service provider while offering a communication service. These services are provided depending on the type of resource involved in the communication.
• #15: X.800 defines a security service as a service provided by a protocol layer of communicating open systems, which ensures adequate security of the systems or of data transfers. Earlier I had discussed about RFC 2828. This document provides an elaborate definition for all the services discussed in this part. X.800 categorises services into 5 divisions which may have further sub-divisions too. The top 5 divisions are Authentication, Access Control, Data Confidentiality, Data Integrity and Nonrepudiation. Let us start with authentication. Authentication is a service that assures that a communication is authentic or trustable. Though the term trust has a different meaning in specific terms of security, we discuss the term in a broader perspective for now. On receipt of a message from the sender, the receiver must be able to assure that the message is from the source which it claims to be. Secondly it has to ensure that the communication is not intiated or interfered by a masquerading adversary. X.800 defines two authentication services namely peer entity authentication and data origin authentication. The peer entity authentication provides the evidence to ensure the identity of a peer entity in an association. This ensures that the authenticating authority is not a masquerade or an unathorised replay of a previous connection. Data origin authentication provides evidence that ensures the source of a data unit. This is used in applications like e-mail. However, the data origin authentication does not provide any protection against duplication or modification. Access control is defined as the ability to limit or control the access of a resource in any system. Any entity trying to gain access over a resource need to be identified or authenticated inoder to get its access rights tailored. Confidentiality is the service which protects a message from passive attacks.This ensures that the message is not accessed by any entity apart from the entities involved in the communication. There are four classes of confidentiality namely connection confidentiality, connectionless confidentiality, selective-field confidentiality and traffic flow confidentiality. Connection confidentiality ensures the protection of all user data in a connection. Connectionless confidentiality provides protection of all user data in a single data block. Selective-field confidentiality provides confidentiality of selected fields within the user data in a connection or a data block. Traffic flow confidentiality protects the infromation about the flow of traffic, which is much effective against the traffic analysis attack mentioned in the passive attack. Data integrity assures that the data received at the destination is exactly the same as sent by an authorised entity. It also ensure that no modification, insertion, deletion or replay is done over the message. Data integrity has five sub-divisions namely connection integrity with recovery, connection integrity without recover, selective-field connection integrity, connectionless integrity and selective-field connectionless integrity. Connection integrity with recovery ensures the integrity of all user data in a connection and detects any modification, insertion, deletion, or replay of any data within an entire data sequence. In case of any modification being detected, recovery will be attempted. Connection integrity without recovery ensures the integrity of all user data in a connection and detects any modification, insertion, deletion, or replay of any data within an entire data sequence. This helps in detection but does not attempt any recovery. Selective-Field Connection Integrity ensures the integrity of selected fields within the user data of a data block transferred over a connection and helps in identification whether the selected fields have been modified, inserted, deleted, or replayed. Connectionless Integrity ensures the integrity of a single connectionless data block and helps in the detection of data modification. Selective-Field Connectionless Integrity ensures the integrity of selected fields within a single connectionless data block. It also helps in the determination of whether the selected fields have been modified or not. Nonrepudiation prevents the sender or receiver from denying a transmitted message. There are two types of nonrepudiation namely nonrepudiation at origin and nonrepudiation at destination. Nonrepudiation at origin displays or ensures the proof that the message was sent by the specified party. Nonrepudiation at destination ensures the proof that the message was received by the specified party. Therefore the event of sending as well reception cannot by denied by the parties involved in the communication. Apart from the above mentioned services there is another service called availability. Both X.800 and RFC 2828 define availability to be the property of a system or a system resource being accessible and usable upon demand by an authorized system entity. Denial or service is an attack that prevents the availability service.
• #16: Peeking into the security mechanisms, the security mechanisms listed as per X.800 are given here. Security mechanisms may basically be classified into Specific Security Mechanisms and Pervasive Security Mechanisms. The functionalities of specific security mechanisms are confined to a particular protocol layer whereas the pervasive security mechanisms are not specific to any service mentioned earlier or to any protocol layer. Let us first go through the specific security mechanisms. Encipherment is a mechanism which utilises mathematical algorithms to transform the data into a form which is not readily intelligible. This may deploy a cryptographic algorithm to achieve this. Digital signature is a mechanism that helps the users to ensure the identity of the sender and to verify the integrity of the message. There are different techniques associated with this which shall be discussed in later parts of this course. Access Control deploys mechanisms like Access Control Lists to enforce access rights to resources. Data integrity deploys mechanisms like checksums, message digests etc. to assure the integrity of a data unit or a stream of data. Authentication exchange is a mechanism intended to ensure the identity of an entity involved in the communication by mutual or trusted information exchange. Traffic padding is a mechanism in which padding bits are inserted in a data stream inorder to divert the adversary from attempts towards traffic analysis. Routing control helps in availing secure routes for certain data. It also helps in change of routes when a security breach is suspected. Notarization is a mechanism which employs a trusted third party to endorse the trustabiliity of data or resources exchanged during a communication. Coming to the pervasive security mechanisms, the trusted functionality ensures that the perceived data or resource is correct with respect to a certain criteria like a security policy. Security label marks a resource to designate it of possessing the security attributes required for the communication. Event detection helps in the recording and detection of security-relevant events like a policy violation. Security audit trail is a mechanism in which the data or events are collected inorder to subject them to a security audit. Security recovery deals with requests from event handling and management functions and takes recovery actions.
• #17: The table described here clearly depicts the relationship between the services and mechanisms. For example, it could be observed that notarization ensures nonrepudiation.
• #18: These are the reference text which I had utilised for the preparation of this material. Apart from these the major reference includes the book "Cryptography and Network Security" authored by William Stallings. You can use any book from 4th edition to 6th edition for these topics to get covered.
• #19: If you have any questions, please feel free to interact through the discussion forum.