engineering electronics and communication

Editor's Notes

• #1: Lecture slides by Lawrie Brown for “Cryptography and Network Security”, 4/e, by William Stallings, Chapter 2 – “Classical Encryption Techniques”.
• #2: Opening quote.
• #3: Symmetric encryption, also referred to as conventional encryption or single-key encryption, was the only type of encryption in use prior to the development of public-key encryption in the 1970s. It remains by far the most widely used of the two types of encryption. All traditional schemes are symmetric / single key / private-key encryption algorithms, with a single key, used for both encryption and decryption. Since both sender and receiver are equivalent, either can encrypt or decrypt messages using that common key.
• #4: Briefly review some terminology used throughout the course.
• #5: Detail the five ingredients of the symmetric cipher model, shown in Stallings Figure 2.1: plaintext - original message encryption algorithm – performs substitutions/transformations on plaintext secret key – control exact substitutions/transformations used in encryption algorithm ciphertext - scrambled message decryption algorithm – inverse of encryption algorithm
• #6: We assume that it is impractical to decrypt a message on the basis of the cipher- text plus knowledge of the encryption/decryption algorithm, and do not need to keep the algorithm secret; rather we only need to keep the key secret. This feature of symmetric encryption is what makes it feasible for widespread use. It allows easy distribution of s/w and h/w implementations. Can take a closer look at the essential elements of a symmetric encryption scheme: mathematically it can be considered a pair of functions with: plaintext X, ciphertext Y, key K, encryption algorithm EK, decryption algorithm DK.
• #7: Cryptographic systems can be characterized along these three independent dimensions.
• #8: Typically objective is to recover the key in use rather then simply to recover the plaintext of a single ciphertext. There are two general approaches: Cryptanalytic attacks rely on the nature of the algorithm plus perhaps some knowledge of the general characteristics of the plaintext or even some sample plaintext-ciphertext pairs. Brute-force attacks try every possible key on a piece of ciphertext until an intelligible translation into plaintext is obtained. On average,half of all possible keys must be tried to achieve success.
• #9: Stallings Table 2.1 summarizes the various types of cryptanalytic attacks, based on the amount of information known to the cryptanalyst, from least to most. The most difficult problem is presented when all that is available is the ciphertext only. In some cases, not even the encryption algorithm is known, but in general we can assume that the opponent does know the algorithm used for encryption. Then with increasing information have the other attacks. Generally, an encryption algorithm is designed to withstand a known-plaintext attack.
• #10: Two more definitions are worthy of note. An encryption scheme is unconditionally secure if the ciphertext generated by the scheme does not contain enough information to determine uniquely the corresponding plaintext, no matter how much ciphertext is available. An encryption scheme is said to be computationally secure if either the cost of breaking the cipher exceeds the value of the encrypted information, or the time required to break the cipher exceeds the useful lifetime of the information. Unconditional security would be nice, but the only known such cipher is the one-time pad (later). For all reasonable encryption algorithms, we have to assume computational security where it either takes too long, or is too expensive, to bother breaking the cipher.
• #11: A brute-force attack involves trying every possible key until an intelligible translation of the ciphertext into plaintext is obtained. On average, half of all possible keys must be tried to achieve success. Stallings Table 2.2 shows how much time is required to conduct a brute-force attack, for various common key sizes (DES is 56, AES is 128, Triple-DES is 168, plus general mono-alphabetic cipher), where either a single system or a million parallel systems, are used.
• #12: In this section and the next, we examine a sampling of what might be called classical encryption techniques. A study of these techniques enables us to illustrate the basic approaches to symmetric encryption used today and the types of cryptanalytic attacks that must be anticipated. The two basic building blocks of all encryption technique are substitution and transposition. We examine these in the next two sections. Finally, we discuss a system that combine both substitution and transposition.
• #13: Substitution ciphers form the first of the fundamental building blocks. The core idea is to replace one basic unit (letter/byte) with another. Whilst the early Greeks described several substitution ciphers, the first attested use in military affairs of one was by Julius Caesar, described by him in Gallic Wars (cf. Kahn pp83-84). Still call any cipher using a simple letter shift a caesar cipher, not just those with shift 3.
• #14: This mathematical description uses modulo (clock) arithmetic. Here, when you reach Z you go back to A and start again. Mod 26 implies that when you reach 26, you use 0 instead (ie the letter after Z, or 25 + 1 goes to A or 0). Example: howdy (7,14,22,3,24) encrypted using key f (ie a shift of 5) is MTBID
• #15: With a caesar cipher, there are only 26 possible keys, of which only 25 are of any use, since mapping A to A etc doesn't really obscure the message! Note this basic rule of cryptanalysis "check to ensure the cipher operator hasn't goofed and sent a plaintext message by mistake"! Can try each of the keys (shifts) in turn, until can recognise the original message. See Stallings Fig 2.3 for example of search. Note: as mentioned before, do need to be able to recognise when have an original message (ie is it English or whatever). Usually easy for humans, hard for computers. Though if using say compressed data could be much harder. Example "GCUA VQ DTGCM" when broken gives "easy to break", with a shift of 2 (key C).
• #16: With only 25 possible keys, the Caesar cipher is far from secure. A dramatic increase in the key space can be achieved by allowing an arbitrary substitution, where the translation alphabet can be any permutation of the 26 alphabetic characters. See example translation alphabet, and an encrypted message using it.
• #17: Note that even given the very large number of keys, being 10 orders of magnitude greater than the key space for DES, the monoalphabetic substitution cipher is not secure, because it does not sufficiently obscure the underlying language characteristics.
• #18: As the example shows, we don't actually need all the letters in order to understand written English text. Here vowels were removed, but they're not the only redundancy. cf written Hebrew has no vowels for same reason. Are usually familiar with "party conversations", can hear one person speaking out of hubbub of many, again because of redundancy in aural language also. This redundancy is also the reason we can compress text files, the computer can derive a more compact encoding without losing any information. Basic idea is to count the relative frequencies of letters, and note the resulting pattern.
• #19: Note that all human languages have varying letter frequencies, though the number of letters and their frequencies varies. Stallings Figure 2.5 shows English letter frequencies. Seberry & Pieprzyk, "Cryptography - An Introduction to Computer Security", Prentice-Hall 1989, Appendix A has letter frequency graphs for 20 languages (most European & Japanese & Malay).
• #20: The simplicity and strength of the monoalphabetic substitution cipher meant it dominated cryptographic use for the first millenium AD. It was broken by Arabic scientists. The earliest known description is in Abu al-Kindi's "A Manuscript on Deciphering Cryptographic Messages", published in the 9th century but only rediscovered in 1987 in Istanbul, but other later works also attest to their knowledge of the field. Monoalphabetic ciphers are easy to break because they reflect the frequency data of the original alphabet. The cryptanalyst looks for a mapping between the observed pattern in the ciphertext, and the known source language letter frequencies. If English, look for peaks at: A-E-I triple, NO pair, RST triple, and troughs at: JK, X-Z.
• #21: Illustrate the process with this example from the text in Stallings section 2.2.
• #22: Consider ways to reduce the "spikyness" of natural language text, since if just map one letter always to another, the frequency distribution is just shuffled. One approach is to encrypt more than one letter at once. The Playfair cipher is an example of doing this.
• #23: The best-known multiple-letter encryption cipher is the Playfair, which treats digrams in the plaintext as single units and translates these units into ciphertext digrams. The Playfair algorithm is based on the use of a 5x5 matrix of letters constructed using a keyword. The rules for filling in this 5x5 matrix are: L to R, top to bottom, first with keyword after duplicate letters have been removed, and then with the remain letters, with I/J used as a single letter. This example comes from Dorothy Sayer's book "Have His Carcase", in which Lord Peter Wimsey solves it, and describes the use of a probably word attack.
• #24: Plaintext is encrypted two letters at a time,according to the rules as shown. Note how you wrap from right side back to left, or from bottom back to top. if a pair is a repeated letter, insert a filler like 'X', eg. "balloon" encrypts as "ba lx lo on" if both letters fall in the same row, replace each with letter to right (wrapping back to start from end), eg. “ar" encrypts as "RM" if both letters fall in the same column, replace each with the letter below it (again wrapping to top from bottom), eg. “mu" encrypts to "CM" otherwise each letter is replaced by the one in its row in the column of the other letter of the pair, eg. “hs" encrypts to "BP", and “ea" to "IM" or "JM" (as desired) Decrypting of course works exactly in reverse. Can see this by working the example pairs shown, backwards.
• #25: The Playfair cipher is a great advance over simple monoalphabetic ciphers, since there are 26*26=676 digrams (vs 26 letters), so that identification of individual digrams is more difficult. Also,the relative frequencies of individual letters exhibit a much greater range than that of digrams, making frequency analysis much more difficult. The Playfair cipher was for a long time considered unbreakable. It was used as the standard field system by the British Army in World War I and still enjoyed considerable use by the U.S.Army and other Allied forces during World War II. Despite this level of confidence in its security,the Playfair cipher is relatively easy to break because it still leaves much of the structure of the plaintext language intact
• #26: One approach to reducing the "spikyness" of natural language text is used the Playfair cipher which encrypts more than one letter at once. We now consider the other alternative, using multiple cipher alphabets in turn. This gives the attacker more work, since many alphabets need to be guessed, and because the frequency distribution is more complex, since the same plaintext letter could be replaced by several ciphertext letters, depending on which alphabet is used. The general name for this approach is a polyalphabetic substitution cipher. All these techniques have the following features in common: A set of related monoalphabetic substitution rules is used. 2. A key determines which particular rule is chosen for a given transformation.
• #27: The best known, and one of the simplest, such algorithms is referred to as the Vigenère cipher, where the set of related monoalphabetic substitution rules consists of the 26 Caesar ciphers, with shifts of 0 through 25. Each cipher is denoted by a key letter, which is the ciphertext letter that substitutes for the plaintext letter ‘a’, and which are each used in turn, as shown next.
• #28: Discuss this simple example from text Stallings section 2.2.
• #29: Implementing polyalphabetic ciphers by hand can be very tedious. Various aids were devised to assist the process. The "Saint-Cyr Slide" was popularised and named by Jean Kerckhoffs, who published a famous early text "La Cryptographie Militaire" (Miltary Cryptography) in 1883. He named the slide after the French National Military Academy where the methods were taught. He also noted that any slide can be expanded into a tableau, or bent round into a cipher disk. The Vigenère Tableau (given in the text as Stallings Table 2.3) is a complete set of forward shifted alphabet mappings.
• #30: The Vigenère & related polyalphabetic ciphers still do not completely obscure the underlying language characteristics. The key to breaking them was to identify the number of translation alphabets, and then attack each separately.
• #31: For some centuries the Vigenère cipher was le chiffre indéchiffrable (the unbreakable cipher). As a result of a challenge, it was broken by Charles Babbage (the inventor of the computer) in 1854 but kept secret (possibly because of the Crimean War - not the first time governments have kept advances to themselves!). The method was independently reinvented by a Prussian, Friedrich Kasiski, who published the attack now named after him in 1863. However lack of major advances meant that various polyalphabetic substitution ciphers were used into the 20C. One very famous incident was the breaking of the Zimmermann telegram in WW1 which resulted in the USA entering the war. In general the approach is to find a number of duplicated sequences, collect all their distances apart, look for common factors, remembering that some will be random flukes and need to be discarded. Now have a series of monoalphabetic ciphers, each with original language letter frequency characteristics. Can attack these in turn to break the cipher.
• #32: Taking the polyalphabetic idea to the extreme, want as many different translation alphabets as letters in the message being sent. One way of doing this with a smallish key, is to use the Autokey cipher. The example uses the keyword "DECEPTIVE" prefixed to as much of the message "WEAREDISCOVEREDSAV" as is needed. When deciphering, recover the first 9 letters using the keyword "DECEPTIVE". Then instead of repeating the keyword, start using the recovered letters from the message "WEAREDISC". As recover more letters, have more of key to recover later letters. Problem is that the same language characteristics are used by the key as the message. ie. a key of 'E' will be used more often than a 'T' etc hence an 'E' encrypted with a key of 'E' occurs with probability (0.1275)2 = 0.01663, about twice as often as a 'T' encrypted with a key of 'T' have to use a larger frequency table, but it exists given sufficient ciphertext this can be broken.
• #33: The One-Time Pad is an evolution of the Vernham cipher, which was invented by Gilbert Vernham in 1918, and used a long tape of random letters to encrypt the message. An Army Signal Corp officer, Joseph Mauborgne, proposed an improvement using a random key that was truly as long as the message, with no repetitions, which thus totally obscures the original message. It produces random output that bears no statistical relationship to the plaintext. Because the ciphertext contains no information whatsoever about the plaintext, there is simply no way to break the code, since any plaintext can be mapped to any ciphertext given some key. The one-time pad offers complete security but, in practice, has two fundamental difficulties: There is the practical problem of making large quantities of random keys. 2. And the problem of key distribution and protection, where for every message to be sent, a key of equal length is needed by both sender and receiver. Because of these difficulties, the one-time pad is of limited utility, and is useful primarily for low-bandwidth channels requiring very high security.
• #34: All the techniques examined so far involve the substitution of a ciphertext symbol for a plaintext symbol. A very different kind of mapping is achieved by performing some sort of permutation on the plaintext letters. This technique is referred to as a transposition cipher, and form the second basic building block of ciphers. The core idea is to rearrange the order of basic units (letters/bytes/bits) without altering their actual values.
• #35: The simplest such cipher is the rail fence technique, in which the plaintext is written down as a sequence of diagonals and then read off as a sequence of rows. The example message is: "meet me after the toga party" with a rail fence of depth 2. This sort of thing would be trivial to cryptanalyze.
• #36: A more complex transposition cipher is to write the message in a rectangle, row by row, and read the message off shuffling the order of the columns in each row. A pure transposition cipher is easily recognized because it has the same letter frequencies as the original plaintext. For the type of columnar transposition just shown, cryptanalysis is fairly straightforward and involves laying out the ciphertext in a matrix and playing around with column positions. Digram and trigram frequency tables can be useful.
• #37: Have seen that ciphers based on just substitutions or transpositions are not secure, and can be attacked because they do not sufficient obscure the underlying language structure So consider using several ciphers in succession to make harder. A substitution followed by a transposition is known as a Product Cipher, and makes a new much more secure cipher, and forms the bridge to modern ciphers.
• #38: The next major advance in ciphers required use of mechanical cipher machines which enabled to use of complex varying substitutions. A rotor machine consists of a set of independently rotating cylinders through which electrical pulses can flow. Each cylinder has 26 input pins and 26 output pins, with internal wiring that connects each input pin to a unique output pin. If we associate each input and output pin with a letter of the alphabet, then a single cylinder defines a monoalphabetic substitution. After each input key is depressed, the cylinder rotates one position, so that the internal connections are shifted accordingly. The power of the rotor machine is in the use of multiple cylinders, in which the output pins of one cylinder are connected to the input pins of the next, and with the cylinders rotating like an “odometer”, leading to a very large number of substitution alphabets being used, eg with 3 cylinders have 263=17576 alphabets used. They were extensively used in world war 2, and the history of their use and analysis is one of the great stories from WW2.
• #39: This photo of an Allied Hagelin machine was taken by Lawrie Brown at Eurocrypt'93 in Norway. Note pen for scale, and the rotating cipher wheels near the front.
• #40: Steganography is an alternative to encryption which hides the very existence of a message by some means. There are a large range of techniques for doing this. Steganography has a number of drawbacks when compared to encryption. It requires a lot of overhead to hide a relatively few bits of information. Also, once the system is discovered, it becomes virtually worthless, although a message can be first encrypted and then hidden using steganography.
• #41: Chapter 2 summary.