Cryptography - Symmetric Encryption Ciphers

Editor's Notes

• #3: 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 5 ingredients of the symmetric cipher model: plaintext encryption algorithm – performs substitutions/transformations on plaintext secret key – control exact substitutions/transformations used in encryption algorithm ciphertext decryption algorithm – inverse of encryption algorithm
• #6: Generally assume that the algorithm is known. This allows easy distribution of s/w and h/w implementations. Hence assume just keeping key secret is sufficient to secure encrypted messages. Have plaintext X, ciphertext Y, key K, encryption alg Ek, decryption alg Dk.
• #10: Unconditional security would be nice, but the only known such cipher is the one-time pad (later). For all reasonable encryption algorithms, have to assume computational security where it either takes too long, or is too expensive, to bother breaking the cipher.
• #11: 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 techniques: substitution and transposition. We examine these in the next two sections. Finally, we discuss a system that combine both substitution and transposition.
• #12: 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. Note: when letters are involved, the following conventions are used in this course: Plaintext is always in lowercase; ciphertext is in uppercase; key values are in italicized lowercase.
• #13: This mathematical description uses modulo arithmetic (ie 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 (5) is MTBID
• #14: 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! cf. 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).
• #17: 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.
• #18: This graph is based on counts done at ADFA in the late 1980's, and used to develop the tables published in Seberry & Pieprzyk [SEBE89]. Note that all human languages have varying letter frequencies, though the number of letters and their frequencies varies. Seberry & Pieprzyk [SEBE89] Appendix A has graphs for 20 languages (most European & Japanese & Malay).
• #19: 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.
• #21: 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. Playfair cipher is an example of doing this.
• #22: Have here the rules for filling in the 5x5 matrix, 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 this, and describes the use of a probably word attack.
• #23: Note the various rules, and how you wrap from right side back to left, or from bottom back to top. Decrypting of course works exactly in reverse. Can see this by working the example pairs shown, backwards.
• #25: 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.
• #26: Simply create a set of caesar cipher translation alphabets, then use each in turn, as shown next.
• #30: See that the key used is 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.
• #31: 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. Since any plaintext can be mapped to any ciphertext given some key, there is simply no way to determine which plaintext corresponds to a specific instance of ciphertext.
• #33: Transposition Ciphers 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.
• #34: Example message is: "meet me after the toga party" with a rail fence of depth 2.