MATH3411 Information, Codes and Ciphers
The Complete Exam Bible · T3 2026MATH3411 Overview
- UNSW Sydney
- Term 3 2026
- Level 3 mathematics
- Open-book lab exam
MATH3411 Information, Codes and Ciphers is a Level 3 mathematics course at UNSW Sydney about the mathematical structure of digital communication. It begins with source and code alphabets, modular arithmetic, probability, ASCII and check digits, then studies how deliberate redundancy detects and corrects transmission errors.
- Assessed by Three tests and a 60% final exam
- Hardest step Choosing the theorem before calculating
- How to prepare Repeat algorithms with fresh values
How MATH3411 is assessed
| Component | Weight | Format |
|---|---|---|
| Test 1 | 10% | Individual online test, 40 minutes, held in a computing lab in Week 4 on 6 October 2026 |
| Test 2 | 15% | Individual online test, 40 minutes, held in a computing lab in Week 7 on 27 October 2026 |
| Test 3 | 15% | Individual online test, 40 minutes, held in a computing lab in Week 10 on 17 November 2026 |
| Final exam | 60% | Individual online exam in a computing lab, 2 hours, open book using supplied Moodle materials; no exam hurdle |
Current MATH3411 dates
| Date | Item | Control |
|---|---|---|
| 6 Oct 2026 | Test 1 | Term 3 2026 Moodle syllabus |
| 27 Oct 2026 | Test 2 | Term 3 2026 Moodle syllabus |
| 17 Nov 2026 | Test 3 | Term 3 2026 Moodle syllabus |
Dates are as published in the current course materials. Confirm exact deadlines and submission settings in the live LMS.
What MATH3411 covers
The course follows a digital message from its mathematical model through error protection and compression to information limits and cryptography. Seven source-derived chapters connect modular arithmetic, linear codes, Huffman trees, entropy, finite fields, BCH methods and RSA.
Mathematical Models and Check Digits
Source alphabets, modular arithmetic, probability and ISBN check equationsThis chapter establishes the mathematical objects used throughout the course: source and code alphabets, binary words, modular arithmetic and conditional probability. It then applies congruences to ASCII, parity and ISBN-style check equations.
- Source alphabets and code alphabets
- Binary words and radix-r notation
- Modular arithmetic and inverses
- Conditional probability in noisy channels
- ASCII parity and ISBN checks
- Single-error and transposition detection
Error Detection Through Linear Codes
Hamming distance, sphere packing, matrices and syndrome decodingThis chapter develops the geometry and algebra of error control. Hamming distance and weight quantify separation, sphere packing limits code size, and perfect codes attain the volume bound.
- Hamming distance and weight
- Detection and correction bounds
- Sphere-packing condition
- Perfect codes
- Parity-check and generator matrices
- Syndrome correction
Compression Using Huffman Trees
Unique decoding, Kraft-McMillan, Huffman trees and Markov sourcesCompression assigns shorter codewords to more likely symbols while preserving unique recovery.
- Unique decodability
- Instantaneous prefix codes
- Kraft-McMillan inequality
- Binary Huffman algorithm
- Radix-r Huffman coding
- Source extensions
Entropy, Information and Channel Capacity
Information measures, source bounds, noisy channels and capacityInformation theory measures uncertainty independently of message meaning. Entropy averages surprise, source-coding results bound compression, and joint, conditional and mutual information track how variables share uncertainty.
- Self-information and entropy
- Maximum entropy
- Source-coding bounds
- Joint and conditional entropy
- Mutual information
- Binary symmetric channels
Number Theory with Finite Fields
Euclid, modular inverses, polynomial fields, primality and factoringThis chapter supplies the algebra behind later codes and ciphers. Euclid and Bezout identities produce greatest common divisors and modular inverses.
- Divisibility and Euclid
- Bezout identities
- Modular inverses
- Polynomial arithmetic over prime fields
- Irreducible and primitive polynomials
- Finite-field construction
BCH Methods for Algebraic Codes
Field powers, code roots, syndromes and error-location methodsBCH codes use roots in a finite extension field to turn decoding into algebra. Evaluating a received polynomial at the designed roots cancels the unknown codeword and leaves syndromes determined by the error pattern.
- Hamming codes revisited
- BCH designed roots
- Field-valued syndromes
- Single-error location
- Error magnitudes
- Double-error equations
Classical Ciphers to Public-Key Cryptography
Frequency analysis, Kasiski's method, trapdoors and RSAThe cryptography chapter moves from classical substitution and transposition systems to modern trapdoor constructions. Frequency analysis attacks monoalphabetic structure, while repeated spacings and index of coincidence expose polyalphabetic periods.
- Classical substitution systems
- Frequency analysis
- Kasiski's method
- Index of coincidence
- One-way and trapdoor functions
- RSA key construction
The coding chapters develop parity checks, Hamming distance, sphere-packing bounds, perfect codes, linear codes, parity-check and generator matrices, and algebraic decoding. Compression reverses the design goal: uniquely decodable and instantaneous codes, Kraft-McMillan, Huffman algorithms and source extensions remove predictable redundancy.
Information theory supplies the limit through entropy, conditional entropy, mutual information and channel capacity. The number-theory chapter builds Euclidean algorithms, modular inverses, polynomial arithmetic and finite fields before BCH codes use field roots and syndromes to locate errors. The final chapter connects classical frequency attacks and Kasiski's method to public-key cryptography and RSA.
Assessment is cumulative in the 60% final, so successful revision means more than remembering definitions. You need to write a valid mathematical model, execute an algorithm line by line, and justify the conclusion with the relevant distance, coding, entropy, field or number-theory result. The course supplies practice tests, tutorial problems, old paper exams and solutions.
Use those resources to learn question shape, but solve with new values so the method remains transferable. A short verification line after each algorithm is often the quickest protection against a lost sign, residue or coordinate.
Worked example · free
Huffman coding with a quick optimality check
- +1Merge the two 1/8 probabilities to make 1/4, then merge the two 1/4 nodes to make 1/2, and finally join the two 1/2 nodes.
- +1One valid prefix assignment has lengths 1, 2, 3 and 3. Literal left and right branch labels may be swapped.
- +1The expected length is 1/2 + 2(1/4) + 3(1/8) + 3(1/8) = 1.75 bits per symbol.
- +1The entropy is also 1.75 bits because every probability is an exact negative power of two, so the code reaches the source bound.
Key terms
- Hamming distance
- The number of coordinates in which two words differ.
- Minimum distance
- The smallest Hamming distance between distinct codewords in a code.
- Syndrome
- The parity-check result of a received word, used to identify its error coset.
- Perfect code
- A code whose correction balls partition the entire ambient word space.
- Instantaneous code
- A code in which no codeword is a prefix of another codeword.
- Kraft sum
- The sum of radix raised to the negative codeword lengths, used to test length feasibility.
- Huffman code
- An optimal prefix code built by repeatedly merging the least likely nodes.
- Entropy
- The average information or uncertainty of a probability distribution.
- Mutual information
- The reduction in uncertainty about one variable gained by observing another.
- Finite field
- A finite algebraic system in which every nonzero element has a multiplicative inverse.
- BCH code
- An algebraic code designed through consecutive roots in a finite extension field.
- Trapdoor function
- A function that is easy to compute and hard to invert without secret information.
MATH3411 FAQ
How hard is this course?
The difficulty comes from switching between proof, algorithm and modular calculation rather than from one long formula list. Errors often arise when a method is applied in the wrong algebraic structure. Regular practice with fresh values makes the workload much more manageable.
How is the course assessed?
Three online lab tests contribute 10%, 15% and 15%. The individual online final exam contributes 60%. The published weights total 100%, and the course information explicitly says there is no exam hurdle.
What is the final exam format?
The final is a two-hour online exam in a computing lab and is expected to contain about seventeen questions, some with sub-questions. It mixes basic concepts, deeper problems, proofs and technical methods, with randomised questions.
Are past papers available?
Yes. Moodle lists paper exams and solutions from several prior offerings, including a Term 3 paper and solution pair. Use them to learn question types and timing, while following the current examinable-content notice and current format.
Is the final open book?
Yes, within strict limits. The lab computer provides the course notes, lecture notes, tutorial problems and solutions through Moodle. Personal notes, printouts, handheld calculators, pre-written code, general internet access and AI are not permitted.
What is the best revision method?
Master the current practice-test questions first, then solve tutorial problems and selected old-paper questions under time. Keep an error ledger for definitions, theorem choice, modular arithmetic and missing verification, and repair the first incorrect step rather than rereading everything.
Questions students ask in each MATH3411 chapter
- Why does the model matter before a calculation?
- It determines whether symbols are integers, vectors, field elements or probabilities, and therefore which operations and cancellation rules are valid. Declaring that structure first prevents familiar notation from licensing the wrong algebra. Chapter 01 · Mathematical Models and Check Digits
- How many errors can distance d correct?
- It corrects every pattern of up to the largest t satisfying 2t+1 at most d, while detection extends to d-1 errors. These are guaranteed worst-case limits rather than predictions about every larger pattern. Chapter 02 · Error Detection Through Linear Codes
- Does a failed prefix test prove ambiguity?
- No. It proves the code is not instantaneous, but a delayed decoder may still recover every concatenation uniquely. To prove ambiguity, exhibit one finite encoded string with two distinct codeword parsings. Chapter 03 · Compression Using Huffman Trees
- Why is a fair binary source most uncertain?
- Its two outcomes are equally plausible, so observing either removes the largest possible average uncertainty for a binary choice. Moving probability toward either endpoint makes one outcome easier to predict and lowers entropy. Chapter 04 · Entropy, Information and Channel Capacity
- When may I cancel in modular arithmetic?
- You may cancel a factor when it is invertible modulo the modulus, equivalently when its gcd with the modulus is one. Otherwise cancellation can discard valid residue classes or create a false conclusion. Chapter 05 · Number Theory with Finite Fields
- Why do codeword terms disappear in a syndrome?
- Every code polynomial vanishes at the designed roots, so evaluating received word equals evaluating error polynomial there. Linearity removes the unknown transmitted codeword and leaves equations in the sparse error parameters. Chapter 06 · BCH Methods for Algebraic Codes
- How does Kasiski analysis suggest a key length?
- Repeated ciphertext fragments create spacings whose common factors are candidates for the repeated-key period. Test those candidates with coincidence statistics because accidental repetitions can produce misleading spacings. Chapter 07 · Classical Ciphers to Public-Key Cryptography
How to study for the exam
Build revision around the communication pipeline. First make modular arithmetic, conditional probability and check equations automatic. Next connect Hamming distance, sphere packing and parity-check matrices so you can translate between geometry and algebra. Practise Huffman coding with a merge ledger and compare expected length with entropy. For information theory, keep logarithm bases and units explicit.
Then alternate integer Euclid with polynomial reduction until inverses and field arithmetic feel routine. In BCH problems, annotate coefficient arithmetic, exponent arithmetic and printed coordinate conventions separately. Finish with classical cipher diagnosis and small RSA key builds. The current exam is open book only through supplied Moodle materials, so practise navigating by theorem name instead of searching aimlessly.
Use the randomised practice tests for speed, tutorial solutions for method checks, and old papers for mixed-problem timing. End every calculation with a structural verification: zero residue, zero syndrome, valid Kraft sum, entropy bound or RSA round trip.
Your AI Mathematics tutor for MATH3411
Stuck on a hard MATH3411 question? Sia is AskSia’s AI Mathematics tutor — ask any MATH3411 Information, Codes and Ciphers question and get a clear, step-by-step explanation grounded in how the course is actually taught and assessed. Read this whole study guide free, then take your hardest questions to Sia.
