Hawkynt

Shannon-Fano Coding

Variable-length prefix-free coding algorithm that predates Huffman coding. Divides symbols recursively by frequency to create binary codes, though not always optimal.

Properties

Property	Value
Category	Compression Algorithms
Sub-category	Statistical
Security status	🎓 Educational Only
Complexity	Intermediate
Inventor	Claude Shannon, Robert Fano
Year	1948
Origin	🇺🇸 United States
Source	algorithms/compression/shannon-fano.js

Security

Status: 🎓 Educational Only

No vulnerabilities are recorded for this implementation.

Documentation

• A Mathematical Theory of Communication
• Shannon-Fano Coding - Wikipedia
• Information Theory Primer
• Data Compression History

References

• MIT Information Theory Course
• Shannon-Fano vs Huffman Analysis
• Rosetta Code Implementation
• Educational Examples

Test vectors

3 vectors ship with this algorithm and run in the test suite. Byte values are hexadecimal.

Vector 1 — Basic frequency encoding

Field	Value
input	414141424243
expected	06000000000000000000000000000000 00000000000000000000000000000000 00000000000000000000000000000000 00000000000000000000000000000000 … (518 bytes; the full value is in the source)

Vector 2 — Alphabet frequency test

Field	Value
input	414243444546
expected	06000000000000000000000000000000 00000000000000000000000000000000 00000000000000000000000000000000 00000000000000000000000000000000 … (518 bytes; the full value is in the source)

Vector 3 — Repeated pattern encoding

Field	Value
input	414241424142
expected	06000000000000000000000000000000 00000000000000000000000000000000 00000000000000000000000000000000 00000000000000000000000000000000 … (517 bytes; the full value is in the source)

---

← All algorithms

This site is open source. Improve this page.