| Symbol | Freq | Probability | Huffman Bit Code | Bit Length |
|---|---|---|---|---|
| Run compression to generate codebook tree. | ||||
Academic Research & Engineering Rationale
Published in Jurnal Ilmiah Teknologi Informasi Terapan (JITTER), Vol. 5, Iss. 3 (2019). This research evaluated the real-world operational viability of combining prefix-free minimum redundancy codes (Huffman) with intermediate binary-to-hexadecimal byte-boundary packing across 50 distinct file types and 26 standard extensions.
1. Frequency-Sorted Min-Heap Tree
The compressor builds an optimal binary prefix tree where high-probability symbols gravitate toward the root (1-3 bits) and rare symbols receive deeper branch paths, breaking the static 8-bit ASCII barrier.
2. Nibble-Aligned Hexadecimal Packing
Continuous variable-length bitstreams are zero-padded to 4-bit boundaries and translated into clean hexadecimal byte sequences, eliminating binary serialization drift during distributed message transfer.
3. Cross-Format Empirical Results
The paper tested text, executables, archives, and multimedia files. Highly redundant structured data (.txt, .log, .json, .csv) achieved compression ratios between 35% and 58%, while pre-compressed formats (.zip, .mp4, .jpg) showed negligible gains due to maximum Shannon entropy.
4. Trade-off vs. Modern Sliding Dictionaries
While standalone Huffman encoding requires negligible memory and zero dictionary lookahead latency, multi-pass LZ77/Deflate algorithms (WinRAR, gzip) outperform it on large continuous texts with repetitive multicharacter substrings.
Erlando Aubrey
Author of peer-reviewed research in data compression & information systems. 8+ years building enterprise high-throughput services, fintech settlement ledgers, and deterministic engines.