International Journal of Computer Science and Technology

International Journal of Computer Science and Technology | Vol. 9, No. 9, September 2007 | pp. 65–72

DOI: 10.46882/2007/IJCST/099006

Research Article

Title: Performance Evaluation of Lossless Data Compression Algorithms for Embedded Systems

Names of Authors: Yukihiro Tanaka¹, Kenji Sato¹

Authors’ Affiliations: ¹Department of Electronic Engineering, Tokyo Institute of Technology, Tokyo, Japan

Abstract: Resource-constrained embedded systems often encounter storage and data transmission bottlenecks, making lightweight data compression algorithms highly essential. This paper conducts a rigorous performance evaluation of four prominent lossless data compression algorithms—Huffman coding, Lempel-Ziv-Welch (LZW), Deflate, and Run-Length Encoding (RLE)—tailored for 16-bit microcontroller microarchitectures. The algorithms were evaluated using diverse sensor-derived datasets, logging metrics across compression ratios, execution times, and dynamic RAM requirements. Our experimental benchmarks reveal that while the Deflate algorithm offers the highest average compression ratio of 2.45:1, it demands excessive memory footprints (exceeding 32 KB RAM) and high computational cycles. In contrast, an optimized Huffman coding implementation yielded a balanced compression ratio of 1.68:1 while utilizing merely 2.8 KB of RAM and running within 14.2 milliseconds per kilobyte of raw sensor data. Statistical t-tests confirmed that the optimized Huffman implementation achieves a significant reduction in power consumption during RF transmission, dropping energy costs by 34.5% ± 1.2% due to minimized packet payloads.

Keywords: Lossless data compression, embedded systems, Huffman coding, LZW algorithm, memory constraints, power optimization

Manuscript Timeline: Received June 11, 2007; Revised July 23, 2007; Accepted August 15, 2007; Published September 4, 2007

Citation: Tanaka, Y., & Sato, K. (2007). Performance Evaluation of Lossless Data Compression Algorithms for Embedded Systems. International Journal of Computer Science and Technology, 9(9), 65–72.

Subscribe to read the full article: https://www.internationalscholarsjournals.org/subscribe-to-read