2025/08/12 by Vatsala Upadhyay, Upadhyay, Vatsala, J. Kokila +3
Computer Science · #Advanced Data Compression Techniques #Algorithms and Data Compression #FOS: Computer and information sciences #Information Theory (cs.IT) #Video Coding and Compression Technologies
paper · pdf · doi:10.48550/arxiv.2508.08840
openalex publication_date 2025/08/12 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
The Internet of Things (IoT) generates vast amounts of heterogeneous data, ranging from sensor readings to log alerts and images, that pose challenges to storage and data transmission in resource-constrained environments. In this context, lossless data compression techniques, like Arithmetic Coding, offer an effective solution owing to their high compression ratio. However, the standard Arithmetic Coding technique is computationally intensive, leading to high memory and processing overhead. This paper proposes an optimized version of Arithmetic coding for the IoT environment that incorporates three improvements using Iterative and Iteration Optimizations for minimizing redundant computations and achieving faster convergence; Principal Component Analysis(PCA) for dimensionality reduction and identifying key features; and lastly, Cardinality reduction for grouping similar probabilities to improve the compression efficiency. The proposed method was evaluated on a dataset of images and demonstrated significant reductions in the time to compress, CPU utilization, and memory consumption, and preserves data integrity as seen through the low RMSE values. The optimized version of the Arithmetic Coding algorithm achieves an impressive compression ratio of 814:1 and 101 ms to compress a single image. This makes the optimized algorithm suitable for real-time applications and resource-constrained environments for efficient data transmission and storage.