DECODING ERRORS IN DIFFERENCE-INVERTIBLE BLOOM FILTERS: ANALYSIS AND RESOLUTION

Abstract

Difference-Invertible Bloom Filters (DIBFs) are widely used in networking, distributed storage, and set reconciliation due to their compact representation and efficient difference computation. However, decoding errors caused by hash collisions and element overlaps limit their reliability and practical deployment in high-assurance applications. This paper presents a systematic framework to enhance the reliability of DIBFs by detecting, analyzing, and resolving decoding errors. The proposed Reliable Difference-Invertible Bloom Filter (R-DIBF) incorporates adaptive hash selection, error detection layers, iterative decoding algorithms, and controlled redundancy to improve error resilience while maintaining memory efficiency. Experimental evaluation demonstrates that R-DIBF significantly reduces decoding failures compared to standard DIBFs across varying set sizes and element distributions, providing a practical and scalable solution for applications requiring accurate set reconciliation and robust distributed data management.