Design and Analysis of Two-Layer Coding Scheme for DNA-Based Data Storage

DNA-based data storage has emerged as a compelling alternative to traditional media due to its ultra-high information density and long-term stability. However, the high read cost caused by the error-prone synthesis, storage, and sequencing processes remains a major bottleneck for practical deployment. To address this challenge, this paper proposes a read-cost-efficient coding framework that enhances reliability without increasing total redundancy. First, a novel two-layer intra-oligo coding scheme based on Bose-Chaudhuri-Hocquenghem (BCH) codes is presented, where index bits and data bits are respectively protected to mitigate base-level errors. Second, a semi-analytical optimization method based on the normal approximation of the finite-length coding rate is developed to allocate redundancy between index and data bits optimally under a fixed total code rate. The inter-oligo protection is further achieved through low-density parity-check (LDPC) codes to combat sequence-level errors. We then present extensive analytical and numerical results to show the effectiveness of the proposed analysis. Finally, we present numerical results to demonstrate that the concatenated code based on the optimized two-layer coding scheme significantly outperforms the concatenated code based on the single-layer coding scheme in terms of frame error rate (FER) under the same sequencing depth and total redundancy. These results underscore the advantages of the two-layer coding scheme and the optimization method for DNA-based data storage systems.