Clustering-Based Dual-Population Co-Evolutionary Algorithm for 2D Segmentation Coding in NAND Flash Memory

In NOT-AND (NAND) flash memory-based solid-state drive (SSD), the error checking and correction (ECC) module is configured to ensure the stability of stored data. With the increase of raw bit error rate (RBER), the traditional error correction coding mode is not only difficult to meet the requirements, but also the efficiency of data repair is extremely low because the error bits cannot be located, when the error correction cannot be completed. Therefore, an innovative error correction coding mode, called two-dimensional (2D) segmentation coding mode, is proposed.To our knowledge, this is the first attempt to apply this coding mode to NAND flash memory. It improves error correction capability and data recovery efficiency by coding information bits in rows and columns separately. We propose a dual-population co-evolution algorithm based on clustering algorithm to optimize the segmentation method for better efficiency. Under the condition of satisfying all the constraints, the solutions obtained by the proposed algorithm achieve the balance of rewrite efficiency, coding efficiency and coding time. Experimental results show that the proposed algorithm performs well compared with six state-of-the-art evolutionary algorithms (EAs) in solving the proposed constrained multi-objective optimization problem.