Multi-Dimensional Decoding for High-Density Memory Error Correction

The patent divides the encoded data stream into multiple segments and processes each segment independently through parallel decoder components. This segmentation allows the system to handle high error rates in dense storage by treating errors locally rather than requiring complex global correction, thus improving reliability without proportionally increasing overall system complexity.

The patent introduces a temporal dimension to error correction by using multiple decode attempts with varying threshold values. Instead of relying solely on spatial redundancy, the system performs iterative decoding across time, refining error correction by adjusting decision thresholds between attempts. This dimensional approach enables effective error correction at high storage densities without linearly increasing complexity.

The patent performs preliminary decoding attempts with relaxed threshold values before finalizing error corrections. By conducting initial decode attempts that are computationally less intensive and progressively refining results in subsequent attempts, the system achieves high error correction accuracy while minimizing total processing time. This preliminary action approach allows the most time-consuming operations to be performed only when necessary.

The patent dynamically adjusts threshold values between decode attempts based on error patterns observed in previous attempts. Rather than using fixed thresholds, the system adapts its decoding parameters in real-time, increasing threshold stringency only when errors are detected. This dynamic approach optimizes the balance between processing speed and correction accuracy by applying computational effort selectively.