Neural Network Decoder Selection for Error Correction Throughput
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Electronic devices face challenges in reliably selecting the optimal error correction decoding algorithm for data integrity, leading to suboptimal throughput and increased power consumption during data operations.
Innovation Solution
An electronic device is equipped with a decoding controller that inputs data on unsatisfied check nodes and correction bits to a trained artificial neural network, which selects between two error correction decoding algorithms based on its output, enabling optimal algorithm selection for error correction decoding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a fixed error correction decoding algorithm is used, then device complexity is reduced, but reliability of data correction deteriorates
Solution Approach 1:
The patent pre-trains multiple error correction decoding algorithms (including BP, Min-sum, and Sum-product algorithms) with different characteristics before runtime. The training phase prepares various decoding strategies in advance, allowing the system to select the most appropriate algorithm based on the specific error patterns encountered, rather than using a single fixed algorithm for all cases.
Solution Approach 2:
The patent implements a dynamic algorithm selection mechanism where the system adaptively chooses different error correction decoding algorithms based on the characteristics of the received signal and error patterns. This dynamic approach allows the decoder to switch between algorithms (e.g., from BP to Min-sum or Sum-product) depending on the situation, optimizing performance for varying channel conditions while managing complexity through intelligent selection rather than simultaneous implementation of all algorithms.
2Reliability
If multiple error correction decoding algorithms are always executed, then reliability improves, but productivity decreases
Solution Approach 1:
The system performs preliminary analysis of the received signal characteristics and error patterns before selecting a decoding algorithm. By pre-evaluating the situation and choosing the most suitable algorithm in advance, the system avoids unnecessarily executing multiple algorithms, thus maintaining high reliability when needed while preserving throughput by selecting the most efficient single algorithm for each specific case.
Solution Approach 2:
The patent implements dynamic algorithm selection that adapts to the specific error conditions. When errors are detected, the system dynamically chooses the most appropriate decoding algorithm from multiple trained options based on the error characteristics, rather than always executing all algorithms. This dynamic approach ensures high reliability for difficult-to-correct errors while maintaining productivity by using simpler, faster algorithms when appropriate.
3Reliability
If multiple error correction decoding algorithms are always executed, then reliability improves, but power consumption increases
Solution Approach 1:
The system performs preliminary assessment of the received signal and error patterns before initiating decoding operations. By pre-identifying the most suitable algorithm based on error characteristics, the system avoids the unnecessary power consumption of executing multiple algorithms, thereby maintaining reliability through appropriate algorithm selection while significantly reducing overall power consumption of the decoding process.
Solution Approach 2:
The patent implements a dynamic power-management approach where the decoder adapts its algorithm selection based on real-time error conditions. When errors require sophisticated correction, the system dynamically activates more complex algorithms; when errors are minimal or patterns are simple, it uses lighter-weight algorithms. This dynamic adaptation ensures reliability is maintained only when necessary while minimizing power consumption during normal operation.
Data Source
AI summary
Devices for using a neural network to choose an optimal error correction algorithm are disclosed. An example device includes a decoding controller inputting at least one of the number of primary unsatisfied check nodes (UCNs), the number of UCNs respectively corresponding to at least one iteration, and the number of correction bits respectively corresponding to the at least one iteration to a trained artificial neural network, and selecting any one of a first error correction decoding algorithm and a second error correction decoding algorithm based on an output of the trained artificial neural network corresponding to the input, and an error correction decoder performing error correction decoding on a read vector using the selected error correction decoding algorithm. The output of the trained artificial neural network may include a first predicted value indicating a possibility that a first error correction decoding using the first error correction decoding algorithm is successful.