BCH Error Correction Circuit with Adaptive Search Module Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Bose-Chaudhuri-Hochquenghem (BCH) data correction circuits face bottlenecks in data throughput due to the long propagation delay of the search module, especially when data bit errors occur simultaneously in multiple data channels, leading to reduced performance and increased size and power consumption when multiple independent data correction circuits are used to alleviate this issue.
Innovation Solution
A data correction system with multiple search modules, each having a respective bit error capacity, an arbiter to select the appropriate search module based on the number of data bit errors, and a correction module to correct errors indicated by the selected search module, allowing simultaneous error location and correction across multiple data units.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiple independent data correction circuits are used to increase throughput when data bit errors occur simultaneously in multiple data channels, then data throughput is improved, but device size and power consumption increase
Solution Approach 1:
The patent implements a single data correction circuit that can handle multiple data channels by incorporating an arbiter that dynamically assigns data units from different channels to shared correction resources (search module, correction module). This allows one circuit to perform the function of multiple circuits through time-multiplexed operation, improving throughput without proportionally increasing device size.
Solution Approach 2:
The arbiter dynamically selects which data channel receives correction service based on current error conditions and circuit availability. This dynamic allocation allows the system to adapt to varying error patterns across multiple channels, maintaining high throughput while utilizing fixed physical resources efficiently through time-varying assignment rather than static dedicated paths.
2Productivity
If the search module processes data units with high bit error capacity, then more errors can be corrected, but propagation delay increases and becomes a bottleneck
Solution Approach 1:
The patent segments the data correction function into distinct modular components: syndrome generator, locator polynomial generator, search module, and correction module. This segmentation allows each module to be optimized independently and enables pipelining where different data units can be processed at different stages simultaneously, reducing the effective propagation delay for high-error-capacity data units.
Solution Approach 2:
The syndrome generator and locator polynomial generator operate in advance to prepare correction information before the search module needs to locate errors. This preliminary computation of syndromes and locator polynomials allows the search module to focus solely on error location without waiting for syndrome computation, reducing overall propagation delay while maintaining the ability to correct multiple bit errors.
Data Source
AI summary
In various embodiments, a data correction system has a data path including search modules. Each of the search modules has a respective bit error capacity for locating a number of data bit errors in a data unit based on a locator polynomial. The data correction system generates a syndrome based on an input data unit, generates a locator polynomial based on the syndrome, and determines a number of data bit errors in the input data unit based on the locator polynomial. Additionally, the data correction system selects one of the search modules having a bit error capacity of at least the number of data bit errors in the input data unit. The selected search module generates an error indicator based on the locator polynomial. The data correction system corrects each data bit error in the input data unit based on the error indicator.