Multi-Level Comma Detection for Data Stream Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data communication systems face challenges in efficiently aligning serial data streams, particularly under standards like SONET, due to the need for extensive circuitry and complex comparisons to locate break points between data packets, which increases logic requirements and can lead to false realignments.
Innovation Solution
A multi-level comma detection and alignment method that reduces logic requirements by first detecting and aligning with a predetermined byte value, then shifting to locate a predetermined sequence, thereby reducing the number of comparisons needed and improving alignment accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If extensive circuitry and complex comparisons are used to locate break points in data streams, then alignment accuracy can be improved, but device complexity and logic requirements increase
Solution Approach 1:
The patent divides the alignment process into two distinct stages: first detecting a comma character (0x19 byte value) to achieve partial alignment, then performing a second-level comparison of 4-byte sequences to achieve final precise alignment. This segmentation reduces the complexity of each individual comparison stage while maintaining overall alignment accuracy.
Solution Approach 2:
The patent performs preliminary alignment by detecting the comma character (0x19) before performing the final alignment. This preliminary action narrows down the possible locations of break points, reducing the search space for the subsequent precise alignment stage and thereby reducing the complexity of the circuit required for the final comparison.
2Measurement precision
If extensive comparisons are performed to locate break points, then alignment accuracy is improved, but the number of logic operations increases
Solution Approach 1:
The comparison process is segmented into two phases: first comparing individual bytes to find the comma character (0x19), then comparing 4-byte sequences for precise alignment. This segmentation reduces the total number of logic operations required compared to performing a single exhaustive comparison of all possible break point locations.
Solution Approach 2:
The patent performs preliminary byte-level comparison to locate the comma character before performing the more complex 4-byte sequence comparison. This preliminary action eliminates many potential false alignment locations, reducing the number of logic operations needed in the final alignment stage.
3Device complexity
If single-level comma detection is used, then device complexity is reduced, but false realignments occur
Solution Approach 1:
The patent implements a two-level detection system: the first level detects the comma character (0x19) to identify potential alignment points, and the second level verifies these points by comparing 4-byte sequences. This segmentation eliminates false realignments that would occur with single-level detection while keeping each detection stage relatively simple.
Solution Approach 2:
The comma character (0x19) serves as an intermediary marker that guides the alignment process. By first locating this known marker and then performing verification comparisons, the system reliably identifies true alignment points without the complexity of direct full-sequence matching from scratch.
Data Source
AI summary
Methods and structures of performing multi-level comma detection and alignment on an unaligned data stream. Each string of N consecutive bits in the unaligned data stream is monitored for a predetermined byte value. When the predetermined byte value is located, the unaligned data stream is aligned with the predetermined byte value, producing a partially aligned data stream. A string of bytes from the partially aligned data stream is then compared with a predetermined sequence of byte values. When the predetermined sequence is located, the partially aligned data stream is aligned based on the location of the predetermined sequence within the partially aligned data stream. The invention also encompasses multi-level comma detection and alignment circuits that can perform, for example, the previously described inventive methods.