Digital Clock and Data Recovery Phase Search Block
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In automotive environments with high noise and interference, existing Digital Clock and Data Recovery (DCDR) systems face challenges in quickly identifying the best sampling phase for signals due to the shortness of the 10SPE frame preamble, which limits their ability to perform reliable data recovery and pass stringent tests like BCI and DPI.
Innovation Solution
The implementation of a phase search block that oversamples received signals using quadrature clocks to identify the boundaries of bits and determine the best initial phase within a few symbols, replacing conventional training stages, allowing DCDR to lock by the start of the preamble and decode the complete frame.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional training stages are used for DCDR, then reliable data recovery can be achieved, but the system cannot lock by the start of the preamble due to the shortness of the 10SPE frame preamble
Solution Approach 1:
The phase search block performs preliminary phase identification by oversampling the preamble at multiple phases and identifying edge samples before formal data recovery begins. This preliminary action allows the system to establish the correct sampling phase in advance, enabling DCDR to lock by the start of the preamble rather than after conventional training stages.
Solution Approach 2:
The preamble is segmented into multiple portions that are oversampled at different phases. The phase search block processes these segmented portions independently to identify edge samples at each phase, then combines this information to determine the optimal sampling phase. This segmentation allows efficient use of the limited preamble duration.
2Measurement precision
If the system oversamples at multiple phases to identify the best sampling phase, then accurate phase detection is achieved, but the processing complexity increases
Solution Approach 1:
The phase search block extracts only the critical edge sample information from the oversampled preamble portions, rather than processing all sample data. By identifying and extracting only the edge samples at each phase, the system achieves accurate phase detection while minimizing processing complexity. This extraction approach focuses computational resources on the most relevant information.
3Measurement precision
If the system uses a longer training stage for phase identification, then more accurate phase detection is achieved, but the 10SPE frame preamble is too short to accommodate it
Solution Approach 1:
The phase search block employs periodic oversampling of the preamble at multiple discrete phases (e.g., four different phases). This periodic sampling approach allows the system to gather sufficient phase information within the limited preamble duration by systematically sampling at regular phase intervals, achieving accurate phase identification without requiring an extended training stage.
Data Source
AI summary
Systems, devices, and methods related to selecting a sample phase of a signal are disclosed. A method includes sampling a signal including a plurality of symbols with a plurality of different sample phases to obtain sample values of each of the plurality of symbols at each of the plurality of different sample phases. The signal is received from a shared transmission medium. The method also includes determining an edge sample phase of the plurality of different sample phases that corresponds to edges of the symbols based on the sample values. The method further includes determining a center sample phase of the plurality of different sample phases based on the determined edge sample phase, and using the determined center sample phase to determine values of the symbols.