Equalization Simulator Training Sequence Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
High-speed serial data link systems face performance degradation due to channel loss, reflection, cross-talk, and noise, with existing equalization techniques failing to account for uncorrelated signal components like periodic and random jitter/noise, leading to suboptimal equalizer tap optimization and potential bit sequence misclassification.
Innovation Solution
An equalization simulator with training sequence detection for oscilloscopes, which generates a digitized waveform record, recovers a clock signal, re-samples the waveform, and aligns it with a training sequence to generate equalizer taps, effectively removing channel effects and optimizing equalization in the presence of severe Inter-Symbol Interference and noise.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If blind equalizer adaptation approach is used with sampled analog waveforms, then the approach can handle severe ISI and noise, but the equalizer taps cannot be optimized correctly and bit sequence classification fails
Solution Approach 1:
The patent applies preliminary action by detecting and removing channel effects from the digitized waveform record before performing equalizer adaptation. The input signal processor de-embeds channel characteristics using S- or T-parameters or a detection equalizer, providing a cleaned signal to the sequence detector. This preliminary processing enables reliable training sequence detection even in severe channel conditions, which then allows correct equalizer tap optimization.
2Measurement precision
If equalization adaptation algorithm uses averaged sampled analog data pattern waveform, then the algorithm can optimize equalizer taps, but signal components not correlated to data pattern (periodic and random jitter/noise) are not accounted for
Solution Approach 1:
The patent introduces an intermediary approach by using the real-time oscilloscope's full-bandwidth captured waveform as an intermediary representation that preserves all frequency-dependent information including uncorrelated components. Instead of directly using averaged sampled data, the system processes the complete waveform through de-embedding and sequence detection, allowing the equalizer adaptation to account for periodic and random jitter/noise while still optimizing taps correctly.
3Measurement precision
If real-time oscilloscope captures signals without aliasing, then all frequency dependent information is preserved for equalizer optimization, but the system complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the complex signal processing task into distinct functional blocks: an input signal processor for de-embedding channel effects, a sequence detector for training sequence identification, and an equalization adapter for tap optimization. This segmentation of the processing chain makes the complex real-time waveform analysis manageable and implementable while preserving all frequency-dependent information from the oscilloscope capture.
Data Source
Figure 1~3
Figure 4
Figure 5
AI summary
An equalization simulator with training sequence detection uses an oscilloscope to acquire digital samples of an analog waveform signal from a serial data link to produce a digitized waveform record. The equalization simulator has training sequence detection receiving the digitized waveform record and generates a training sequence, an equalization adapter receiving the digitized waveform record and the training sequence and generating equalizer taps, and an equalizer receiving the equalizer taps and the digitized waveform record and generating an equalized digitized waveform record.