.png){kind=icon}
.png){kind=icon}
Return
A Robust Sliding Window-Based Edge Detection Algorithm for Multilevel PAM Clock Recovery
.png){kind=icon}
PRE
AI
DOI：10.1109/LCOMM.2026.3677063.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
As data rates continue to increase in high-speed serial communication systems, signal integrity (SI) issues are of growing concern. Jitter and eye diagram analysis serve as two fundamental means of evaluating SI and system performance, both relying on accurate reference clock extraction via clock recovery (CR). In software-based CR implemented in oscilloscopes, accurate edge detection is a critical prerequisite, since the recovered clock depends directly on correct transition identification. However, reduced signal-to-noise ratio (SNR) and SI impairments such as glitches can cause conventional threshold-based edge detection methods to misidentify transitions, leading to CR failure and invalid SI analysis. The adoption of multilevel pulse-amplitude modulation (PAM) further complicates edge detection by introducing multiple transition types. To bridge this gap, a sliding window-based edge detection method is introduced, enabling the CR algorithm to be reliably applied to PAM signaling. Designed for deployment in oscilloscopes, the algorithm has been validated through simulations under varying noise levels and bandwidth-limited conditions, demonstrating reliable edge detection performance down to an SNR of 14 dB while exhibiting robustness against inter-symbol interference (ISI).
Keywords:
Clock recovery (CR)
edge detection
eye diagram
pulse-amplitude modulation (PAM)
serial communication
signal integrity (SI)
Journal
IF:
4.4
Papers:
1.3W
Citations:
2.2W