University of Illinois Urbana-Champaign
Signal integrity diagnosis and physical-based circuit modeling for 5G/6G connectors in high-speed electrical links
Abstract
dc:descriptionHigh-speed 5G/6G connectors have become a critical bottleneck in next-generation digital systems due to escalating data rates and stringent signal integrity (SI) requirements. This dissertation addresses these challenges by introducing a physics-based signal integrity diagnosis and modeling framework for multi-gigabit connectors. The approach combines field-based resonance analysis with a novel distributed physical-based transmission-line (dPBTL) circuit model, including an extended mixed-mode dPBTL (mm-dPBTL) formulation, to efficiently capture and interpret complex connector behaviors. Key contributions include explicit modeling of differential/common-mode signal paths loaded with ground-cavity and signal stub resonant structures, which are often responsible for narrowband SI degradations, and a mixed-mode analysis methodology to evaluate inter-mode and inter-pair interactions with resonant features. The proposed equivalent-circuit models faithfully reproduce full-wave connector responses over broad frequencies while maintaining physical transparency into resonant mechanisms. Applied to state-of-the-art connectors (e.g., PCIe 5.0 and OSFP), the framework demonstrates excellent agreement with 3D electromagnetic simulations and measurement data, enabling accurate prediction of S-parameters, NRZ/PAM-4 eye diagrams, and industry compliance metrics without resorting to time-intensive full-wave solves. The results show that this fast, interpretable SI tool can guide design pathfinding by pinpointing root causes of reflection, loss, crosstalk, mode conversion, and resonance issues and evaluating mitigation strategies. Ultimately, the developed modeling approach accelerates the design cycle for 5G/6G interconnects and provides engineers with deeper insight into achieving reliable high-speed link performance.
Degree
thesis:*- Name thesis:degree_name
- Ph.D.
- Level thesis:degree_level
- Dissertation
- Discipline thesis:degree_discipline
- Electrical & Computer Engr
- Grantor
- University of Illinois Urbana-Champaign
- Year dc:date
- 2025
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- He, Yulin
- Contributors dc:contributor
-
- Feng, Milton
- Jin, Jianming
- Dallesasse, John
- Schutt-Aine, Jose E
- Zhao, Yang
Subjects
dc:subject × 13Rights
dc:rights- Statement dc:rights
-
- Copyright 2025 Yulin He
- Language dc:language
- en, eng
Identifiers
dc:identifier.*- Handle dc:identifier
- https://hdl.handle.net/2142/130038