I am a PhD Student in Data Science at Tsinghua Shenzhen International Graduate School (SIGS), Tsinghua University, in Data Science and Information Technology. I am advised by Prof. Ercan Engin Kuruoglu. My research focuses on graph signal processing and time-varying graphs, robust state estimation, and statistical signal processing.

My recent work includes robust Kalman filtering under hybrid Gaussian and heavy-tailed measurement noise, graph-based adaptive learning for drone networks, and efficient speech-enhancement methods for microphone arrays.

Research Interests

Graph Signal Processing and Time-Varying Graphs

I am interested in signal processing and learning on time-varying graphs, including graph-based modeling, filtering, adaptive learning, and distributed inference for networked and multi-agent systems.

Robust State Estimation

I study robust Bayesian filtering and Kalman-type methods for state estimation under non-Gaussian, impulsive, and hybrid measurement noise. A central goal is to build estimators that remain reliable while retaining efficient recursive updates.

Statistical Signal Processing

My broader interests include statistical modeling, Bayesian inference, and learning for signal processing. I have applied these ideas to acoustic sensing, microphone-array processing, beamforming, direction-of-arrival estimation, and real-time speech enhancement.

Selected Publications

SARC: Structured Analytic-Residual Correction for Lightweight Multichannel Speech Enhancement
Gurui Jin, Xiao-Ping Zhang, Le Xiao, Yue Wang, Zhenyu Liu
Submitted to ICASSP 2027.
Publication · Code

Robust Kalman Filter For Additive Gaussian-Student’s t Distribution
Gurui Jin, Pengcheng Hao, Zhenyu Liu, Xiao-Ping Zhang, Ercan E. Kuruoglu
2026 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2026.
Publication · Code

Robust Kalman Filters For Additive Gaussian-GSM Distributions
Gurui Jin, Pengcheng Hao, Zhenyu Liu, Xiao-Ping Zhang, Ercan E. Kuruoglu
Manuscript, to be submitted.
Publication · Code

Current & Recent Activities

• 2026 — SARC, a lightweight multichannel speech-enhancement method, submitted to ICASSP 2027.
• 2026–Present — PhD research in Data Science and Information Technology at Tsinghua SIGS, focusing on time-varying graphs and graph-based adaptive learning for drone networks.
• 2026 — Paper on robust Kalman filtering for additive Gaussian-Student’s t noise presented at ICASSP 2026.
• 2025–2026 — Algorithm Engineering Intern at Shokz, working on microphone-array beamforming and real-time speech enhancement.

For a fuller overview of my work, see Research, Publications, or my CV.