About me
Welcome to my page! I am Zheng (Will) Xing (邢正), currently pursuing a PhD at the School of Science and Engineering, The Chinese University of Hong Kong, Shenzhen. I am fortunate to be affiliated with the Laboratory for Wireless Communication and Intelligent Signal Processing (WiseLab), under the guidance of Prof. Junting Chen (陈俊挺).
Research Interests
-
Sequential Analysis in an Unsupervised Manner:
Exploring the segmentation of human motion sequences through temporal learning of neighboring frames. Additionally, analyzing wireless channel knowledge collected via trajectory data, with a focus on visualizing data collection trajectories. -
Wireless Indoor Localization:
Developing techniques for recovering data collection trajectories through RSS, CSI, AoA, and ToA measurements, with the goal of constructing radio maps to enable indoor localization of mobile users. -
Wireless Vehicle Trajectory Tracking:
Developing techniques for recovering data collection trajectories through CSI measurements from 5G MIMO base stations, in order to recover vehicle trajectories and construct radio maps for real-time vehicle tracking. -
High-Dimensional Data Clustering:
Investigating the neighbor relationships within data, constructing similarity matrices, and analyzing their block-diagonal properties to enhance clustering algorithms.
🔥 News
- 2025.01: 🎉🎉 One paper about vehicle trajectory map matching is accepted by IEEE Transactions on Intelligent Transportation Systems (T-ITS).
- 2025.01: 🎉🎉 One paper about outdoor radio map construction is accepted by IEEE International Conference on Communications (ICC).
📖 Educations
-
2020.09 - 2025.07
Doctor of Philosophy, Computer and Information Engineering, School of Science and Engineering, The Chinese University of Hong Kong, Shenzhen, China. Supervised by Assistant Professor Junting Chen (陈俊挺), Associate Professor Jie Xu (许杰), Professor Shuguang Cui (崔曙光). -
2017.09 - 2020.01
Master of Science, Control Science and Control Engineering, School of Electrical Science and Automation, Beihang University, Beijing, China. Supervised by Associate Professor Zhiyue Xu (徐志跃), Professor Yabin Liu (刘亚斌). -
2013.09 - 2017.07
Bachelor of Engineering, Electrical Engineering and Automation, School of Electrical Science and Automation, Ocean University of China, Qingdao, China. Supervised by Professor Ling Zhang (张玲), Professor Ming Li (黎明).
Experience
2018.09 - 2019.01, Universiti Malaya, Kuala Lumpur, Malaysia.
2014.09 - 2016.01, Dalian University of Technology, Dalian, China
📝 Publications
Journal Papers
Zheng Xing, Weibing Zhao
IEEE Transactions on Intelligent Transportation Systems (T-ITS), vol. 0, no. 0, pp. 0 - 0, January 2025, 中科院一区 (2023), JCR Q1
This paper introduces a method for vehicle trajectory reconstruction (VTR) using received signal strength (RSS) measurements, bypassing the need for specialized equipment. By leveraging a hidden Markov model-based technique and spatial-temporal correlations, the approach overcomes RSS data's inherent noise, outperforming state-of-the-art methods on real-world data.
Zheng Xing, Weibing Zhao
IEEE Internet of Things Journal (IoT-J), vol. 0, no. 0, pp. 0 - 0, November 2024, 中科院一区 (2023), JCR Q1
This paper presents a calibration-free indoor positioning method using received signal strength (RSS) measurements, eliminating the need for location labels or IMUs. By applying regional channel tracing (RCT) and subspace clustering, it accurately estimates locations and path loss models, achieving performance comparable to IMU-based methods in real-world environments.
Zheng Xing, Weibing Zhao
IEEE Transactions on Image Processing (TIP), vol. 0, no. 0, pp. 0 - 0, November 2024, 中科院一区 (2023), CCF A, JCR Q1
This paper introduces the Temporal Learning of Subspace Variety Model (TL-SVM) for human motion sequence segmentation and completion. It enhances segmentation by incorporating temporal priors, addressing missing entries through a spatio-temporal assignment consistency constraint, and completing sequences via subspace clustering. Extensive experiments demonstrate its superior performance.
Zheng Xing, Weibing Zhao
IEEE Transactions on Knowledge and Data Engineering (TKDE), vol. 36, no. 11, pp. 5709 - 5722, November 2024, 中科院一区 (2022), CCF A, JCR Q1
This paper introduces a refined DBSCAN algorithm utilizing block-diagonal similarity graphs, addressing challenges in high-dimensional data and parameter sensitivity. It employs a self-representation model, gradient descent, and a traversal algorithm to generate enhanced cluster orderings, achieving superior clustering performance on real-world datasets.
Zheng Xing, Junting Chen
IEEE Transactions on Signal Processing (TSP), vol. 72, no. 22, pp. 2512 - 2526, February 2024, 中科院一区 (2022), JCR Q1
This paper proposes an unsupervised method for constructing region-based radio maps without location labels, using RSS measurements. The method integrates subspace clustering with sequential priors, overcoming challenges posed by noisy and multipath-affected data. It achieves superior localization performance, outperforming supervised techniques like KNN, SVM, and DNN.
Conference Papers
Zheng Xing, Junting Chen
IEEE International Conference on Communications (ICC), 2025.
This paper proposes an unsupervised method for constructing angular power maps in massive MIMO networks using CSI data without location labels. By leveraging a hidden Markov model to estimate mobile trajectories, the approach enables accurate radio map construction and improves CSI prediction performance by over 20% compared to traditional methods.
Zheng Xing, Weibing Zhao
Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), vol. 38, no. 6, pp. 6270–6278, 2024.
This paper presents a fast, unsupervised action segmentation framework that divides videos into semantically consistent actoms. Using a subspace-based similarity measure, it efficiently splits and merges actoms, ensuring coherent segmentation. The method outperforms existing approaches in both accuracy and learning time across four benchmark datasets.
Zheng Xing, Junting Chen
IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 7060–7064, 2024.
This paper presents an HMM-based CSI embedding method for trajectory recovery in outdoor vehicular communication scenarios. By mapping CSI measurements to vehicle locations, it accurately reconstructs user trajectories with a 23-meter localization error. The method provides a promising approach for efficient radio map construction and localization.
Zheng Xing, Junting Chen, Yadan Tang
IEEE Global Communications Conference (GlobeCom), pp. 5360–5365, 2022.
This paper develops an integrated segmentation and subspace clustering method for RSS-based indoor localization with blind calibration. It utilizes sequential data and signal subspace structure to classify regions. The approach reduces region localization error by 50%, outperforming traditional and supervised methods, demonstrating superior performance in real-world scenarios.
Zheng Xing, Haoyun Li, Wenjie Liu, Zixiang Ren, Junting Chen, Jie Xu, Cai Qin
IEEE Wireless Communications and Networking Conference (WCNC), pp. 2136–2141, 2022.
This paper develops a model-assisted, data-driven approach for predicting spectrum efficiency (SE) in 5G networks, utilizing RSRP from drive test data. A joint interference and SE prediction network, along with a classification-assisted model, improves accuracy by approximately 2%, outperforming purely data-driven methods.
Patents
- 邢正,陈俊挺,一种基于接收信号强度的轨迹恢复及无线电地图构建方法, 中国, 1775494.6 1, 2024年9月27日, CN 117768843 B
- 邢正,陈俊挺, 一种室内定位方法及子空间特征提取方法, 中国, 1480461.4 2, 2023年08月18, CN 115988634 B
Projects
Projects
-
Outdoor Wireless Communication Spectrum Prediction
Role: Principal Investigator
Duration: 2020-2022
Collaborating Institution: Huawei Technologies, Shenzhen.
This project focused on predicting spectrum efficiency in outdoor wireless communication environments using data collected from mobile devices mounted on vehicles in Shenzhen, Chengdu, and Seoul. The data collected included CSI (Channel State Information), RSRP (Reference Signal Received Power), RSSI (Received Signal Strength Indicator), SINR (Signal-to-Interference-plus-Noise Ratio), CQI (Channel Quality Indicator), MCS (Modulation and Coding Scheme), Throughput, and Rank data from 1-6 neighboring cells and one main serving cell. I led the development of a neural network model that used this data to predict future spectrum efficiency upon receiving RSRP signals. These predictions were then transmitted back to the base station, allowing for dynamic antenna adjustments to enhance spectrum efficiency and optimize overall wireless communication performance in real-time. -
Air-to-Air Missile Effectiveness Testing
Role: Principal Investigator
Duration: 2017-2020
Collaborating Institutions: China Aerospace Corporation (First and Fifth Research Institutes), China Electronics Technology Group, China Air-to-Air Missile Research Institute.
The project aimed to test the effectiveness of both newly produced and stockpiled air-to-air missiles. As the principal investigator, my responsibilities included conducting performance tests on the missile’s guidance system (seeker). This involved developing a comprehensive testing platform, which included creating an environment with temperatures ranging from -40°C to 60°C, designing a vibration simulation environment, and providing stable 16V and 28V DC power supplies. Additionally, I was responsible for ensuring the proper functioning of signal interfaces, executing missile launch sequence tests to confirm that the guidance system transmitted control signals correctly, and designing a Windows-based graphical user interface (GUI) using C, C++, and C# to facilitate test operation and monitoring.
🎖 Honors and Awards
- Excellent Graduate of the Beihang University, 2020
- Excellent League Member of the Beihang University, 2018
- First-Class Scholarship of the Beihang University, 2017-2020
- Excellent Undergraduate of the Ocean University of China, 2017
- Excellent Undergraduate Thesis of the Ocean University of China, 2017
- Excellent Undergraduate of Shandong Province, 2017
- Silver Award in National Electronic Design Competition 2016
- Excellent Student of the Ocean University of China, 2016
- National Scholarship, 2016-2017
- Excellent Student of the Ocean University of China, 2014-2016
- First-Class Scholarship, 2013-2017
💻 Internships
- 2017.09 - 2020.01, China Aerospace Science and Technology, Beijing, China.
Academic Service
Reviewer for Journals:
- IEEE Journal on Selected Areas in Communications (JSAC)
- IEEE Transactions on Signal Processing (TSP)
- IEEE Transactions on Wireless Communications (TWC)
- IEEE Transactions on Mobile Computing (TMC)
- IEEE Transactions on Communications (TCOM)
- IEEE Internet of Things Journal (IOTJ)
- The Journal of Supercomputing
Reviewer for Conferences:
- IEEE International Conference on Communications (ICC)
- IEEE Global Communications Conference (Globecom)
- International Conference on Acoustics, Speech, and Signal Processing (ICASSP)
- International Conference on Learning Representations (ICLR)
- International Conference on Machine Learning (ICML)
- AAAI Conference on Artificial Intelligence (AAAI)
- IEEE/CIC International Conference on Communications in China (ICCC)
- IEEE International Conference on Machine Learning in Communications and Networking (ICMLCN)
- International Joint Conference on Neural Networks (IJCNN)
Teaching Experience
- MAT1001:Calculus I
- MAT2002:Ordinary Differential Equations
- ECE2810:Digital Systems Design Laboratory
- EIE3510:Digital Signal Processing
- EIE4512:Digital Image Processing
- EIE4005:Introduction to Coding and Cryptography
- EIE4006:Performance Evaluation of Communication Networks