Zhanjun Hao
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Zhanjun Hao has a total of 158 co-authors. Here are the top co-authors by number of shared publications:
- Xiaochao Dang
College of Computer Science and Engineering, Northwest Normal University, Lanzhou, Gansu, China
43 shared publications - Zhongyu Ma
Northwest Normal University
12 shared publications Xiaochao Dang
Northwest Normal University, Lanzhou, China
10 shared publications- Guowei Wang
Lanzhou University
6 shared publications - Hao Yan
College of Computer Science & Engineering, Northwest Normal University, Lanzhou 730070, China
5 shared publications - Qun Guo
College of Electrical and Information Engineering, LanZhou University of Technology, LanZhou, China
5 shared publications
Latest Publications
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 13 (21) , 50065-50079
CoTGait: A Co-Training-Based Semi-Supervised Framework for mmWave Radar Gait Recognition
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 13 (21) , 50065-50079
CoTGait: A Co-Training-Based Semi-Supervised Framework for mmWave Radar Gait Recognition
- 12 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Molecular, Biological, and Multi-Scale Communications
- Vol. 12, 563-575
Molecular Type Permutation Absorption Modulation: Block-Transmission, Detection, and Optimal Location
Unlike most existing modulation schemes for molecular communications, which transmit information through the properties of message molecules, molecular absorption shift keying is a newly proposed scheme that not only conveys information but also provides energy for molecule harvesting nodes. However, the principles of molecular absorption shift keying are tailored for scenarios with a single molecule harvesting node, limiting its versatility. In this paper, we introduce a novel modulation technique called molecular type permutation absorption modulation (MTPAM), where the transmitter encodes the information into the permutation of molecular types of the message molecules and the passive/absorbing states of molecule harvesting nodes. To enhance reliability and transmission rates while reducing the global synchronization requirements among harvesting nodes, we develop a flexible block-based transmission mechanism. Additionally, we implement a majority voting system on each passive link to improve performance. We design both an ideal maximum likelihood detector and a conventional maximum likelihood detector for MTPAM systems. The upper bounds for the bit error rate (BER) of these two detectors are derived. To effectively balance the performance of active and passive links, we utilize Bayesian optimization to quickly identify the optimal locations of harvesting nodes. Simulation results show that our scheme significantly outperforms the existing modulation schemes in terms of BER.
- 12 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Molecular, Biological, and Multi-Scale Communications
- Vol. 12, 563-575
Molecular Type Permutation Absorption Modulation: Block-Transmission, Detection, and Optimal Location
Unlike most existing modulation schemes for molecular communications, which transmit information through the properties of message molecules, molecular absorption shift keying is a newly proposed scheme that not only conveys information but also provides energy for molecule harvesting nodes. However, the principles of molecular absorption shift keying are tailored for scenarios with a single molecule harvesting node, limiting its versatility. In this paper, we introduce a novel modulation technique called molecular type permutation absorption modulation (MTPAM), where the transmitter encodes the information into the permutation of molecular types of the message molecules and the passive/absorbing states of molecule harvesting nodes. To enhance reliability and transmission rates while reducing the global synchronization requirements among harvesting nodes, we develop a flexible block-based transmission mechanism. Additionally, we implement a majority voting system on each passive link to improve performance. We design both an ideal maximum likelihood detector and a conventional maximum likelihood detector for MTPAM systems. The upper bounds for the bit error rate (BER) of these two detectors are derived. To effectively balance the performance of active and passive links, we utilize Bayesian optimization to quickly identify the optimal locations of harvesting nodes. Simulation results show that our scheme significantly outperforms the existing modulation schemes in terms of BER.
- Open Access
- 4 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 85-91
SARLiquid: Through-package Liquid Leakage Detection based on mmWave SAR Imaging
- 4 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 85-91
Open AccessSARLiquid: Through-package Liquid Leakage Detection based on mmWave SAR Imaging
- 2 January 2026
- Published by Springer Nature
- p. 147-159
Enhancing Indoor Trajectory Tracking with XGBoost-Based Classification on mmWave Radar Point Clouds
- 2 January 2026
- Published by Springer Nature
- p. 147-159
Enhancing Indoor Trajectory Tracking with XGBoost-Based Classification on mmWave Radar Point Clouds
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Communications Letters
- Vol. 30, 2610-2614
XAI-Aided Hybrid Neural Network Detection for Probabilistic Shaping-Based Molecular Communication
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Communications Letters
- Vol. 30, 2610-2614
XAI-Aided Hybrid Neural Network Detection for Probabilistic Shaping-Based Molecular Communication
- 31 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 25 (6) , 8432-8449
mm-ARnet: Exploring Millimeter Wave Radar Point Clouds for Human Action Recognition
Human Action Recognition (HAR) offers a wide range of applications, including smart home, smart health, entertainment, security, and surveillance. Traditional vision-based HAR systems face significant limitations due to privacy concerns, lighting dependency, and poor performance in complex environments. Millimeter-wave radar-based activity recognition systems have attracted considerable attention due to their superior sensing capabilities, device-free deployment, privacy preservation, and robustness to environmental variations. However, existing approaches struggle with the inherent sparsity and noise in mmWave radar data, particularly for diverse activity categories spanning from full-body movements to subtle localized gestures. This study proposes mm-ARnet, a comprehensive millimeter-wave point-cloud-based framework for recognizing 16 diverse human activities across three distinct behavioral categories: full-body movements, posture transitions, and localized body movements. Our approach leverages 4D point cloud sequences and introduces a multi-frame fusion with stochastic sampling strategy to enhance point cloud density and mitigate sparsity effects. The core innovation lies in our lightweight TCN+Bi-LSTM temporal modeling pipeline integrated with a novel Temporal Pattern Attention (TPA) mechanism. Extensive experiments conducted across three real-world scenarios with 10 participants demonstrate that mm-ARnet achieves 97.42% accuracy, outperforming state-of-the-art methods while maintaining superior temporal performance.
- 31 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 25 (6) , 8432-8449
mm-ARnet: Exploring Millimeter Wave Radar Point Clouds for Human Action Recognition
Human Action Recognition (HAR) offers a wide range of applications, including smart home, smart health, entertainment, security, and surveillance. Traditional vision-based HAR systems face significant limitations due to privacy concerns, lighting dependency, and poor performance in complex environments. Millimeter-wave radar-based activity recognition systems have attracted considerable attention due to their superior sensing capabilities, device-free deployment, privacy preservation, and robustness to environmental variations. However, existing approaches struggle with the inherent sparsity and noise in mmWave radar data, particularly for diverse activity categories spanning from full-body movements to subtle localized gestures. This study proposes mm-ARnet, a comprehensive millimeter-wave point-cloud-based framework for recognizing 16 diverse human activities across three distinct behavioral categories: full-body movements, posture transitions, and localized body movements. Our approach leverages 4D point cloud sequences and introduces a multi-frame fusion with stochastic sampling strategy to enhance point cloud density and mitigate sparsity effects. The core innovation lies in our lightweight TCN+Bi-LSTM temporal modeling pipeline integrated with a novel Temporal Pattern Attention (TPA) mechanism. Extensive experiments conducted across three real-world scenarios with 10 participants demonstrate that mm-ARnet achieves 97.42% accuracy, outperforming state-of-the-art methods while maintaining superior temporal performance.
- Open Access
- 22 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 128-133
Molecular Type Permutation Absorption Modulation
- 22 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 128-133
Open AccessMolecular Type Permutation Absorption Modulation
- 14 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 12 (24) , 54189-54200
SonicFER: Facial Expressions Tracking Through a Commercial Smartphone Speaker
Facial expression recognition (FER) technology plays a significant role in advancing the intelligence and personalization of human–computer interaction. Although ultrasonic-based expression recognition methods already exist, most rely on phase shifts or Doppler effects, which suffer from insufficient resolution to accurately capture subtle facial expression variations. To address this issue, this article proposes an innovative FER system, SonicFER, which utilizes smartphones to emit ultrasonic waves and receive echoes, achieving fine-grained perception of facial expressions through real-time monitoring and analysis of channel impulse response (CIR). To effectively detect facial expression movements while eliminating static interference and minor motion artifacts, this article employs a differential operation combined with variance calculation, along with setting appropriate thresholds for filtering. Through rigorous experimental evaluation, SonicFER achieves a high accuracy of 91.2% in recognizing six common facial expressions and outperforms state-of-the-art technologies across various real-world scenarios.
- 14 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 12 (24) , 54189-54200
SonicFER: Facial Expressions Tracking Through a Commercial Smartphone Speaker
Facial expression recognition (FER) technology plays a significant role in advancing the intelligence and personalization of human–computer interaction. Although ultrasonic-based expression recognition methods already exist, most rely on phase shifts or Doppler effects, which suffer from insufficient resolution to accurately capture subtle facial expression variations. To address this issue, this article proposes an innovative FER system, SonicFER, which utilizes smartphones to emit ultrasonic waves and receive echoes, achieving fine-grained perception of facial expressions through real-time monitoring and analysis of channel impulse response (CIR). To effectively detect facial expression movements while eliminating static interference and minor motion artifacts, this article employs a differential operation combined with variance calculation, along with setting appropriate thresholds for filtering. Through rigorous experimental evaluation, SonicFER achieves a high accuracy of 91.2% in recognizing six common facial expressions and outperforms state-of-the-art technologies across various real-world scenarios.