About Me

I’m Zhiqiang Wei who is working at China Mobile (Suzhou) Software Technology Co.,Ltd now. As a researcher, I focus on deep learning-based network protocol fuzzing test techniques. Currently, I investigate how to use apply deep generative model(such as GAN, VAE) for message queues fuzzing test case generation. I have proposed a novel GAN-based fuzzing architecture which aims at detecting vulnerabilities in MQTT protocol. Before joining China Mobile, I worked as an NLP engineer in Microsoft(Suzhou) Bing Map team. I’m also interested in multimodal machine learning approaches. I have designed machine learning applications to compensate the missing modality from multisensory data under real-world indoor localization scenarios.

Research Interests

• Fuzzing Test
• Multimodal Machine Learning
• Location Tracking System
• On-device Machine Learning
• Time-Series Forecasting
• Deep Generative Model
• Natural Language Generation

News

[Feb. 2024] Paper accepted at IEEE Access, 2024. Work with Xijia Wei and Xinghua Zhao, “SGANFuzz: A Deep Learning-based MQTT Fuzzing Method using Generative Adversarial Networks”. Link

[Dec. 2022] Paper accepted at ACAI 2022. Work with Xinghua Zhao, “An LSTM-based method for Message Queue Throughput Prediction”. Link

[Oct. 2021] Journal accepted at Sensors. Work with Xijia Wei and Valentin Radu, “Sensor-fusion for Smartphone Location Tracking using Hybrid Multimodal Deep Neural Networks”. Link

[Sept. 2021] Paper accepted at IPIN, 2021. Work with Xijia Wei and Valentin Radu, “MM-Loc: Cross-sensor Indoor Smartphone Location Tracking using Multimodal Deep Neural Networks”. See you in Lloret de Mar, Spain. Link