A5100427215- Privacy-Preserving Technologies in Data
- Irrigation Practices and Water Management
- Cryptography and Data Security
- Stroke Rehabilitation and Recovery
- Agronomic Practices and Intercropping Systems
- Polymer-Based Agricultural Enhancements
- Vehicular Ad Hoc Networks (VANETs)
- Neural Networks and Applications
- IoT Networks and Protocols
- Multimedia Communication and Technology
- Wireless Communication Networks Research
- Distributed and Parallel Computing Systems
- Cloud Data Security Solutions
- Cooperative Communication and Network Coding
- Chaos-based Image/Signal Encryption
- Internet Traffic Analysis and Secure E-voting
- Human Pose and Action Recognition
- Multimodal Machine Learning Applications
- IoT and Edge/Fog Computing
- Wireless Body Area Networks
- Soil Mechanics and Vehicle Dynamics
- Catalysis and Oxidation Reactions
- Advanced MIMO Systems Optimization
- Advanced Malware Detection Techniques
- Plant responses to water stress
Nanjing Forestry University
2023-2025
Beijing Electronic Science and Technology Institute
2024
Xidian University
2022
University of Massachusetts Boston
2018
Federated learning (FL) is a machine technique that allows for on-site data collection and processing without sacrificing privacy transmission. Heterogeneity key challenge in federated settings. Recently, cross-technology communication (CTC) has emerged as solution Internet of Things (IoT) heterogeneity, enabling direct between different wireless devices the need hardware modifications or gateway intervention. For example, sophisticated WiFi device can serve central coordinator other...
Reinforcement learning with human feedback (RLHF) fine-tunes a pretrained large language model (LLM) using preference datasets, enabling the LLM to generate outputs that align preferences. Given sensitive nature of these datasets held by various clients, there is need implement RLHF within federated (FL) framework, where clients are reluctant share their data due privacy concerns. To address this, we introduce feasible framework in which collaboratively train binary selector our proposed...
Massive computing tasks have been generated with the widespread applications of big data analysis in vehicular edge (VEC) networks. However, offloading process VEC networks suffers a threat information leakage. The physical layer security (PLS) technology is an effective solution to protect confidential information. Furthermore, contradiction between massive transmission and limited communication resources promotes urgent need for proper scheme improve resource utilization. In this paper, we...
Cross-technology communication(CTC) enables seamless interactions between diverse wireless technologies. Most existing work is based on reversing the transmission path to identify appropriate payload generate waveform that target devices can recognize. However, this method suffers from many limitations, including dependency specific technologies and necessity for intricate algorithms mitigate distortion. In work, we present NNCTC, a Neural-Network-based Cross-Technology Communication...
Autonomous vehicles (AV) need an access to a big amount of data from online roadmap servers and local sensors make real-time control decision for safety. Considering the may change in cases road maintenance, new/broken signs, accidents, traffic jams, AV should keep its updated with servers. In this paper, we propose efficient privacy-preserving update scheme AV, which achieves privacy preservation route information while minimizing overhead. We define new type sensitivity segment based on...
The topic of non-contact diagnosis became a hot during COVID-19 and online consultation gained popularity. In this research, deep learning-based autonomous limb evaluation system is developed for remote rehabilitation training people with physical limitations. Its main goal to collect analyze information about states. patient can evaluate the state at home using mobile app, doctor view data connect via web's chat module offer diagnostic opinions. Deep learning used Start/End Attitude...
AbstractBecause traditional slow-release fertilizers cannot effectively retain soil heat and moisture, their effect on increasing crop yields is limited. To solve this problem, novel sprayable liquid mulch films (LSA) were prepared with a urea function, created by emulsion polymerization between lignin (extracted from papermaking waste liquor), starch, acrylate monomers. It was noted that the introduction of transformed LSA into pseudoplastic fluid, enhancing sprayability emulsion. Lignin...
Smart contracts play a vital role in the Ethereum ecosystem. Due to prevalence of kinds security issues smart contracts, contract verification is urgently needed, which process matching contract's source code its on-chain bytecode for gaining mutual trust between developers and users. Although services are embedded both popular browsers (e.g., Etherscan Blockscout) official platforms (i.e., Sourcify), gain great popularity ecosystem, their trustworthiness remain unclear. To fill void, we...
Privacy concerns have led to a surge in the creation of synthetic datasets, with diffusion models emerging as promising avenue. Although prior studies performed empirical evaluations on these models, there has been gap providing mathematical characterization their privacy-preserving capabilities. To address this, we present pioneering theoretical exploration privacy preservation inherent discrete (DDMs) for dataset generation. Focusing per-instance differential (pDP), our framework...
Physical rehabilitation assessment is currently faced with problems such as complex content, lack of modules, and being easily limited by resources. This paper designs an automated lower limb evaluation system, which includes the tasks for limbs "follow walk", "squat down", "pickup", "throw pick", realize virtual dog using HoloLens2.0 that can be covered on service robot DJI RoboMaster in real-time. The walk" "pickup" require to automatically set target point lead patient through automatic...