Junchao Fan

ORCID: 0000-0003-0617-8101
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Privacy-Preserving Technologies in Data
  • UAV Applications and Optimization
  • Digital Rights Management and Security
  • Advanced Steganography and Watermarking Techniques
  • Digital and Cyber Forensics
  • Digital Media Forensic Detection
  • IoT and Edge/Fog Computing
  • Stochastic Gradient Optimization Techniques
  • Vehicular Ad Hoc Networks (VANETs)
  • Internet Traffic Analysis and Secure E-voting
  • Adversarial Robustness in Machine Learning
  • Facility Location and Emergency Management
  • Vehicle Routing Optimization Methods
  • Advanced Neural Network Applications
  • Supply Chain and Inventory Management
  • Elevator Systems and Control
  • Transportation and Mobility Innovations
  • Autonomous Vehicle Technology and Safety
  • Evacuation and Crowd Dynamics
  • Wireless Communication Security Techniques
  • Distributed Control Multi-Agent Systems
  • Cryptography and Data Security
  • Security and Verification in Computing
  • Robotic Path Planning Algorithms
  • Infrastructure Resilience and Vulnerability Analysis

Beijing Jiaotong University
2021-2025

This article investigates a hierarchical aerial computing system, where both high-altitude platforms (HAPs) and unmanned vehicles (UAVs) provision computation services for ground devices (GDs). Different from the existing works which ignored UAV task offloading to HAPs suffered long transmission delay between GDs, in our UAVs are responsible collecting tasks generated by GDs. Considering limited resources constrained coverage, need cooperatively allocate their (including spectrum, caching,...

10.1109/jiot.2023.3240173 article EN IEEE Internet of Things Journal 2023-01-27

This paper investigates a dual-unmanned aerial vehicle (UAV) aided communication system to improve the security of between ground devices and UAVs. Different from existing works which ignored mobility just considered one-way UAVs, we allow be mobile consider bi-directional ground-UAV security. Specifically, one UAV server communicates with devices, other jammer is invoked confuse eavesdroppers. Our objective maximize worst-case average secrecy rate by joint optimization trajectory sender...

10.1109/tvt.2022.3184804 article EN IEEE Transactions on Vehicular Technology 2022-06-20

Traditional covert transmission (CT) approaches have been hindering CT application while blockchain technology offers new avenue. Current blockchain-based require off-chain negotiation of critical information and often overlook the dynamic updating session keys, which increases risk message key leakage. Additionally, in some transactions exhibit obvious characteristics that can be easily detected by third-parties. Moreover, most do not address issue decreased reliability attack scenarios....

10.1109/ton.2024.3519860 article EN 2025-01-01

Recently, big data has seen explosive growth in the Internet of Things (IoT). Multi-layer FL (MFL) based on cloud-edge-end architecture can promote model training efficiency and accuracy while preserving IoT privacy. This paper considers an improved MFL, where edge layer devices own private join process. iMFL improve resource utilization also alleviate strict requirement end devices, but suffers from issues Data Reconstruction Attack (DRA) unacceptable communication overhead. aims to address...

10.1109/jiot.2024.3360007 article EN IEEE Internet of Things Journal 2024-01-30

Unmanned aerial vehicles (UAVs) are being broadly employed to assist in efficient data collection for Internet of Things (IoT) networks. Studies have been conducted ensure the effectiveness and safety UAVs process. However, they only considered part challenges energy consumption, collision avoidance, mobility IoT devices. In this article, we study a UAV path planning optimization problem UAV-assisted maximize amount collected data. Different from these existing works, not considers all...

10.1109/jiot.2024.3448537 article EN IEEE Internet of Things Journal 2024-08-23

The Virtual Machine (VM)-based Trusted-Execution-Environment (TEE) technology, like AMD Secure-Encrypted-Virtualization (SEV), enables the establishment of Confidential VMs (CVMs) to protect data privacy. But CVM lacks ways provide trust proof its running state, degrading user confidence using CVM. technology virtual Trusted Platform Module (vTPM) can be used generate for However, existing vTPM-based approaches have weaknesses lack a well-defined root-of-trust, vTPM protection, and vTPM's...

10.48550/arxiv.2405.01030 preprint EN arXiv (Cornell University) 2024-05-02

Traditional covert transmission (CT) approaches have been hindering CT application while blockchain technology offers new avenue. Current blockchain-based require off-chain negotiation of critical information and often overlook the dynamic session keys updating, which increases risk message key leakage. Additionally, in some transactions exhibit obvious characteristics that can be easily detected by third-parties. Moreover, most do not address issue decreased reliability attack scenarios....

10.48550/arxiv.2405.04046 preprint EN arXiv (Cornell University) 2024-05-07

Emergency supplies allocation plays a crucial role in post-disaster relief operations, requiring prompt and efficient strategies. As powerful tool for addressing decision-making problems, reinforcement learning (RL) has been explored to address emergency problems. However, existing RL-based approaches tend overlook the uncertainty demand, which is prevalent factor real disaster scenarios. In this paper, we first introduce multi-period multi-objective problem with demand formulate as...

10.2139/ssrn.4836786 preprint EN 2024-01-01

Data provision, referring to the data upload and access, is one key phase in vehicular digital forensics. The unique features of Driverless Taxi (DT) bring new issues this phase: 1) efficient verification integrity when diverse Providers (DPs) data; 2) DP privacy preservation during upload; 3) both INvestigator (IN) under complex ownership accessing data. To end, we propose a novel Lightweight Privacy-preserving Provision (LPDP) approach consisting three mechanisms: Privacy-friendly Batch...

10.48550/arxiv.2409.14039 preprint EN arXiv (Cornell University) 2024-09-21

Despite significant advancements in deep reinforcement learning (DRL)-based autonomous driving policies, these policies still exhibit vulnerability to adversarial attacks. This poses a formidable challenge the practical deployment of driving. Designing effective attacks is an indispensable prerequisite for enhancing robustness policies. In view this, we present novel stealthy and efficient attack method DRL-based Specifically, introduce adversary designed trigger safety violations (e.g.,...

10.48550/arxiv.2412.03051 preprint EN arXiv (Cornell University) 2024-12-04

Recently, big data has seen explosive growth in the Internet of Things (IoT). Multi-layer FL (MFL) based on cloud-edge-end architecture can promote model training efficiency and accuracy while preserving IoT privacy. This paper considers an improved MFL, where edge layer devices own private join process. iMFL improve resource utilization also alleviate strict requirement end devices, but suffers from issues Data Reconstruction Attack (DRA) unacceptable communication overhead. aims to address...

10.48550/arxiv.2309.13864 preprint EN other-oa arXiv (Cornell University) 2023-01-01

This paper investigates a maintenance worker scheduling system for charging pile faults, where workers are scheduled to repair the faulty piles. The existing works mostly adopt traditional optimization methods personnel scheduling, which cannot adapt dynamic environments and suffers high computation complexity large-scale problems. Our objective is minimize repairing time by reasonably workers. To this end, problem first formulated as partially observable Markov decision process (POMDP)...

10.1109/ispa-bdcloud-socialcom-sustaincom57177.2022.00018 article EN 2022-12-01
Coming Soon ...