Lei Zhen

Mr. Lei Zhen


Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
6 November 2024 in Applied Sciences
Verified
Journal Article
Open access
PDF file
ZZZhiyong Zha
Zhiyong Zha
JHJianwen He
Jianwen He
LZLei Zhen
Lei Zhen
MYMingyang Yu
Mingyang Yu
CDChenxi Dong
Chenxi Dong
KPKai Peng
Kai Peng
With the development of IoT technology, central cloud servers and edge-computing servers together form a cloud–edge communication network to meet the increasing demand for computing tasks. The data transmitted in this network is of high value, so the ability to quickly and accurately predict the traffic load of each link becomes critical to ensuring the security and stable operation of the network. In order to effectively counter the potential threat of flood attacks on network stability, we combine the Bi-directional Gated Recurrent Unit (BiGRU) model with the Dung Beetle Optimizer (DBO) algorithm to design a DBO-BiGRU short-term traffic load prediction model. Experimental validation on a public dataset shows that the proposed model has better prediction accuracy and fit than the mainstream models of RNN, LSTM, and TCN.

2 January 2025 in Symmetry
Verified
Journal Article
Open access
PDF file
GWGeng Wu
Geng Wu
LJLu Jing
Lu Jing
DHDai Hou
Dai Hou
LZLei Zhen
Lei Zhen
KLKaihong Li
Kaihong Li
KPKai Peng
Kai Peng
In recent years, unmanned aerial vehicles (UAVs, also known as drones) have gained widespread application in fields such as data collection and inspection, owing to their lightweight design and high mobility. However, due to limitations in battery life, UAVs are often unable to independently complete large-scale data collection tasks. To address this limitation, vehicle–drone collaborative data collection has emerged as an effective solution. Existing research, however, primarily focuses on collaborative work in static task scenarios, overlooking the complexities of dynamic environments. In dynamic scenarios, tasks may arrive during the execution of both the vehicle and UAV, and each drone has different positions and remaining endurance, creating an asymmetric state. This introduces new challenges for path planning. To tackle this challenge, we propose a 0–1 integer programming model aimed at minimizing the total task completion time. Additionally, we introduce an efficient dynamic solving algorithm, referred to as Greedy and Adaptive Memory Process-based Dynamic Algorithm (GAMPDA). This algorithm first generates an initial global data collection plan based on the initial task nodes and dynamically adjusts the current data collection scheme using a greedy approach as new task nodes arrive during execution. Through comparative experiments, it was demonstrated that GAMPDA outperforms SCAN and LKH in terms of time cost, vehicle travel distance, and drone flight distance and approaches the ideal results. GAMPDA significantly enhances task completion efficiency in dynamic scenarios, providing an effective solution for collaborative data collection tasks in such environments.

21 April 2025 in Applied Sciences
Verified
Journal Article
Open access
PDF file
DHDai Hou
Dai Hou
LWLan Wei
Lan Wei
LZLei Zhen
Lei Zhen
GWGeng Wu
Geng Wu
HJHu JiaXing
Hu JiaXing
KPKai Peng
Kai Peng
With the gradual application of blockchain technology in the domain of Social Internet of Vehicles (SIoV), the increasing volume of blockchain data has imposed significant storage pressure on roadside units (RSUs). Collaborative storage schemes, which organize RSUs into clusters to jointly store content for vehicles, have been explored. However, existing collaborative storage solutions in IoV primarily focus on caching content and are not well-suited to the deployment constraints of blockchain networks. Building on blockchain’s decentralized characteristics and data integrity mechanisms, this paper proposes a collaborative storage scheme that reduces RSU storage loads while sustaining distributed ledger operations in SIoV. Specifically, the RSU Access Preference-based Spectral Clustering Algorithm (RAPSCA) is proposed to address RSU clustering by analyzing both the RSUs’ access preferences for blockchain data and their resource availability. Subsequently, the Vehicle Service Priority-based Greedy Block Allocation Algorithm (VSPGBAA) is devised for intra-cluster storage allocation, which considers vehicles’ dwell times and block access probabilities to reduce overall access costs. Experimental results indicate that, compared to baseline algorithms, the proposed method achieves a 27.7% reduction in cost and a 3.5-fold decrease in execution time, thereby demonstrating the feasibility of collaborative storage optimization in blockchain-enabled SIoV.

10 April 2025 in Applied Sciences
Verified
Journal Article
Open access
PDF file
DHDai Hou
Dai Hou
ZYZhenkai Yang
Zhenkai Yang
LZLei Zhen
Lei Zhen
BJBo Jin
Bo Jin
HXHuan Xu
Huan Xu
KPKai Peng
Kai Peng
In-context learning (ICL) enables large language models (LLMs) to adapt to new tasks using only a few examples, without requiring fine-tuning. However, the new privacy and security risks brought about by this increasing capability have not received enough attention, and there is a lack of research on this issue. In this work, we propose a novel membership inference attack (MIA) method, termed Neighborhood Deviation Attack, specifically designed to evaluate the privacy risks of LLMs in ICL. Unlike traditional MIA methods, our approach does not require access to model parameters and instead relies solely on analyzing the model’s output behavior. We first generate neighborhood prefixes for target samples and use the LLM, conditioned on ICL examples, to complete the text. We then compute the deviation between the original and completed texts and infer membership based on these deviations. We conduct experiments on three datasets and three LLMs and further explore the influence of key hyperparameters on the method’s performance and their underlying reasons. Experimental results show that our method is significantly better than the comparative methods in terms of stability and achieves better accuracy in most cases. Furthermore, we discuss four potential defense strategies, including increasing the diversity of ICL examples and introducing controlled randomness in the inference process to reduce the risk of privacy leakage.