BQBin Qu

Bin Qu

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Bin Qu has a total of 14 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Q-Learning Enabled Ant Colony (Q-ANT)-Based Navigation Optimization Framework for UAVs

    by Xing Wei, Ning Li, Hua Yang, Bin Qu

    Flying ad-hoc networks (FANETs) composed of unmanned aerial vehicles (UAVs) which are serving in both military as well as civilian domains. The environmental changes and alterations in network topology present considerable challenges to communication among drone swarms. Consequently, there is an imperative need to investigate a high-performance routing scheme for UAV based FANETs. In this study, a routing scheme is introduced that employs the reinforcement Q-learning algorithm to enhance the ant colony optimization routing protocol (Q-ANT). The primary innovation of this scheme lies in the implementation of a neighborhood link selection mechanism that incorporates link residual time and link signal strength. This approach addresses the inadequacy of relying solely on a single indicator for assessing link residual survival time and expedites the convergence velocity of the ant colony routing algorithm. Q-ANT possesses the capability to optimize its routing strategy in accordance with the prevailing network environment by utilizing the Q-learning algorithm to attain the ideal option strategy in response to environmental fluctuations. The proposed scheme is implemented within the NS3 simulation environment and scenario-based simulation experiments have been conducted on three critical network metrics: network lifetime, end-to-end delay, and packet delivery rate. The experimental results exhibit the superiority of Q-ANT in terms of minimal energy consumption, minimal latency, and an enhanced packet delivery rate to the existing approaches such as FAnt-HocNet and Q-learning based QRF.


  • Optimizing Dynamic Cache Allocation in Vehicular Edge Networks: A Method Combining Multisource Data Prediction and Deep Reinforcement Learning

    by Bin Qu, De Li, Xianxian Li

    Currently, most studies on content caching strategies for vehicular networks rely on the Zipf distribution model of content popularity. This approach, however, often fails to accommodate the time-varying nature of content popularity and vehicular traffic. Although some studies have begun considering time-varying content popularity, they overlook the implications of time-varying vehicle traffic, which significantly influences content allocation caching policies and could potentially reduce edge server rental costs for service providers. Addressing this gap, our study takes both time-varying content popularity and vehicle traffic data into account. From the perspective of content service providers, we need to strike a tradeoff in optimizing the hit rate (HR) based on the foundation of utility enhancement. Accordingly, we model the problem as a multiobjective issue that jointly optimizes HR, utility, and replacement cost. By employing a scalar function, we transform this multiobjective problem into a single-objective one, allowing for the modulation of optimization objective priorities via weight tuning. We employ two gated recurrent unit networks to predict both vehicle traffic volume and content popularity for the next moment. We use the predicted area traffic flow volume and content popularity as the system state for our cache allocation decision optimization, which is accomplished using a deep reinforcement learning model. We conducted a series of evaluation experiments using real data sets, and our proposed approach significantly outperforms the baseline scheme.


  • Open Access

    GCBlock: A Grouping and Coding Based Storage Scheme for Blockchain System

    by Bin Qu, Li-E Wang, Peng Liu, Zhenkui Shi, Xian Xian Li

    To achieve the tamper-proof, reliability and traceability of transactions in a trustless environment, the blockchain requires each peer node to store the whole global ledger. However, as transactions keep increasing over time, the storage cost of each node increases. In addition, many schemes have been proposed to boost rapid transactions which will even lead transactions to grow explosively. The problem of storage is becoming one challenge of blockchain since the storage overhead of each node increase rapidly. Reducing the storage overhead of each participant is very urgent and worthy. In this paper, we present GCBlock: a grouping overlay network storage scheme for the blockchain which can reduce the storage overhead of nodes and cut the whole storage cost of the blockchain greatly while keeping the underlying protocols. In our scheme, we try to group the nodes according to their physical fuzzy distance to reduce the overall delay when tracing. We set rules of autonomous check to deal with evil behavior within the group. To further enhance the stability of our scheme, we propose the transcript fractional repetition code which is newly constructed based on the fractional repetition code to encode data. Finally, we make a comprehensive evaluation of GCBlock and the results show that it is workable and reasonable.