Scholarlink

<p>The rapid proliferation of Internet of Things (IoT) devices and the exponential increase in generated data necessitate advanced distributed computing systems to enhance performance and ensure scalability. By processing data closer to IoT devices, Fog Computing has emerged as a promising paradigm to improve Quality of Service (QoS). Within this domain, Mobile Crowdsensing (MCS) has demonstrated its efficacy in gathering environmental and social data through active user participation. However, challenges such as optimal user selection and efficient task allocation continue to hinder system performance, affecting task coverage, platform costs, and reliability in task execution. This study introduces ADOPT (Aquila-Differential Optimizer for Efficient Task Allocation and QoS Enhancement). To address these challenges, this novel optimization framework synergizes the Aquila Optimizer (AO) and Differential Evolution (DE) algorithms. The AO algorithm ensures robust exploration and identifies near-optimal task allocation solutions, while the DE algorithm enhances exploitation by refining solutions, avoiding local optima, and accelerating convergence. ADOPT is evaluated through extensive MATLAB simulations against state-of-the-art methods, including PTAM-GA, TSA, CSTA, and FITMCS. The evaluation comprises two key scenarios: one where the number of users exceeds the number of tasks, and another where tasks outnumber users. Each scenario is analyzed across three scales—small, medium, and large—to comprehensively assess the adaptability and scalability of the proposed approach. Results indicate that ADOPT consistently outperforms competing methods, achieving a 56.11% improvement in task coverage, an 18.01% reduction in platform costs, and a 12.05% enhancement in execution reliability. Notably, in scenarios with an overwhelming number of tasks relative to users, ADOPT demonstrates exceptional resilience and efficiency, confirming its robustness and scalability. This work establishes a strong foundation for optimizing MCS systems, paving the way for more efficient task allocation and superior QoS in dynamic IoT environments.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

ADOPT—the aquila-differential optimizer for efficient task allocation and QoS enhancement in mobile crowdsensing

  • 期刊论文

  • 发表日期：

    2025年9月15日

查看全文
  • Hadi Ghahremani,
  • Masumeh Damrudi,
  • Ali Ghaffari,
  • Kamal Jadidy Aval

摘要

The rapid proliferation of Internet of Things (IoT) devices and the exponential increase in generated data necessitate advanced distributed computing systems to enhance performance and ensure scalability. By processing data closer to IoT devices, Fog Computing has emerged as a promising paradigm to improve Quality of Service (QoS). Within this domain, Mobile Crowdsensing (MCS) has demonstrated its efficacy in gathering environmental and social data through active user participation. However, challenges such as optimal user selection and efficient task allocation continue to hinder system performance, affecting task coverage, platform costs, and reliability in task execution. This study introduces ADOPT (Aquila-Differential Optimizer for Efficient Task Allocation and QoS Enhancement). To address these challenges, this novel optimization framework synergizes the Aquila Optimizer (AO) and Differential Evolution (DE) algorithms. The AO algorithm ensures robust exploration and identifies near-optimal task allocation solutions, while the DE algorithm enhances exploitation by refining solutions, avoiding local optima, and accelerating convergence. ADOPT is evaluated through extensive MATLAB simulations against state-of-the-art methods, including PTAM-GA, TSA, CSTA, and FITMCS. The evaluation comprises two key scenarios: one where the number of users exceeds the number of tasks, and another where tasks outnumber users. Each scenario is analyzed across three scales—small, medium, and large—to comprehensively assess the adaptability and scalability of the proposed approach. Results indicate that ADOPT consistently outperforms competing methods, achieving a 56.11% improvement in task coverage, an 18.01% reduction in platform costs, and a 12.05% enhancement in execution reliability. Notably, in scenarios with an overwhelming number of tasks relative to users, ADOPT demonstrates exceptional resilience and efficiency, confirming its robustness and scalability. This work establishes a strong foundation for optimizing MCS systems, paving the way for more efficient task allocation and superior QoS in dynamic IoT environments.