Scholarlink

<p>Mobile crowdsensing (MCS) faces fundamental challenges in task allocation due to unpredictable user mobility and uncertain participation, which can degrade sensing quality and increase energy waste. This work introduces the first unified task-allocation framework that jointly predicts users’ future movement paths and block-level participation rates addressing both challenges simultaneously. User mobility and participation are modeled in an offline phase using bidirectional GRU networks, and the predictions guide task assignment in the allocation phase. Three population-based metaheuristic algorithms (BE-WOA, UBPSO, and ACS) are employed to maximize data quality while minimizing users’ energy consumption. Experiments on 11 task-allocation scenarios using two real-world datasets (110 simulation runs) show that the proposed framework achieves up to 70% energy savings and a 37.6% improvement in coverage compared to prior research, with BE-WOA outperforming random, greedy, UBPSO, and ACS methods based on the cost function. The results show that integrating mobility and participation prediction leads to more reliable, energy-efficient, and high-quality task allocation for practical MCS deployments.</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

Energy-efficient task allocation in mobile crowdsensing using mobility and participation prediction

  • 期刊论文

  • 发表日期：

    2026年3月4日

查看全文
  • Seyed Reza Amerizadeh,
  • Mohammad Reza Khayyambashi

摘要

Mobile crowdsensing (MCS) faces fundamental challenges in task allocation due to unpredictable user mobility and uncertain participation, which can degrade sensing quality and increase energy waste. This work introduces the first unified task-allocation framework that jointly predicts users’ future movement paths and block-level participation rates addressing both challenges simultaneously. User mobility and participation are modeled in an offline phase using bidirectional GRU networks, and the predictions guide task assignment in the allocation phase. Three population-based metaheuristic algorithms (BE-WOA, UBPSO, and ACS) are employed to maximize data quality while minimizing users’ energy consumption. Experiments on 11 task-allocation scenarios using two real-world datasets (110 simulation runs) show that the proposed framework achieves up to 70% energy savings and a 37.6% improvement in coverage compared to prior research, with BE-WOA outperforming random, greedy, UBPSO, and ACS methods based on the cost function. The results show that integrating mobility and participation prediction leads to more reliable, energy-efficient, and high-quality task allocation for practical MCS deployments.