Scholarlink

<p>In large-scale IoT and mobile networks, broadcasting is an important method of data dissemination to present real-time information efficiently to many devices. This paper introduces a new ML-Based Adaptive Broadcast Scheduling (ML-ABS) framework. The proposed framework combines mobility prediction, zone-aware indexing, and secure indexing for data broadcasting. By using LSTM-based mobility prediction and Random Forest zone classification, the scheduler dynamically allocates bandwidth and prioritizes data dissemination based on the predicted zone density. We conducted experiments using CRAWDAD mobility datasets. The results show that the LSTM model consistently outperforms Random Forest, achieving accuracies between 89% and 93%. The ML-ABS scheduler achieves the lowest tuning time and reduces client-side access delay by up to 34% compared to static scheduling and 25% compared to round-robin broadcasting.</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

ML-Based Location Prediction and Adaptive Broadcast Scheduling in IoT Networks

  • 期刊论文

  • 发表日期：

    2026年5月9日

查看全文
  • Vikas Goel,
  • Amit Kumar Goyal,
  • Narendra Kumar,
  • Mayank Singh

摘要

In large-scale IoT and mobile networks, broadcasting is an important method of data dissemination to present real-time information efficiently to many devices. This paper introduces a new ML-Based Adaptive Broadcast Scheduling (ML-ABS) framework. The proposed framework combines mobility prediction, zone-aware indexing, and secure indexing for data broadcasting. By using LSTM-based mobility prediction and Random Forest zone classification, the scheduler dynamically allocates bandwidth and prioritizes data dissemination based on the predicted zone density. We conducted experiments using CRAWDAD mobility datasets. The results show that the LSTM model consistently outperforms Random Forest, achieving accuracies between 89% and 93%. The ML-ABS scheduler achieves the lowest tuning time and reduces client-side access delay by up to 34% compared to static scheduling and 25% compared to round-robin broadcasting.