Scholarlink

<p>Energy efficiency is a key challenge in ultra-dense networks (UDNs) due to increasing traffic demand and dense base station (BS) deployments. Conventional centralized approaches often face scalability limitations and delayed response to dynamic traffic conditions. To address these challenges, this paper proposes DEEPCHAIN, a decentralized framework that combines edge-based traffic prediction with blockchain-enabled coordination for adaptive BS control. In the proposed approach, each BS utilizes a hybrid Deep Belief Network–Long Short-Term Memory (DBN–LSTM) model to predict future traffic demand. Based on these predictions, BSs determine sleep, activation, and coverage adjustment actions, which are validated through a Practical Byzantine Fault Tolerance (PBFT)-based consensus mechanism to ensure coordinated operation. Simulation-based evaluation under varying network conditions indicates that DEEPCHAIN can improve energy efficiency and overall network performance compared to baseline approaches. The framework achieves up to 32–35% net energy savings, along with improvements in throughput and spectral efficiency, while maintaining high coverage levels. The prediction model also demonstrates a 4.1% reduction in prediction error compared to reference methods. These results suggest that integrating predictive edge intelligence with decentralized coordination can provide a scalable and energy-aware solution for next-generation cellular networks, although further validation with real-world datasets remains an important direction for future work.</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

DEEPCHAIN: Decentralized Energy-Efficient Predictive Framework for Next-Generation Networks

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年5月9日

查看全文
  • Nilakshee Rajule,
  • Mithra Venkatesan,
  • D. G. Bhalke

摘要

Energy efficiency is a key challenge in ultra-dense networks (UDNs) due to increasing traffic demand and dense base station (BS) deployments. Conventional centralized approaches often face scalability limitations and delayed response to dynamic traffic conditions. To address these challenges, this paper proposes DEEPCHAIN, a decentralized framework that combines edge-based traffic prediction with blockchain-enabled coordination for adaptive BS control. In the proposed approach, each BS utilizes a hybrid Deep Belief Network–Long Short-Term Memory (DBN–LSTM) model to predict future traffic demand. Based on these predictions, BSs determine sleep, activation, and coverage adjustment actions, which are validated through a Practical Byzantine Fault Tolerance (PBFT)-based consensus mechanism to ensure coordinated operation. Simulation-based evaluation under varying network conditions indicates that DEEPCHAIN can improve energy efficiency and overall network performance compared to baseline approaches. The framework achieves up to 32–35% net energy savings, along with improvements in throughput and spectral efficiency, while maintaining high coverage levels. The prediction model also demonstrates a 4.1% reduction in prediction error compared to reference methods. These results suggest that integrating predictive edge intelligence with decentralized coordination can provide a scalable and energy-aware solution for next-generation cellular networks, although further validation with real-world datasets remains an important direction for future work.