Scholarlink

<p>The rapid growth of IoT-driven digital advertising requires secure, transparent, and intelligent frameworks to resolve issues like data privacy, fraud, and centralized control. This paper introduces a Federated Reinforcement Learning with Blockchain-based Smart Contracts (FRL-BSC) framework that integrates federated reinforcement learning, blockchain, and edge computing to enhance targeted ad delivery. In the proposed approach, IoT devices locally train reinforcement learning agents, while blockchain smart contracts manage access, ensure transparency, and secure immutable records of model updates. Edge nodes aggregate models, reducing latency and improving scalability in large-scale networks. Experimental results demonstrate that FRL-BSC improves security by 93.1%, increases campaign identification accuracy by 8%, lowers the false positive rate (FPR) by 3.8%, enhances local model accuracy by 92.8%, reduces latency by 38%, and improves revenue sharing by 94.9%. These outcomes establish FRL-BSC as a scalable, privacy-preserving, and fraud-resistant solution for next-generation IoT-based digital advertising ecosystems.</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

Blockchain-integrated AI framework for secure IoT-based digital advertising ecosystems

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年11月28日

查看全文
  • Yegen Ouyang,
  • Enwei Cao,
  • Bingfeng Liu

摘要

The rapid growth of IoT-driven digital advertising requires secure, transparent, and intelligent frameworks to resolve issues like data privacy, fraud, and centralized control. This paper introduces a Federated Reinforcement Learning with Blockchain-based Smart Contracts (FRL-BSC) framework that integrates federated reinforcement learning, blockchain, and edge computing to enhance targeted ad delivery. In the proposed approach, IoT devices locally train reinforcement learning agents, while blockchain smart contracts manage access, ensure transparency, and secure immutable records of model updates. Edge nodes aggregate models, reducing latency and improving scalability in large-scale networks. Experimental results demonstrate that FRL-BSC improves security by 93.1%, increases campaign identification accuracy by 8%, lowers the false positive rate (FPR) by 3.8%, enhances local model accuracy by 92.8%, reduces latency by 38%, and improves revenue sharing by 94.9%. These outcomes establish FRL-BSC as a scalable, privacy-preserving, and fraud-resistant solution for next-generation IoT-based digital advertising ecosystems.