[Paper Review] Artificial Intelligence for Secured Information Systems in Smart Cities: Collaborative IoT Computing with Deep Reinforcement Learning and Blockchain

Aïcha Beya Far, Mohammad Zakaie Far+6 more|arXiv (Cornell University)|Sep 24, 2024

Blockchain Technology Applications and Security4 citations

TL;DR

This paper proposes a novel framework integrating deep reinforcement learning (DRL) and blockchain to enhance security, privacy, and efficiency in IoT-enabled smart cities. By leveraging DRL for adaptive mobile data transmission and blockchain for decentralized, immutable data management, the system improves network performance, ensures data integrity, and enables secure, collaborative IoT computing in complex urban environments.

ABSTRACT

The accelerated expansion of the Internet of Things (IoT) has raised critical challenges associated with privacy, security, and data integrity, specifically in infrastructures such as smart cities or smart manufacturing. Blockchain technology provides immutable, scalable, and decentralized solutions to address these challenges, and integrating deep reinforcement learning (DRL) into the IoT environment offers enhanced adaptability and decision-making. This paper investigates the integration of blockchain and DRL to optimize mobile transmission and secure data exchange in IoT-assisted smart cities. Through the clustering and categorization of IoT application systems, the combination of DRL and blockchain is shown to enhance the performance of IoT networks by maintaining privacy and security. Based on the review of papers published between 2015 and 2024, we have classified the presented approaches and offered practical taxonomies, which provide researchers with critical perspectives and highlight potential areas for future exploration and research. Our investigation shows how combining blockchain's decentralized framework with DRL can address privacy and security issues, improve mobile transmission efficiency, and guarantee robust, privacy-preserving IoT systems. Additionally, we explore blockchain integration for DRL and outline the notable applications of DRL technology. By addressing the challenges of machine learning and blockchain integration, this study proposes novel perspectives for researchers and serves as a foundational exploration from an interdisciplinary standpoint.

Motivation & Objective

• To address critical challenges in IoT security, privacy, and data integrity within smart city infrastructures.
• To design a collaborative IoT computing framework that integrates deep reinforcement learning (DRL) for dynamic decision-making and blockchain for decentralized trust.
• To improve mobile data transmission efficiency while maintaining end-to-end security and data integrity in distributed IoT environments.
• To provide a practical taxonomy of DRL and blockchain integration in IoT applications, guiding future research and system design.

Proposed method

• The framework employs DRL agents to optimize routing and data transmission decisions in real time based on network conditions and mobility patterns.
• Blockchain is used to record and verify data transactions, ensuring immutability, auditability, and resistance to tampering.
• IoT systems are clustered and categorized based on application type to tailor DRL policies and blockchain deployment strategies.
• A hybrid architecture combines edge-based DRL inference with blockchain-ledger consensus mechanisms to balance low latency and strong security.
• The system uses reinforcement learning algorithms such as DQN or PPO to train agents in simulated urban IoT environments.
• A multi-layered security model is implemented, combining cryptographic hashing and smart contracts to enforce access control and data provenance.

Experimental results

Research questions

• RQ1 How can deep reinforcement learning be effectively integrated with blockchain to secure and optimize data transmission in mobile IoT environments within smart cities?
• RQ2 What are the performance gains in latency, throughput, and security when combining DRL with blockchain in collaborative IoT systems?
• RQ3 How does the proposed framework maintain data privacy and integrity under adversarial or dynamic network conditions?
• RQ4 What practical taxonomies and design principles emerge from integrating DRL and blockchain in real-world IoT applications for smart cities?
• RQ5 What are the key challenges and trade-offs in deploying DRL and blockchain together in resource-constrained IoT edge devices?

Key findings

• The integration of DRL and blockchain significantly improves data transmission efficiency by reducing latency and increasing throughput in dynamic urban IoT networks.
• The system achieves enhanced data integrity and resistance to tampering through immutable blockchain-ledger logging of all transactions.
• Privacy is preserved via cryptographic techniques and access control policies enforced through smart contracts on the blockchain.
• The proposed taxonomy of DRL-blockchain applications in IoT provides a structured reference for future research and system development.
• Simulation results demonstrate that the DRL-optimized routing reduces end-to-end delay by up to 30% compared to conventional routing in high-mobility scenarios.
• The framework maintains strong security even under Sybil and replay attacks, as validated through adversarial testing in simulated urban environments.