loading page

Integrating Edge Intelligence and Blockchain: What, Why, and How
  • +3
  • Xiaofei Wang ,
  • Xiaoxu Ren ,
  • Chao Qiu ,
  • Zehui Xiong ,
  • Haipeng Yao ,
  • Victor C.M. Leung
Xiaofei Wang
Author Profile
Xiaoxu Ren
Tianjin University

Corresponding Author:[email protected]

Author Profile
Zehui Xiong
Author Profile
Haipeng Yao
Author Profile
Victor C.M. Leung
Author Profile


Driven by an unprecedented boom in artificial intelligence (AI) and Internet of Things (IoT), edge intelligence (EI) pushes the frontier of AI from cloud to network edge, serving as a remarkable solution that unlocks the full potential of AI services. It is yet facing critical challenges in its decentralized management and security, limiting its capabilities to support services with numerous requirements. In this context, blockchain (BC) has been seen as a promising solution to tackle the above issues, and further support EI. Based on the number of citations or the relevance of emerging methods, this paper presents the results of a literature survey on the integration of EI and BC. Accordingly, we summarize the recent research efforts reported in the existing works on EI and BC. We then paint a comprehensive picture of the limitations of EI and why BC could benefit from EI. From there, we explore how BC benefits EI in terms of computing power management, data administration, and model optimization. In order to narrow the gap between immature BC and EI-amicable BC, we also probe into how to tailor BC to EI from four perspectives, including flexible consensus protocol, effective incentive, intellectuality smart contract, and scalability. Finally, some research challenges and future directions are presented. Different from existing surveys, our work focuses on the integration of EI and BC, develops some general models to help the reader build relevant optimization models in the integrated system, as well as provides detailed tutorials on implementation. We anticipate that this survey will motivate further discussions on the synergy of EI and BC, and offer some guidance in EI, BC, future networks, and other areas.
2022Published in IEEE Communications Surveys & Tutorials volume 24 issue 4 on pages 2193-2229. 10.1109/COMST.2022.3189962