Title
Blockchain and AI Empowered Trust-Information-Centric Network for Beyond 5G
Abstract
As the next-generation network, beyond fifth generation (B5G) provides transmission capability up to terabits and processes hundreds of exabytes of content data per day from the internet of Everything. From 5G to B5G, the information-centric network (iCN) is expected to play a vital role due to the strong capabilities of content distribution, caching, and processing. As security is a major concern in B5G, content trust of iCN is of critical importance. Lack of content trust leads to the untrustworthiness and maliciousness of services and applications in B5G, such as malicious accidents resulting from the untrusted content of vehicle navigation and autonomous system. To deal with this issue, we propose a blockchain and artificial intelligence (Ai) empowered trust-information- centric network architecture for B5G. First, we design a blockchain-based trust evaluation and circulation scheme for B5G nodes called TrustCoin, which quantifies the credibility of B5G nodes in a dynamic and fine-grained way, and manages trust quotas of B5G nodes as well as trust-coin circulation. Second, to obtain the content credibility, we devise a credibility decision method based on content status and B5G nodes' behaviors by exploiting the excellent properties of deep reinforcement learning, which provides the intelligent allocation criterion for TrustCoin. Third, we propose a smart incentive mechanism for the endogenous trust of B5G networks according to the allocation criterion, thereby establishing the trust-information-centric network. Experimental results have verified the effectiveness of our proposed mechanism.
Year
DOI
Venue
2020
10.1109/MNET.021.1900608
IEEE Network
Keywords
DocType
Volume
content status,B5G nodes,endogenous trust,next-generation network,iCN,artificial intelligence,blockchain-based trust evaluation,trust quotas,trust-coin circulation,content credibility,smart incentive mechanism,trust-information-centric network architecture,deep reinforcement learning,TrustCoin
Journal
34
Issue
ISSN
Citations 
6
0890-8044
3
PageRank 
References 
Authors
0.39
0
5
Name
Order
Citations
PageRank
Qianqian Pan161.09
Jun Wu245675.01
Jian-hua Li355898.16
Yang Wu46922.62
Zhitao Guan516622.58