IEEE Access (Jan 2022)

CTB-PKI: Clustering and Trust Enabled Blockchain Based PKI System for Efficient Communication in P2P Network

  • Amrutanshu Panigrahi,
  • Ajit Kumar Nayak,
  • Rourab Paul,
  • Bibhuprasad Sahu,
  • Shashi Kant

DOI
https://doi.org/10.1109/ACCESS.2022.3222807
Journal volume & issue
Vol. 10
pp. 124277 – 124290

Abstract

Read online

The decentralization feature of public and private blockchain-based applications is achieved by selecting different nodes as validators or Certificate Authority ( $CA$ ) for each transaction. Public blockchain uses Proof of Work ( $PoW$ ) to search for the validator. $PoW$ causes an enormous amount of energy. Therefore, Proof of Stake ( $PoS$ ), and Proof of Authority ( $PoA$ ) emerged as alternate solutions. Selection of a new $CA$ using $PoS$ or $PoA$ algorithms for each transaction may improve transaction security. However, a network may have a large number of transactions and participants. Selecting a $CA$ for each transaction using $PoS$ or $PoA$ may cause a significant amount of block propagation delay, which can reduce network efficiency drastically. This paper proposes a different approach to increase the efficiency of Blockchain-based Public Key Infrastructure ( $BC-PKI$ ). The proposed approach creates clusters of participant nodes based on their validation time, response time, and trust. This method selects a cluster based on the budget of response time and validation time given by the node that intends to start a transaction. Thereafter, the node which has the highest trust in that cluster is chosen as a $CA$ for the next transaction. Instead of searching on all participant nodes, our approach searches on the nodes of the chosen cluster which reduces the searching space of the $CA$ selection process. This research work adopts a trust evaluation approach where the trust factor is quantified based on its experience and reputation. The node trust is reevaluated after every successful and unsuccessful transaction. A node that performs more successful transactions has more trust value. The node that has a higher trust value has a higher probability to be selected as a $CA$ for a transaction. The trust reevaluation process is followed by the clustering process. The result shows the proposed approach can reduce ~38.5% response time and ~2.2% validation time as compared to infrastructure which does not implement clustering. Additionally, the proposed $CTB-PKI$ can be used in Blockchain 2.0 and Blockchain 3.0-related applications.

Keywords