Enabling machine learning‐based side‐chaining for improving QoS in blockchain‐powered IoT networks. Issue 4 (23rd December 2021)

• Record Type: Journal Article
• Authors: Vairagade, Rupali Sachin; S. H., Brahmananda
• Other Names: Cheng Xiaochun guestEditor.; Liu Zheli guestEditor.; Ning Yongsheng guestEditor.
• Is Part Of: Transactions on emerging telecommunications technologies. Volume 33:Issue 4(2022)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2021-12-23
• Subjects: Telecommunication -- Periodicals; 384.05
• DOI: 10.1002/ett.4433
• Languages: English
• ISSNs: 2161-5748

Abstract

Abstract: Problem: The Internet of Things (IoT) makes developers for integrating various collection of device, saving, and processing a huge amount of data, allowing new services for improving several personal and professional activities. But, the privacy problems occur with a large amount of data produced and therefore the solutions on the basis of blockchain with sidechain technology have been developed to overcome these problems. Aim: Blockchain powered IoT networks have high security levels, and are resilient to most network attacks. In order to deploy blockchains, a large set of hashing, encryption, and linking algorithms are implemented in the network. One of the recent developments is the sidechains and these sidechains reduce computational complexity during block creation, and thereby reduce the effect of high delay and high‐power consumption in the IoT network. But maintaining these sidechains requires an additional computational component, which adds to the computational complexity of the underlying system. To reduce the complexity of this maintenance while keeping the inherent advantages of side chaining networks, this article proposes a sidechain creation, updation, merging, and scanning processes. Methods: In this article, the sidechain creation is done by modified two‐way peg protocol and updated by hybrid delegated practical byzantine fault tolerance‐delegated proof of stake (DPBFT‐DPOS), merging via continuous network information analysis and scanning via graph‐based searching mechanism to improve the Quality of Service (QoS) network and security. Results: The proposed methodology achieves an accuracy and F 1 ‐score of about 98.6% and 99.5%, reduces end‐to‐end communication delay by 10% while increasing the energy efficiency by 15%, and improving the throughput by 15% when compared to other existing methods. Conclusion: The IoT provides the possibility of creating real time data and the combination of the IoT with blockchain moves beyond authorization and financial recording. Then the utilization of sidechain provide better communication among devices and better sophistication of the data processing. Abstract : Purpose : Blockchain powered IoT networks have high security levels, and are resilient to most network attacks. In order to deploy blockchains, a large set of hashing, encryption, and linking algorithms are implemented in the network. Nowadays blockchain‐based systems have faced many challenges related to security, scalability, and privacy, and so forth. Different approaches have been developed by the research and professional communities to overcome these drawbacks. Approach : One of the recent developments is the sidechains and these sidechains reduce computational complexity during block creation, and thereby reduce the effect of high delay and high‐power consumption in the IoT network. But, maintaining these sidechains requires an additional computational component, which adds to the computational complexity of the underlying system. To reduce the complexity of this maintenance while keeping the inherent advantages of side chaining networks, this article proposes a sidechain creation, updation, merging, and scanning processes. Originality : Here the sidechain creation is done by modified two‐way peg protocol and updated by hybrid delegated practical byzantine fault tolerance‐delegated proof of stake (DPBFT‐DPOS), merging via continuous network information analysis and scanning via graph‐based searching mechanism to improve the Quality of Service (QoS) network and security. Findings : The proposed methodology achieves an accuracy and F 1 ‐score of about 98.6% and 99.5%, reduces end‐to‐end communication delay by 10% while increasing the energy efficiency by 15%, and improving the throughput by 15% when compared to its single‐chained counterpart.