An Incremental Bioinspired-learning Model for integrating Dynamic Security in Blockchain-based distributed Cloud deployments
Blockchain-based distributed cloud deployments generally use static methods to model encryption, hashing, consensus and sharding techniques. Existing dynamic security models for blockchains either have higher complexity, or lower performance efficiency when applied to real-time deployments. To overcome these issues, this text proposes design of an incremental bioinspired-learning model for integrating dynamic security in blockchain-based distributed cloud deployments. The proposed model uses Q-Learning to dynamically select different encryption & hashing. This assists in dynamically modifying security performance for different data-level attacks. Model proposes use of a novel Proof-of-Distributed-Cloud-Performance-Trust (PoDCPT), which integrates temporal execution performance of distributed Virtual Machines (VM) for selection of optimal miner nodes during consensus. These miner nodes are selected based on computational delay, execution efficiency of VMs, and their temporal mining performance under different attacks. The selected miner nodes assist in adding new blocks to the sharded chains. Sharding configurations of these chains are selected using hybrid Ant Lion Optimization (ALO) with Teacher-Learning?based-Optimization (TLbO) process. This hybrid combination assists in formation of delay & complexity-aware sidechain configurations. This enables the model to reduce computational delay by 3.5%, reduce energy consumption by 5.4%, improve throughput by 8.3%, while enhancing mining efficiency by 4.5% under different cloud attacks.
Email Address of Submitting Authorhimanshu.firstname.lastname@example.org
Submitting Author's InstitutionGovernment College of Engineering, Amravati, India
Submitting Author's Country