| Abstrakt: |
Blockchains are a secure alternative for long-length data storage & security deployments. Though blockchains are unbounded, their quality of service (QoS) performance reduces after adding a certain number of blocks. Thus, sidechains have become essential for making the system decentralized, secure, and effective for use thereby improving their scalability & QoS performance. Sidechains reduce data storage and extraction delays. However, all sidechains of a single blockchain are the same in size, capacity, and security. This limits their application to real-world use cases that require dynamic security. To overcome this limitation, the authors propose a meta-heuristic approach to design a system that produces customized sidechains based on the quantity and quality of data stored on the chain. The model uses a machine-learning approach to find the best possible sidechain configuration for different data types. It makes the system fast and scalable and improves storage & memory efficiency. The proposed model is tested on multiple data sets and compared with various state-of-art sidechain deployments. An improvement of 18% in terms of mining speed, 27% in terms of energy efficiency, 10% with regards to throughput, and 8% concerning packet delivery ratio is observed. The model is tested under various attacks and faulty nodes to validate its security performance. A consistent QoS performance is observed for the proposed model under these attack types, thereby validating its resiliency for different attacks. This enhancement in performance makes the proposed model eligible for deployment in high-speed, low-energy, and high-security applications like IoT, mobile ad-hoc networks, and sensor networks. [ABSTRACT FROM AUTHOR] |
