Ghost-Weight protocol: a highly efficient blockchain consensus for IoT

Blockchain technology, with its decentralized nature, offers a more secure, reliable, and transparent solution for the Internet of Things (IoT). However, directly deploying existing consensus protocols in many IoT environments can limit scalability, lead to high computational costs and delays, and fail to meet the high responsiveness required for network communication. Additionally, IoT devices vary widely in performance, and lower-performance devices participating as nodes in a decentralized network may not be able to meet the demands of the consensus process, resulting in consensus delays and decreased system stability. This proposal introduces an improved consensus protocol called Ghost-Weight, based on the original Ghost protocol, designed to accommodate the performance-heterogeneous IoT environment with frequent information exchanges. Ghost-Weight introduces the concept of a Dominant Block, which, through a mechanism called the Confinement Edge, incorporates forked blocks within the network in a graph-based manner. The ForkBlock Process then organizes these blocks, removes conflicting transactions, and ultimately achieves global consensus based on weight before publishing them to the network. Furthermore, the protocol proposes a node partitioning strategy, ensuring that devices of all performance levels can participate in the consensus process. Through experiments and theoretical comparisons, we demonstrate that this protocol achieves a 52% increase in throughput, a 23% improvement in transaction confirmation time, and a 56% higher efficiency in processing forked blocks compared to the original Ghost protocol. We also designed a series of experiments to showcase Ghost-Weight's success in resisting selfish mining attacks and balancing attacks. This proposal effectively addresses the challenge of deploying consensus mechanisms in performance-heterogeneous IoT environments with high responsiveness requirements. Compared to previous designs, it significantly improves network scalability, transaction speed, network efficiency, and overall reliability.