Blockchain based computing power sharing in urban rail transit: System design and performance improvement

With the development of urban rail transit (URT), many latency-sensitive and computationally intensive tasks arise. Edge computing can provide low-latency computing service in URT systems. Edge servers cannot always process all incoming computing tasks in a timely manner when operating independently due to limited computing power resources. They need to collaborate frequently through peer-to-peer offloads. However, it is challenging for the server to select the appropriate computing power resources and corresponding network connections to fulfill its performance and cost requirement. More importantly, edge servers are deployed and managed by different computing departments, putting the task offload process at risk. We propose a blockchain-based computing power sharing system to achieve secure and efficient computing power sharing in URT systems. The blockchain provides auditing and checking functions to guarantee the security of computing power resource sharing. We further propose a method to optimize the computing power sharing strategy and node selection strategy in the computing power sharing workflow. The numerical findings reveal that the proposed scheme provides significant improvements in both departmental utility and business processing capability.