Blockchain has recently been applied in many applications such as bitcoin,smart grid, and Internet of Things (IoT) as a public ledger of transactions.However, the use of blockchain in mobile environments is still limited becausethe mining process consumes too much computing and energy resources on mobiledevices. Edge computing offered by the Edge Computing Service Provider can beadopted as a viable solution for offloading the mining tasks from the mobiledevices, i.e., miners, in the mobile blockchain environment. However, amechanism needs to be designed for edge resource allocation to maximize therevenue for the Edge Computing Service Provider and to ensure incentivecompatibility and individual rationality is still open. In this paper, wedevelop an optimal auction based on deep learning for the edge resourceallocation. Specifically, we construct a multi-layer neural networkarchitecture based on an analytical solution of the optimal auction. The neuralnetworks first perform monotone transformations of the miners' bids. Then, theycalculate allocation and conditional payment rules for the miners. We usevaluations of the miners as the data training to adjust parameters of theneural networks so as to optimize the loss function which is the expected,negated revenue of the Edge Computing Service Provider. We show theexperimental results to confirm the benefits of using the deep learning forderiving the optimal auction for mobile blockchain with high revenue
展开▼
