A new differential evolution algorithm for joint mining decision and resource allocation in a MEC-enabled wireless blockchain network

This paper studies a mobile edge computing-enabled wireless blockchain network, in which a set of Internet of Things (IoT) devices can act as miners to participate in mining. In this blockchain network, we jointly optimize the mining decision and resource allocation to maximize the total profit of all miners. When using evolutionary algorithms to solve this problem, each individual usually represents the mining decisions and resource allocations of all miners, which results in the redundant search space due to the fact that not all miners participate in mining. In this paper, we propose a new differential evolution (DE) algorithm, called DEMiDRA. In DEMiDRA, each individual represents the resource allocation of a participating miner and the resource allocations of all participating miners constitute the whole population. Then, DE is adopted to optimize the resource allocation. As for the optimization of the mining decision, we need to select miners to participate in mining and update the number of participating miners. Since the population size is equal to the number of participating miners, we transform the update of the number of participating miners into the adjustment of the population size and design an adaptive strategy. Besides, a tabu strategy is developed to prevent unpromising miners from participating in mining. The effectiveness of DEMiDRA is verified by comparing it with three other algorithms on a set of instances.