A Genetic Algorithm for Task Offloading of Edge Computing towards Meteorological Radar Data Computing Scenario

We propose a genetic algorithm for task offloading of meteorological radar data computing with Edge Computing (EC) in this paper. The increasing volume of radar data and the complexity of computing tasks make it necessary to offload the computation to remote computing resources. The proposed Randomized Greedy Algorithm (RA) and Group Genetic Algorithm are designed to determine the optimal task offloading strategy, which considers the trade-off between computation cost and communication overhead. The algorithm uses a fitness function that considers the task transmission delay, and data loss rate. The algorithm also employs crossover and mutation operations to generate new offloading strategies and improve the system's overall performance. We evaluate the performance of the proposed algorithm using simulation experiments and demonstrate that it can effectively reduce the computation offloading delay and data loss rate, which is compared with the RA task distribution