A CRCLA Task Partition Algorithm Combining Genetic Algorithm and Clustering Based Partitioning Algorithm

In order to find the global optimal solution of the task partition results of data flow graph(DFG) based on reconfigurable cryptographic logic array, a task partition algorithm CBPGA combining genetic algorithm and cluster partition algorithm is proposed. The algorithm is first improved on the basis of the CBP algorithm, combined with the hardware structure of the array and the particularity of the cryptographic algorithm operation, the probability-based absorption method is adopted for the absorption of the node operator to generates initial population, and then use genetic algorithm(GA) to solve the global optimal solution. Using the number of divided blocks and communication cost as evaluation indicators, experiments show that the algorithm has a certain improvement compared to CBP algorithm and LSCBP algorithm.