Test Bus Assignment of System on Chip by Using Cultural Particle Swarm Optimization

The system on chip (SoC) must be tested in order to insure the correctness of functions. The test of SoC involves the test of cores and the test of interconnects among cores. The test set of cores must be transported to the inputs of cores through the test access architecture in SoC. Therefore, in order to make the total test time to be minimal, it is necessary to determine an assignment of cores to test access architecture. In this paper, a new method is presented for test bus assignment of system on chip, which is based on cultural particle swarm optimization. In this method, the feasible solutions of test bus assignments are represented by the individuals. The better schemes of test bus assignments are obtained by the evolution of the individuals. The implementation of the method consists of a population space, a belief space, and a communication protocol that describes the exchange mode of knowledge between the population space and belief space. The particle swarm optimization with selection strategy is used for carrying out the evolution of individuals in population space. A lot of experimental results for SoC benchmark circuits demonstrate that the proposed method in this paper can effectively obtain the test bus assignments in shorter time than conventional genetic algorithms.