Abstract: In order to obtain higher fault coverage with shorter sequence length and no additional hardware overhead,we presented a weighted random test approach of digital integrated circuits based on Cellular Automata (CA). The approach adopted probabilistic testability measure to establish the Testability Cost Function (TCF) which reflects the cost of fault diagnosis.By an optimization procedure for TCF,weights at primitive inputs of circuit under test are obtained.With the aid of one dimensional hybrid CA,random sequence corresponding to the weight is real-ized.Experimental results on benchmark circuits prove the effectiveness and convenience ofthe proposed test generation methodology for BISTapplication.