Algorithm to Determine the Optimum Number of Smart Pseudo Random Patterns to Generate During Test Generation
Original Publication Date: 1988-Sep-01
Included in the Prior Art Database: 2005-Feb-15
Producing the optimum percent test coverage with the optimum number of smart pseudo random patterns during test generation using Enhanced Test Generation (ETG) is accomplished in the following. Percent test coverage is a function of the number of smart pseudo random patterns generated and tends to take the shape shown in the graph in Fig. 1. Prior to the knee the change in coverage Y' is large for a small change in patterns X', giving a large slope Y'/X'. Beyond the knee the change in coverage Y" is small for a large change in patterns X" giving a small slope Y"/X". Therefore, the optimum change in coverage with the optimum change in patterns occurs in the area of the knee of the curve giving the optimum slope.