Embedded program testing in untestable mobile environment: An experience of trustworthiness approach

Comparing actual output with the expected output of some controlled input is a fundamental principle of program correctness testing. However, in some situations, the input is uncontrollable or even undetectable during testing and, hence, it is impossible to decide the expected output or the test oracle. We encountered this problem when we developed programs to extract network data from various mobile stations in the Mobile Location Estimation System project. We propose testing the trustworthiness of the programs instead. Since the input is uncontrollable and undetectable, program output is analyzed and challenged against with the intrinsic properties, environment, the program output itself and their application results, to find evidence that the output is suitable to be used for the planned purposes. Furthermore, in the case of mobile software development, it is common that different programs of the same specification have to be developed for mobile stations of different models. These different implementations provide another source of reference for trustworthiness tests. Our experience of applying trustworthiness test in extracting network data from mobile stations is reported in this paper.