| Popis: |
In recent years, wireless networks have presented a significant evolution in their technology. While first-generation networks, based on analog signalling, were targeted primarily at voice and data communications occurring at low data rates, we have recently seen the evolution of secondand third-generation wireless systems that incorporate the features provided by broadband networks (Garg, 2001). In addition to supporting mobility, broadband networks also aims to support multimedia traffic, with quality of service (QoS) assurance. Therefore the evolution from 2G to 3G wireless technology bears the promise of offering a wide range of new multimedia services to mobile subscribers (De Vriendt et al., 2002). In this context, handset devices are following this ongoing network evolution and taking advantage of this technological update to offer a broad variety of resources and applications to their users. In fact, handset development has evolved into a complex engineering process, mainly because of the recent network capabilities supported by new mobile communication and computer technology advances (memory speed and size, processing power, better resources for information delivery, etc.). This scenario has increased the demand on the test phase of handset development, which is required to apply more extensive and efficient evaluation procedures so that the final product meets the fast time-to-market goals and can compete in the global marketplace. While the number and complexity of tests are increasing, test centers need to decrease their test execution time. The quicker a specific handset is evaluated and delivered to the market, the better will be its chances to compete with other models. Therefore we have a contradiction: we need to increase the number of tests and decrease the test time. Furthermore, this contradiction can lead us to reduce the quality of our test processes. Test automation is one alternative to this emerging scenario, because it enables tests to be launched and executed without the need for user intervention. Thus, common delays and errors associated with the manipulation of test parameters by humans can be avoided. |
