Using quality attributes to identify and specify embedded system requirements

Embedded systems are present in nearly every aspect of life today including transportation, healthcare, and entertainment. Despite the prevalence of these systems, processes and resources specific to their development have, until recently, focused primarily on the software portion of the system. The purpose of this research is to identify the quality attributes of embedded systems as a whole and use them as the basis for creating a mechanism that can be used to identify and specify the requirements of embedded systems.;Software quality attributes (QAs) are one example of a commonly used software-based resource. QAs are used to describe and quantify the characteristics of how the functional requirements of a software product will behave when implemented. While software QAs do a good job of covering the quality characteristics of the software of an embedded system, they do not cover many that are required of the system as a whole, such as size, weight, and durability. As part of this research, embedded system trade studies from industry were evaluated to identify the QAs used to make design decisions. The list of QAs identified from the trade study evaluation was combined with existing industry standard lists of software QAs to create a list of QAs applicable to both the hardware and the software of embedded systems.;General scenarios, which are patterns for specific requirement scenarios, exist for several software QAs. They are helpful in identifying the requirements of software systems. Based on the results of the trade study evaluation, this research updates existing software QA general scenarios and develops new embedded system QA scenarios in order to create Embedded System QA General Scenarios. A list of Generic Questions for each General Scenario was also created.;The benefit of using the Embedded System QA General Scenarios was validated using a two part requirements identification activity for a simple embedded system. On average, the number of requirements identified by industry representatives increased by 104% using the Embedded System QA General Scenarios and the Generic Questions after first identifying requirements using traditional methods. The control group found an additional 17% on average.