A systematic review on the engineering of software for ubiquitous systems

• Published in: The Journal of systems and software 2016-08, Vol.118, p.251-276
• Main Authors: Sánchez Guinea, Alejandro, Nain, Grégory, Le Traon, Yves
• Format: Article
• Language: English
• Subjects: Computer programs; Development methods; Empirical software engineering; Evidence-based software engineering; Feedback; Life cycle analysis; Life cycles; Literature reviews; Pervasive systems; Phases; Research synthesis; Software; Software development; Software engineering; Software development cycle; Studies; Systematic review; Ubiquitous systems
• ISSN: 0164-1212; 1873-1228
• DOI: 10.1016/j.jss.2016.05.024

•A systematic literature review on engineering software for ubiquitous systems.•We identified 132 approaches addressing issues on different phases of the software engineering cycle for ubiquitous systems.•Implementation, evolution/maintenance, and feedback phases have been the most studied.•The testing phase needs to receive more attention, especially in what respect to simulations. Context: Software engineering for ubiquitous systems has experienced an important and rapid growth, however the vast research corpus makes it difficult to obtain valuable information from it. Objective: To identify, evaluate, and synthesize research about the most relevant approaches addressing the different phases of the software development life cycle for ubiquitous systems. Method: We conducted a systematic literature review of papers presenting and evaluating approaches for the different phases of the software development life cycle for ubiquitous systems. Approaches were classified according to the phase of the development cycle they addressed, identifying their main concerns and limitations. Results: We identified 128 papers reporting 132 approaches addressing issues related to different phases of the software development cycle for ubiquitous systems. Most approaches have been aimed at addressing the implementation, evolution/maintenance, and feedback phases, while others phases such as testing need more attention from researchers. Conclusion: We recommend to follow existing guidelines when conducting case studies to make the studies more reproducible and closer to real life cases. While some phases of the development cycle have been extensively explored, there is still room for research in other phases, toward a more agile and integrated cycle, from requirements to testing and feedback.