Organizational evolution of digital signal processing software development
2006
A base station, as a network element, has become an increasingly software-intensive system. Digital signal processing (DSP) software is hard real-time software that is a part of the software system needed in a base station. This article reports practical experiences related to organizing the development of embedded software in the telecommunication industry, at Nokia Networks. The article introduces the main factors influencing the development of DSP software and also compares the evolutionary process under study with both selected organizational models for a software product line and a multistage model for the software life cycle. We believe it is vitally important to formulate the organization according to the software architecture, and it is essential to have a dedicated development organization with long-term responsibility for the software. History shows that without long-term responsibility, there is no software reuse. In this paper we introduce a new organizational model for product line development. This new hybrid model clarifies long-term responsibilities in large software organizations with hundreds of staff members and formulates the organization according to the software architecture. Our case needs a couple more constraints to keep it in the evolution stage of the software life cycle. Thus, we extend the evolution phase in the multistage model to make it relevant for embedded, hard real-time software. Copyright © 2006 John Wiley & Sons, Ltd.
Keywords:
- Software peer review
- Computer engineering
- Software construction
- Systems engineering
- Social software engineering
- Package development process
- Software design description
- Backporting
- Software product line
- Software sizing
- Computer science
- Software development
- Software engineering
- Software measurement
- Personal software process
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
8
References
2
Citations
NaN
KQI