Self-organizing maps for embedded processor selection

Abstract In this work we show the application of self-organizing maps for analysing large microprocessor selection tables, like those used for device selection. We explain how to apply this artificial neural network to extract relevant information from microprocessor device databases, showing some case studies as an illustration. The resulting ‘microprocessor maps’ can be a valuable tool for the design engineer in the task of selecting a specific microprocessor or microcontroller device for a particular embedded application. In addition, these ‘micromaps’ can be used for microprocessor market analysis.