In order to let Ambient Intelligence (AmI) become a reality, efficient implementation of the required AmI functionality - especially "ubiquitous computing" and "ubiquitous communication" - is of central importance. The key driving technology of AmI is therefore microelectronic technology as the implementation technology.
AmI encompasses systems with very diverse requirements regarding bandwidth and calculation complexity. These may vary by several orders of magnitude depending on the application. The same is true for performance consumption. Energy efficiency, in particular, (= required computing power in proportion to performance consumption) requires innovative new system and architecture concepts accompanied by new design methodologies. Target sizes for portable AmI assistants, e.g., are in the range of 10-50 Mops/mW, for Smart Sensors in that of 100 Mops/mW. This means that energy efficiency is improved by two orders of magnitude when compared to the state-of-the-practice. Even for hard-wired solutions this requirement will only be achievable with 65 nm technologies. Furthermore, AmI systems require great flexibility, since various applications must be executed on the same hardware. Flexibility and low energy consumption, however, are contradictory design goals. New architecture approaches are therefore needed to enable efficient signal processing with great energy efficiency and flexibility at the same time. These are modular multi-processor architectures whose energy consumption is optimized via intelligent operating systems. Progress in these areas will have direct effects on such application areas as health telematics, but also on Smart Buildings and Smart Cars.
Another challenge lies in the heterogeneity of AmI systems (sensors, actuators, mixed signal and signal processing components on the same chip or in the same package). This affects both the design methodology and the cost-efficient production of such System-on-Chips (SoC) resp. System-in-Packages (SiP). This progress will result in technologies for continuous process control in manufacturing areas (all the way to integration in Supply Chain Management) becoming affordable even for SMEs.