Uniquely identifying objects and items are becoming an issue with ever increasing importance on our path towards the internet of things (IoT). So far the technology for connecting unique identifiers (e.g. integer numbers, codes) to objects in radio frequency identification (RFID) was either based on EEPROM or fuse technologies. Both of them require additional process complexity during mass production, thus adding significant contributions to the basic costs of identification technologies. In addition, fuses' states' are easily being read out and thos do not provide any means for storing a secret value. A radically new approach towards identification - exploiting so called "Physically Uncloneable Functions (PUFs)" - is based on the unique variations of electrical characteristics inherently present in every material and appearing in every manufacturing process. These variations are usually not aimed for and only tolerated within certain limits, not to influence the specified behaviour and functionality of the produced devices. The idea to make use of the device specific uniqueness of material properties could bring a big improvement in terms of technology costs and logistics complexity avoiding time consuming programming steps as well as area overheads like today's non volatile memories. In addition these characteristics are supposed to be much less likely being maipulated and thus are adding an extra level of security to the ID technology. To go beyond the identification purpose secret values from PUFs may serve as key material to secure communications beween participants in the IoT. The PUCKMAES project is aiming at research on reliable, easy-to-extract, secure PUFs that can be used as a base material for device dependent individual secret key generation on silicon, thus pushing further the foundations of a secure internet of things.
|Tatsächlicher Beginn/ -es Ende||1/02/09 → 31/03/12|