Tailored security and safety for pervasive computing

Pervasive computing makes high demands on security: devices are seriously resource-restricted, communication takes place spontaneously, and adversaries might control some of the devices. We claim that 1.) today's research, studying traditional security properties for pervasive computing, leads to inefficient, expensive, and unnecessary strong and unwanted security solutions. Instead, security solutions tailored to the demands of a user, the scenario, or the expected adversary are more promising. 2.) Today's research for security in pervasive computing makes naive, inefficient, and unrealistic assumptions regarding safety properties, in particular the quality of basic communication. Therefore, future security research has to consider safety characteristics and has to jointly investigate security and safety for efficient, tailored solutions.