A Framework for Dynamic Optimization of Security and Performances

Implementing security solutions in embedded systems is challenging due to their intrinsic limitations on performances as well as on available energy. Furthermore, traditional security solutions, designed by assuming static environments, do not provide the ability to tail security to the current operational conditions of the system. Thus, the efficiency and the applicability of security solutions to embedded dynamic applications, such as the ones often used in wireless sensor networks, are often limited. In this paper we introduce a solution to this problem by proposing a framework in which gradual adaptation of security is provided: the system is able to adjust security and workload settings depending on the current operating conditions. The adaptation is performed moving through adjacent configurations. In this paper we discuss the policies that can be used to control the adaptations and we present the results obtained when implementing a case study on Sun SPOT nodes. The results show that the use of the framework increases the energy efficiency of the network nodes. Furthermore, they show the effects of different adaptation policies on the behavior of nodes.