Abstract
In this paper, we design good security metrology to solve the problem when, in a network, there is a node accusing another one of misbehaving. This problem is not easy because bad nodes tend to use false accusations to disrupt the network and the result is disastrous. We set up a standard, namely the security ratings, and use it to resolve such accusations. We use approaches of negative-credit, and mixed-credit (positive-credit), respectively, to solve this problem. We exclude the use of public key infrastructure and use only symmetric ciphers and hash functions to reduce the computational overhead of the security metrol-. ogy. Our results prove to be practical and robust against node compromise. The communication and computational overhead also prove to be small and suitable for real world applications. © Springer-Verlag Berlin Heidelberg 2006.