Stochastic geometric analysis of black hole attack on smart grid communication networks

Abstract

In this article, we develop a stochastic geometric framework for the performance analysis of a large scale smart grid communication network. Our proposed model caters for both topological and channel dynamics. More specifically, we consider a smart grid communication network where an arbitrary smart meter communicates with the metering head-end in a multi-hop manner. Spatial configuration of data aggregation points, which act as relays for the smart meter transmission, is captured using a homogeneous Poisson point process. Optimization of coverage by adaptation of a device level parameter such as transmit power and/or a network level parameter such as aggregation point density is also briefly discussed. The proposed framework is employed to quantify the performance degradation encountered in the presence of malicious black hole attackers. It is shown that the performance degradation can be measured in terms of “welfare loss” or equivalently in terms of outage probability difference. The cross layer approach adapted in this paper demonstrates that the end-to-end outage probability in the presence of attackers can be minimized by adapting the desired signal to noise ratio threshold or equivalently the transmission rate.