Joint Traffic Routing and Virtualized Security Function Activation in Wireless Multihop Networks

Abstract

Decentralized security function deployment (activation) is an important solution to achieve reliable and secure operations of wireless networks. In this paper, we consider the joint problem of security function activation and traffic routing in a wireless multihop network with the aim of minimizing total energy consumption of all nodes in the network while meeting the security requirements of the users. Firstly, we analyze the tradeoff of energy consumption between preventing security damages with protection functions and remedying it with recovery functions. The analysis takes the uncertainty of attacks and network capacity availability into account. Secondly, if the protection function is to be activated, the activation has to be performed along the route of traffic flows, which means that the traffic routing should also be jointly optimized. Therefore, we formulate this problem as a stochastic optimization model. To obtain the optimal solution, we apply a distributed algorithm, i.e., accelerated distributed augmented Lagrangian algorithm which is ensured to converge. Furthermore, we analytically demonstrate that the solution obtained by the accelerated distributed augmented Lagrangian algorithm is optimal to our stochastic optimization model. Based on the setting from real experiments, the performance evaluation reveals that the optimal solution depends largely on the energy budget and energy consumption of the nodes for transferring traffic and running security functions.