On wireless network infrastructure optimisation for cyber-physical systems in future smart buildings

摘要

Today, most cyber-physical systems (CPS) in smart buildings require a wireless-based network infrastructure for sensing, communication, and actuation. The energy expenditure and hence battery lifetime of such a wireless network infrastructure depend heavily upon the placement of the base stations (BS). However, in indoor environments, BSplacement is challenging due to the impact of building structures and floor/wall separations. In this paper, we study the problem of jointly optimising BS placement and power control to prolong the battery lifetime of sensors in the CPS network infrastructure. It turns out that this problem can be formulated as a mixed-integer non-convex program (MINCP), which is NP-hard and difficult to solve especially for large-sized networks. To address this difficulty, we propose a novel efficient algorithm called expansion-clustering-projection-contraction (ECPC). Our theoretical analysis and numerical results show that ECPC achieves competitive solutions compared to the true optimal solutions obtained by the branch-and-bound (BB) method.