Integrated Placement and Routing of Relay Nodes for Fault-Tolerant Hierarchical Sensor Networks

Two-tiered sensor networks have gained popularity in recent years, due to their ability to facilitate load-balanced data gathering, fault-tolerance as well as increased network connectivity and coverage. Using higher-powered relay nodes as cluster beads can lead to further improvements in network performance. It is important to determine an appropriate placement scheme for such relay nodes, in order to achieve specified coverage and connectivity requirements with as few relay nodes as possible. A significant amount of work has been done in this area in recent years. However, existing placement strategies typically do not consider energy dissipation due to routing and are not capable of optimizing the routing scheme and placement concurrently. In this paper, we propose an integrated integer linear program (ILP) formulation that determines the minimum number of relay nodes, along with their locations and a suitable communication strategy such that i) all sensor nodes are able to connect to at least k(g) relay nodes, ii) the upper tier relay node network is at least k(r)-connected and iii) the network has a guaranteed lifetime. We also present an intersection based scheme for creating the initial set of potential relay node positions, which are used by our ILP, and evaluate its performance under different conditions. Experimental results on networks with hundreds of sensor nodes show that our approach leads to significant improvements over existing energy-unaware placement schemes.