3D-CCD: a Novel 3D Localization Algorithm Based on Concave/Convex Decomposition and Layering Scheme in WSNs

摘要

Wireless sensor network (WSN) localization is an important issue which attracts significant research attentions in recent years. However, the majority of the current localization techniques are designed assuming 2D deployments, which obviously cannot meet the requirements in the practical application. In general, those techniques are difficult to be extended under the environment of 3D deployment. In this paper, we propose a novel 3D localization algorithm (named 3D-CCD) based on concave/convex decomposition and layer scheme in WSNs. The localization can be completed by the following steps. First, the deployment area of WSN is split horizontally into some logical layers according to the height of nodes. Then, in the individual layer, a concave/convex decomposition based technique is designed to rationally divide the whole irregular network into several sub-areas, resulting in minimizing the position error caused by concave/convex degree of the network land-form. At last, the RSSI distance and trilateration measurement are used to globally localize the unknown nodes. We conduct simulations based upon the model of mountain surface network. The simulation results show that our algorithm decreases the less position error by about 10%, 17%, and lowers calculation overhead by about 30%, 18%, typically than the state-of-the-art localization algorithm (i.e., SV [1], COLA [2]). Thus, our algorithm is more potential in practical WSNs which are the characteristic of limited energy and 3D deployment.