Connectivity-Based Joint Parameter Estimation in One-Dimensional Wireless Sensor Networks

In this paper, we propose a novel joint estimation scheme for the range, path-loss exponent (PLE), and inter-node distances based on the received signal strength (RSS) and the network's information of an anchor-less (i.e., without anchor nodes) curvilinear wireless sensor network (CLWSN). Assuming a random node distribution along a one-dimensional curve (e.g., in deployments over gas/oil/water pipelines, railway tracks, underground mine tunnels, subway networks, city sewage networks, street/road lights, etc.), and adopting a propagation model that combines both large-scale PLE and log-Normal shadowing, we derive new analytical expressions for the node's communication range as a function of the network's connectivity and density and for the PLE as a function of the node's range and both the network's connectivity and density. Once we calculate the node's range and the PLE, we estimate all distances between nodes using an RSS-based method. We illustrate by simulations the superior accuracy of our new joint range, PLE, and distance estimation technique against state-of-the-art benchmarks in terms of the normalized mean absolute error (NMAE). We thereby highlight the major contributions of this work by demonstrating both: i) the ability of our new solution to jointly estimate the node's range, the PLE, and all inter-node distances with relatively high accuracy using solely the RSS and network's connectivity and density; and ii) its significant potential for enabling very cost-effective yet highly accurate positioning over anchor-less (e.g., in GPS-denied harsh environments) wireless sensor networks (WSN)s.