Abstract
Within the context of the Internet of Things (IoT) and the Location of Things (LoT) service, this paper presents an interactive tool to quantitatively analyze the performance of cooperative localization techniques for wireless sensor networks (WSNs). In these types of algorithms, nodes help each other determine their location based on some signal metrics such as time of arrival (TOA), received signal strength (RSS), or a fusion of them. The developed tool is intended to provide researchers and designers a fast way to measure the performance of localization algorithms considering specific network topologies. Using TOA or RSS models, the Crámer-Rao lower bound (CRLB) has been implemented within the tool. This lower bound can be used as a benchmark for testing a particular algorithm for specific channel characteristics and WSN topology, which allows determination if the necessary accuracy for a specific application is possible. Furthermore, the tool allows us to consider independent characteristics for each node in the WSN. This feature allows the avoidance of the typical “disk graph model,” which is usually applied to test cooperative localization algorithms. The tool allows us to run Monte-Carlo simulations and generate statistical reports. A set of basic illustrative examples are described comparing the performance of different localization algorithms and showing the capabilities of the presented tool.
Highlights
Internet of Things (IoT) is a paradigm intended as a network of a very large number of sensor nodes deployed over physical space
The Crámer-Rao lower bound (CRLB) has been implemented in wsnLocalize, providing researchers insight about the behavior of a specific localization algorithm given a specific network topology
Node estimated and true locations (b). This plot shows the absolute positioning of the nodes and depends on the localization algorithm applied and all the parameters that affect to the pairwise distance estimations (WSN topology, average connectivity, etc.)
Summary
Internet of Things (IoT) is a paradigm intended as a network of a very large number of sensor nodes deployed over physical space These sensors form a wireless sensor network (WSN) and have capabilities to monitor the physical environment and collect and report data for a specific application. Simulating the performance of a localization algorithm in a specific scenario can help designers define the minimum sensor requirements for a set of positioning applications. We present an interactive simulation tool, named wsnLocalize, for testing different localization algorithms in WSNs, those that belong to cooperative localization techniques, which allow sensors that are not in range of known-location devices ( called anchors) to be located by means of the estimation of pairwise distances. An appendix is included containing the main stress functions used for the performance analysis of the localization algorithms
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
