Self-Localization Algorithms for Wireless Sensor Networks in the Smart City

We deploy simulatively a network of unattended sensor transceiver nodes with known locations in an urban environment, and investigate the ability to localize the position of a mobile unit based on its RF transmission. The radio propagation is simulated using ray-tracing, involving reflection and diffraction from buildings, to model the multipath phenomenon, which is the origin of localization errors. The algorithms which are examined and evaluated are based on the sensors’ positions, RSSI and AoA. The localization algorithm’s performance is presented by means of probabilistic range error. Additionally, Pareto front analysis reveals the impact of reducing the number of sensors on the algorithm’s product. Furthermore, sensitivity analysis for the location of sensors on the buildings and the density of the buildings in the urban environment are demonstrated.