Collinearity for sensor network localization

In the context of wireless sensors networks, the knowledge of the device positions with the minimum possible error is required. In this paper we estimate sensor node positions throughout the measurement of the received powers from nodes characterized too have a well-know positions, called beacons. The estimation is performed by using an approximated maximum likelihood strategy, where the beacons are selected by means of two estimation mechanisms, involving a new parameter called normalized collinearity. This merit figure indicates to the sensor how much the terns of beacons, used as reference, are aligned with respect to the average distance from the sensor. By means of simulations and using the normalized collinearity to choose the best available tern for the sensor position estimation, we have found that terns which have higher values of normalized collinearity (that means terns less aligned) give the best position estimation.