Polarization diversity improves RSSI based location estimation for wireless sensor networks

Abstract

Signal strength based wireless location estimation suffers accuracy in indoor environments since the distance to RSSI (received signal strength indicator) function is not monotonic and gives multiple distance estimates for a specific signal strength value. In this paper an advancement of the existing estimation techniques is proposed by establishing the use of polarization diversity in getting robust location estimation. The measurements done in indoor environment clearly bring to fore the need to compute the signal strengths on different polarization vectors of radiated E-field, for prediction of location of wireless communication nodes with a better accuracy. Experiments conducted for both line-of sight and non-line of sight cases show that an optimal weighted sum of the respective signal strengths for different polarizations lead to a monotonic path loss expression which can be used for predicting the locations even with minimal initial training sequence. This novel approach enables precision indoor localization with a minimal set of measurements by exploiting the easily available RSSI information.