Abstract
Localization in Industrial Wireless Sensor Networks (IWSNs) promotes innovations in manufacturing applications, such as structural status mapping, instrument fault diagnosing and oriented automation system associating. However, dispersive distortion, impulse noise and interference effects causing unpredictable and time-variant effects of the signal, decrease the localization accuracy in harsh manufacturing environments. In this paper, we propose a noise reduction localization scheme in IWSNs, called Support Vector Semidefinite (SVSD), based on practical industrial wireless channel measurements. We introduce an e-insensitive error function to evade the effects of impulse noise and interference by applying a new statistical path-loss model obtained from the measurements. We further relieve the noise effects by estimating the boundaries of the sensor locations before addressing the localization. Considering the boundaries, we obtain the sensor locations by utilizing semidefinite programming (SDP) relaxation. Simulation results show that the SVSD scheme provides higher location estimation accuracy than the SDP scheme under varying noise effects.
Sign-in/Register to access full text options 
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
