| Localization is an important aspect of Wireless Sensor Networks. Information regarding the position of the sensor nodes is not always known. Without the position information of the sensors, the data reported by the sensors is of little use. Various approaches have been used to perform localization using some information about the sensor node. Potential field approach for localization, using distance information has been successfully tested with satisfactory results.; However in case of noisy environment, the range measurements have greater inaccuracy. In such cases, localization using the above algorithm can provide some inaccuracy. To rectify such erroneous localization situations, Extended Kalman Filters are used to estimate the position. The Extended Kalman Filter has been used as the process for estimation of coordinates is a non-linear process. The EKF is a recursive filter which only needs the information from the previous state to predict the next state. |
