IMM-EKF Algorithm-Based Indoor Moving Target Localization

It is a challenging task to improve the accuracy of the mobile localization in LOS (line-of-sight) and NLOS (non-line-of-sight) mixed environment. When the MN (moving node) moves in indoor environment, due to the obstacles such as walls, doors, and furniture, the communicating signal between MN and ANs(anchor nodes) change between LOS and NLOS frequently and randomly, which has negative effect on the accuracy of MN location estimation. To guarantee the accuracy, a KF (Kalman filter) based IMM (interacting multiple model) is proposed to filter the measured distance under the LOS/NLOS mixed environment. Due to the different characteristic of ranging errors between LOS and NLOS, two parallel KFs with different parameters are employed in order to suit for LOS mode and NLOS mode, both of the mode probabilities are calculated by the mode likelihoods and history probabilities. The modes transition between LOS/NLOS modes is based on Markov chain and mode probabilities. The weighted mean of the two modes filtering results is taken as the estimated distance of IMM. Once the estimated distances are obtained, the EKF (extended Kalman filter) is applied to locate the MN. The simulation results demonstrate the IMM can significantly mitigate the positive range error and achieve high localization accuracy.