中文摘要：

卡尔曼滤波算法是用来解决定位中滤波的问题的一个重要内容,但由于预测和测量值之间的误差比较大,算法并没有达到最优,因为在室内定位中温湿度（高斯白噪声）对其有影响,以及非平面中的位置信息影响人员物品的位置定位精确度。针对卡尔曼滤波算法的这一问题,引进模拟退火算法。结合模拟退火算法的降温思想,采用迭代选取最优解,以此为基础,得到的最优解用于卡尔曼的初始值;将得到的最优距离作为对象,并以此建立邻域,最后再用线性插值法得到坐标。仿真实验表明,此种方法有效提高了室内定位精确度,减小降低了各种因素的干扰。

英文摘要：

Kalman filtering algorithm is used to solve the problem of the positioning filter, however the error isrelatively large between the measured value and predicted value. The algorithm does not reach the optimal solution becausethere exists temperature and humidity （we can call it Gaussian white noise） as well as the location information of non-planar affects positioning accuracy of human＇s site in the indoor location. We introduce the simulated annealing algorithmto the significant difference of Kalman filtering algorithm. We combine the cooling of simulated annealing algorithm anduse the iterative to select the optimum solution. On this basis, we can get the optimal solution for initial value of theKalman. Getting the optimal as an object and we set up neighborhood based on the object. At the end we can getcoordinate by linear interpolation method. Simulation experiments show that this method can improve effectively thepositioning accuracy and reduce the interference of various kinds of factors.