The Research Of LANDMARC Positioning System And Its Algorithm

Abstract:With the rapid development of mobile communications and wireless technologies,there are more and more technologies and applications that is relate to location. For outdoor positioning, GPS and other global satellite positioning system, wireless cellular positioning systems and other ground-based navigation system has been widely used in all sides of life. But in indoor positioning system, although there are many positioning technologies,such as ZigBee, Infrared,Ultrasound, Bluetooth, Wi-Fi technology,but the complex and volatile indoor environment make positioning result of these techniques is not very good.And the characteristics of Radio Frequency Identification technology (RFID) on low cost, non-line,non-contact has a very important significance for indoor positioning research. Radio Frequency Identification technology combining with communication technologies and Internet make it possible to track goods and share information in a wide range.This thesis summarize the key technology of indoor positioning firstly,then make an intensive study of traditional LANDMARC indoor positioning system and observe its parameters which may have a effect on the result of positioning so that we can make some improvements based on it.In this paper,we mainly improve the LANDMARC algorithm from two aspects.First is the definition of the weights.As some research indicate that given two tags with a fixed distance,when those tags are near the reader, it can be found that the difference in RSS of those tags becomes larger.In other words,when tags are close the reader,the RSS values become non-linear.So the weights defined on such traditional LANDMARC system has obvious drawback.Thus, in order to account for this effect, a weighting factor,which is proportional to the RSS value, is added into the weights of traditional LANDMARC to adjust the positioning accuracy. Second is the large errors of the received signal strength value and this is inevitable because of the multipath interference.We need to add the feedback correction to avoid the problem of declining of position accuracy caused by the inaccurate receiving signal strength.Theoretical analysis and the simulation results on MATLAB all demonstrate that the improved algorithm’s positioning error is less than0.2m and it’s maximum error is less than0.3m.Overall,the improved indoor positioning algorithm has higher accuracy compared with traditional LANDMARC algorithm.