A Study of Fuzzy Set Theory in Triangulation Position Estimation–for Zigbee

• Main Authors: Wei-Yang Yang, 楊維揚
• Other Authors: Ping-Teng Chang
• Format: Others
• Language: zh-TW
• Published: 2008
• Online Access: http://ndltd.ncl.edu.tw/handle/44319520970676495630

碩士 === 東海大學 === 工業工程與經營資訊學系 === 96 === This study demonstrates the use of fuzzy set theory in triangulation algorithm, using distance estimation (lateration) method, to estimate locations of objects, dependent on the measurement of signal strengths. The locations are taken as fuzzy numbers. The degree of the membership function of the fuzzy numbers attached describes the level of confidence for an object at an estimated position, so that it reduces the impact of positioning error caused by the variability and uncertainty of the received signal strength. Previous researches indicate that most of the studies apply data training method to reduce the positioning estimation error considering the characteristic value of the signal probability distribution. They seldom discuss the study regarding the fuzzy set theory used for location estimation. Furthermore, the use of fuzzy set theory in the study of location estimation was mainly presenting the fuzzy rules based on the fuzzy logic to infer the determination of the location but not directly estimating the locations which are fuzzy numbers from the signal strength fuzzy number. This study defines the signal strength fuzzy numbers according to the link quality indications and the time based on dataset operations. The distance fuzzy numbers between the objective and the known location are calculated, and then the distance fuzzy numbers in the lateration method of triangulation algorithms are applied to estimating the location and the scope of possible region containing the real location of the objective. In order to evaluate the effectiveness of this method for reducing the impact of positioning error caused by the uncertain changes of the received signal strength, it was experimented with practical data to demonstrate the operation and to statistics the encountered triangulation forms caused by the distortion of the received signals during positioning. In the experiments, fuzzifiered received signal strength was compared with crisp received signal strength used in triangulation location estimation. The results showed that the stability of the fuzzifiered received signal strength method is better, and that the estimation error is less.