| Summary: | 碩士 === 國立臺北科技大學 === 電子工程系研究所 === 103 === Recently, the coverage of the Wi-Fi network become wider in indoor environments, many researchers have focused on developing indoor positioning system based on Wi-Fi received signal strength (RSS), which is a low cost and easy-accessible technique. However, the RSS are changeable over time by obstacles and multipath effect, which may affect the accuracy of positioning. In this paper, we proposed a method based on AP selection and adaptive pattern matching for Wi-Fi indoor positioning policy (ASAPM), using the box plot to remove the outlier of the RSS samples form APs to smooth the RSS. Then we use the standard deviation to analysis the variation of RSS, and select the Top-N APs having interference. Moreover, we uses the history of positioning results to estimate the next moving direction and distance of users, as the pattern matching range of next positioning to reduce the numbers of pattern matching and computational overhead of the positioning system.
The simulation results showed that ASAPM has less the average positioning error, the maximum positioning error and the average pattern matching times than RADAR, SRIGT and APSS. In the average positioning error, ASAPM is better than the others about 36 %. In the maximum positioning, ASAPM is better than the others about 51 %. Finally, in the average pattern matching times, ASAPM is better than the others about 57 %. These showed that ASAPM could reduce computational overhead and more suitable for indoor positioning service in the mobile device.
|
|---|
