Application of Kalman Filter and Adaptive-Network-Based Fuzzy Inference System in Indoor Localization

• Main Authors: Wei-Chieh Hsiao, 蕭維頡
• Other Authors: Chin-Yin Huang
• Format: Others
• Language: zh-TW
• Published: 2014
• Online Access: http://ndltd.ncl.edu.tw/handle/11884914828064826829

碩士 === 東海大學 === 工業工程與經營資訊學系 === 102 === Location-based services are widely integrated in our lives, such as inventory management, personal tracking or healthcare. With increasing applications of wireless localization, accuracy and stability of location estimation have become more critical. However, indoor localization suffers from multipath interference that affects traditional algorithm based on received signal strength indicator (RSSI). In this research, an indoor localization algorithm which combined Kalman filter and adaptive-network-based fuzzy inference system (ANFIS) was proposed. This localization algorithm is based on an assumption of the relationship between the RSSI and the distance is the same within the same distance in the same environment. The proposed localization algorithm utilizes Kalman filter to eliminate noises between transmitters and receivers, and ANFIS to get better environment parameters for the unknown target. For these advantages, the proposed algorithm could improve traditional algorithm based on RSSI which signal may affected by noises. This research established an experiment in an indoor environment to verify the proposed indoor localization algorithm, and it also compared with other localization algorithms. In experiment results, it is expected to significantly improve the accuracy and stability of location estimation.