WiFi fingerprint indoor positioning method

Abstract

The invention relates to a WiFi fingerprint indoor positioning method, belongs to the technical field of indoor fingerprint positioning, and discloses a WiFi fingerprint indoor positioning method based on a clustering algorithm and a least square support vector regression machine. The method comprises the steps of offline data acquisition; partitioning, namely training a least square support vector regression machine model according to the training subsets; and carrying out online position calculation. According to the method, the problem that the k value cannot be reasonably determined in a k-means clustering algorithm is solved; the defect that a k-means clustering algorithm cannot determine a reasonable initial value to shorten the algorithm execution time is overcome, and the defect that a common clustering algorithm cannot select a more reasonable clustering standard is overcome; the problems that a common positioning algorithm cannot reasonably select sub-regions in an online position calculation stage and cannot comprehensively calculate the position are solved. The positioning accuracy is high.

Description

technical field [0001] The invention relates to a WiFi fingerprint indoor positioning method, belongs to the technical field of indoor fingerprint positioning, and is a WiFi fingerprint indoor positioning method based on a clustering algorithm and a least square support vector regression machine. Background technique [0002] With the development of wireless communication, wireless positioning technology has been widely concerned by people. In the outdoor environment, the positioning effect of GPS and other positioning and navigation technologies is almost perfect. Therefore, positioning and navigation technologies such as GPS cannot play a role. Therefore, research and development personnel often use positioning methods based on wireless sensor networks, WiFi, infrared rays, ultrasonic waves, ultra-wideband and other technologies to solve the problem of indoor positioning. Among them, WiFi The network has been widely used in today's society, and it has the natural advantage...

AI Technical Summary

Problems solved by technology

[0011] The purpose of the present invention is to provide a WiFi fingerprint indoor positioning method for the above-mentioned unresolved problems, which is a clustering method that uses a fixed hypersphere volume to determine the number of cluster centers, and uses the Euclidean distance and coordinate distance of received signal strength as metrics. Class algorithm, sub-area training least squares support vector regression model, using a new type of kernel function, using the target and clustering center signal strength Euclidean distance and model training standard deviation weighted positioning WIFI fingerprint indoor positioning method, to solve the existing Some common algorithms fail to solve the following problems: 1. The k value cannot be determined in the k-means clustering algorithm; 2. A reasonable initial value cannot be determined in the k-means clustering algorithm to shorten the execution time of the algorithm; 3. Common clustering algorithms fail to select reasonable clustering criteria; 4. Common positioning algorithms fail to select sub-regions reasonably in the online position calculation stage, and fail to reasonably calculate position coordinates; 5. Common kernel functions fail to optimize the model

Images