Research On WiFi Indoor Fingerprint Location Algorithm Based On A Daptive Gaussian Kernel

With the widespread application of wireless fidelity(WiFi)technology,indoor positioning technology based on WiFi has become a hot research topic,and the location fingerprint method is widely studied and adopted.Therefore,the algorithms for indoor location fingerprint positioning technology are deeply researched in this paper.Through deeper research,it is found that the accuracy and stability of traditional indoor location fingerprint positioning algorithms are susceptible to the indoor environment and positioning scenes.To solve these problems,this paper proposed the following optimization algorithms to improve positioning accuracy and algorithm stabilityThe positioning process based on the received signal strength indication(RSSI)in WiFi location fingerprint positioning technology includes two stages:the offline stage and the online stage.In the offline stage of fingerprint database construction,this paper proposes a hybrid filtering algorithm to preprocess fingerprint data and uses the Kernel Principal Component Analysis(KPCA)to reduce the dimension of the fingerprint database.The hybrid filtering algorithm can correct the noise signal data generated by the complex indoor environment,reduce the influence of the noise signal on the accuracy of the fingerprint database,and then improve the positioning accuracy.And the KPCA algorithm can solve the problem of"dimension of dimensionality" caused by the excessive dimension of the access points(AP)in the fingerprint database,which not only improves the matching efficiency of the fingerprint database but also improves the positioning accuracyIn the online stage,this paper proposes a weighted K-nearest neighbor algorithm based on adaptive Gaussian Kernel(AGK-WKNN).The positioning algorithm uses a Gaussian kernel function with the strong anti-interference ability to calculate the weight value of the weighted K nearest neighbor algorithm and combines the RSSI indoor fingerprint positioning feature to calculate the kernel width value of the Gaussian kernel function so that the kernel width value can be adaptive to the distribution of reference points in the fingerprint database.Moreover,the algorithm uses the selection method to calculate the optimal value of K by finding the potential correlation between reference points.In the process of fingerprint matching,the AGK-WKNN algorithm improves the positioning accuracy and algorithm stability by improving the rationality of the weight distribution of the K n earest reference points and reducing the error caused by the K value selection by experience.Finally,in the experimental part,the optimization algorithm of fingerprint database construction is used to construct the fingerprint database,and the positioning accuracy of the algorithm is compared and analyzed,which verifies that the positioning matching algorithm proposed in this paper not only has higher positioning accuracy but also has wide application value.