Optimal Sampling Interval Acquisition Method for WiFi Fingerprint-Based Localization Based on Monte Carlo Method and Multi-objective Optimization

In recent years, with the gradual popularization of Wireless Local Area Networks (WLAN) and mobile devices, WiFi-based fingerprint localization technology has won wide attention from researchers due to its simple principle and high universality. Although there are numerous performance analysis and optimization works for this technology, the mechanism of error formation is still unclear, and it is still very challenging to optimize the localization performance, especially the relationship between the site survey and localization error is very complicated, and the sampling interval not only affects the localization error but also determines the survey workload. To address the above problems, this paper obtains the relationship between localization error and sampling interval in two-dimensional scenes based on Monte Carlo method, and then defines a survey cost formula based on multi-objective optimization theory to weigh localization error and site survey workload, and then obtains the optimal sampling interval. The simulation results prove the correctness of this study, which is consistent with the previous conclusions obtained through experiments.